Special Issue

(IJMRAE, Vol.6, No. II, April 2014)

Dedicated

To

ICPEDS-NITTE 2014
International Conference on
Power Electronics & Drives

Nitte Mahalinga Adyanthaya Memorial Institute of Technology

(An Autonomous Institute under VTU Belgaum)
Nitte- 574 110, Karkala Taluk,
Karnataka, INDIA.

International Conference on
Power Electronics & Drives [ICPEDS-NITTE 2014]
(Funded by TEQIP Phase II)

10 - 11 April 2014

Organized by:
Electrical & Electronics Engineering Department,
N.M.A.M.I.T, Nitte-574110,
Karnataka, India
Vol. 6 No. II, ISSN 0975-7074

l
April 2014

INTERNATIONAL JOURNAL OF
MULTI DISCIPLINARY RESEARCH
AND ADVANCES IN
ENGINEERING

(IJMRAE)

NT PUBLICAT
CE
AS

IO

@
Nl

ASCENT
PUN
E, INDIA

[Link]
INTERNATIONAL JOURNAL OF MULTIDISCIPLINARY
RESEARCH AND ADVANCES IN ENGINEERING
IJMRAE, ISSN 0975-7074

EDITORIAL BOARD

• Dr. S. M. Khairnar (India)

• Dr. Paulo Coimbra (Portugal)

• Dr. S. M. Fakhrahmad (Iran)

• Dr. Lorenzo Capra (Italy)

• Dr. Lorenzo Capra (Spain)

• Dr. Fatiha Merazka (Algeria)

• Dr. S. Franko (Turkey)

• Dr. Y. Takeuchi (Japan)

• Dr. Jianxin Tang (USA)

• Dr. ZHANG Zhong-cai (China)

• Dr. S. H. Wallace (Malaysia)

• Dr. K. L. Mak (Hong Kong)

• Dr. M. Imbaby (Egypt)

• Dr. Irfan Anjum Manarvi (Taxila)

• Dr. Abdallah. M. Hasna (UAE)

• Dr. R. Beguenane (Canada)

• Dr. Ing. J. Gausemeier (Germany)

• Dr. D. Pearce (UK)

International J. of [Link] & Advcs. in Engg. (IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 1-16

DESIGN AND IMPLEMENTATION OF AN AUTONOMOUS

“UNMANNED GROUND VEHICLE “USING MACHINE
VISION FOR MILITARY APPLICATION

JESHWIN DSOUZA PRABHU

M. Tech, Dept of ECE,
NMAM Institute of technology NITTE,
VTU, Karnataka, India.

Abstract
In this project we propose to design of a remotely controlled robot/unmanned ground vehicle
(predator) capable of reaching the site of threat or security hazard. Real time down link parameters are
projected on a video monitor and counter actions are divided accordingly in real time in threat
scenario like border security, tracking remote places etc. The camera used here is a wireless Night
vision camera which works during day as well as night. The robot has three significant contributions
firstly a remote controlled robot for military application in case of threats , secondly automatic battery
charging using the voltage generated by solar energy and thirdly in case of detection of mines in
threat areas. This robot is movable on any terrain due to its wheel belt and can move preventing
obstacles. These are a sort of mobile robots which are the focus of a great deal in current research
scenario. These are found in industry, military and security environments.
The scope of machine vision is broad and Comprehensive, and it also includes quality assurance and
provides a definite solution. The primary uses of machine vision are automatic inspection and robot
guidance. The machine vision incorporated here increases the security in the border areas or restricted
areas
---------------------------------
Keywords : Machine Vision; Unmanned ground vehicle; Solarenergy; border security automatic
inspection; robot guidance.
© http: //[Link]
2 JESHWIN DSOUZA PRABHU

1. INTRODUCTION
Robotics is one of the most exciting areas of electronics. Robotics is the field of controlling
of electronic machines that can be substitute in the place of human actions. This field has
become so advanced that in the near future robots can imitate human behavior. Machine
vision (mv) is the process of applying a range of technologies and methods to provide
imaging-based automatic inspection, process control and robot guidance in industrial
applications.[1]
Bridging the gap between the world of computer and that of its user has always been one of
the chief goals of robotics. Graphical interface, input devices, speech generators,
handwriting recognition system and face recognition system are just a few examples of how
computers have become more accessible. Motion detection is a more recent thrust in this
direction and represents a major step in bringing the computer into our world.
This project focuses entirely on three specific challenges, the task of detecting and
recognizing the target, and to charge the battery using the power generated by the solar
panel. The last decade has been a time of incredible advancement in motion detection and
recognition. Computers have become fast enough to perform computationally intensive
image processing tasks and storage devices have grown to allow the accumulation of large
database of images.

2. TECHNOLOGY
2.1 Machine Vision concept
Machine vision methods are defined as both the process of defining and creating an MV
solution, and as the technical process that occurs during the operation of the solution. This
includes user interfaces, interfaces for the integration of multi-component systems and
automated data interchange. Nonetheless, the first step in the mv sequence of operation is
acquisition of an image, typically using cameras, lenses, and lighting that has been designed
to provide the differentiation required by subsequent processing[1].
MV software packages then employ various digital image processing techniques to allow the
hardware to recognize what it is looking at, extract the required information, and often make
decisions (such as pass/fail) based on the extracted information.
The imaging device (e.g. Camera) can either be separate from the main image processing
unit or combined with it in which case the combination is generally called a smart camera or
DESIGN AND IMPLEMENTATION OF AN AUTONOMOUS… 3

smart sensor. When separated, the connection may be made to specialized intermediate
hardware, a frame grabber using either a standardized (camera link) or custom interface. MV
implementations also have used digital cameras capable of direct connections.
Techniques used in mv image processing include: thresholding (converting a gray scale
image to black and white, or using separation based on a gray scale value), segmentation,
blob extraction, pattern recognition, barcode and data matrix code reading, optical character
recognition, gauging (measuring object dimensions), positioning, edge detection, color
analysis, filtering (e.g. Morphological filtering) and template matching (finding, matching,
and/or counting specific patterns).
The most common outputs from machine vision systems are pass/fail decisions. These
decisions may in turn trigger mechanisms that reject failed items or sound an alarm. Other
common outputs include object position and orientation information from robot guidance
systems. Additionally, output types include numerical measurement data, data read from
codes and characters, displays of the process or results, stored images, alarms from
automated space monitoring mv systems, and process control signals.
2.2 Image processing concepts
An image basically is a two dimensional array of pixels.[2] These are the tiny little dots of
colour you see on your screen, and the smallest possible size any image can get. When an
image is stored, the image file contains information on every single pixel in that image. This
information includes two things: colour, and pixel location. Where a pixel has different
number of channels and each channel has some value which specifies the colour of that
pixel. In case of a multi-channel pixel, the resulting colour of the pixel is the overall result of
the values of all the channels in it.[2]
Image processing and image analysis tend to focus on 2d images, how to transform one
image to another, e.g., by pixel-wise operations such as contrast enhancement, local
operations such as edge extraction or noise removal, or geometrical transformations such as
rotating the image. This characterization implies that image processing/analysis neither
require assumptions nor produce interpretations about the image content.[3]
4 JESHWIN DSOUZA PRABHU

Fig [Link] the pixel representation

In image processing, high resolution means greater amount of processing. Thus, for image
processing optimum resolution is used large enough to retrieve required data and small
enough for quick processing. In this project a resolution of 720*480 is set.
2.3 Battery Charging Concept.
Here Battery charging mechanism is established using a solar panel and the power generated
by the solar panel based on the solar energy emitted on it is used to charge the battery, The
panel power generated is connected to a buck boost converter The buck–boost converter is
a type of DC-to-DC converter that has an output voltage magnitude that is either greater than
or less than the input voltage magnitude. The voltage generated is compared to the battery
voltage .If voltage is greater than the battery voltage the battery gets charged, this helps is
movement of the robot as the robot is wireless making the Robot (predator) complete.

3. Algorithms
The robot uses two types of algorithms for object detection:
3.1 Sad algorithm
 This algorithm 1st gets the consecutive video frames.
 Coverts them to yuvcolour space and store them in an array.
 Takes there variance, put it in a array and compare those frames.
 By this the robot can detect the motion.
 The process is carried out for entire image.
DESIGN AND IMPLEMENTATION OF AN AUTONOMOUS… 5

3.2 Cross correlation algorithm

 It is same as that of sad algorithm, but here the entire image is spitted into portions
and same procedure is carried out to those portions of image.
 It is a very accurate algorithm and speed process also.
These two algorithms were uses a log files to store the record of captured images.
3.3 Charging and Discharging System
In the first scenario we see the robot and its Image processing using machine vision in the
second part we see the management of charging process of the battery which acts a supply to
the robot.A Buck Boost converter is used for as the charger system, this device is a DC- DC
converter .The charger is connected to the solar panel from where the voltage is generated
and given directly to the battery unit which acts as the power source.[4]

4. Hardware Design
When designing a wireless robot, it is necessary to define what skills are most important to
that machine. About where it should move and shoot, how to communicate, when it should
move and when to shoot? The robot should visualize target and alarm the remote place for
further actions. So the main concern here is the objects movement. The source for the image
can be a camera operating in real time OS or it can be a stored video sequence.[5]
The system is provided with the manual interface, as in the case of target recognition. The
system operated with camera is stored with many images. The internally implemented
algorithm takes the present input image as the input and compares with the previous image
in that array of frames. It alarms the remote user. Later user executes suitable action based
on the recognition result. The action may allow or deny the shooting at target.
The system design consists ofpowerful DC motors will be employed to make the unit move
in different directions. 1st DC motor will be utilized for the left/right controlling mechanism
and 2nd DC motor will be used for driving the unit in forward/reverse direction. 3rd DC motor
for the movement of arm (up/down). Eight relays are used for switching purpose of
movement of dc motors and gun. The relays are mounted with gun (laser) and motors. The
camera used here is wireless night vision web camera. The video captured by it is used for
object detection. The program in the computer takes each frame from the camera and
searches for target by checking change in pixels in consecutive frames. There by any
changes found in pixels, the processor gives out the information (alarm) to remote control
6 JESHWIN DSOUZA PRABHU

room. Then the robot takes necessary [Link] sensor is mounted in front to prevent any
obstacles when the robot moves in any terrain. Metal sensor is used on the robot for
detection of mines and bombs. The other use of the robot is automatic charging here LDR
sensors are used for absorbing the sunlight rays during day noon and evening and the solar
panel rotates respectively in the direction of the sunlight based on what the LDR sensor
detects, the voltage generated by the solar panel is given to a buck boost converter which
charges the battery. The robot can be controlled manually and also works automatically here
RF transceivers are used to control the robot manually from the base station.
Base Station

Robot

Figure 2 : Block diagram of machine vision for military application robot

DESIGN AND IMPLEMENTATION OF AN AUTONOMOUS… 7

Figure 3 : the relays &the power supply unit will be on board and will be used to power the
entire unit .
4.1 Motion detection

Figure 4 : flow chart for motion detection

The camera which was mounted will scan the specified area and capture the frames of video.
Take the variance of those frames. Put them into an array and compare the frames i.e. Pixel
comparison. If there is any change in consecutive frames then the motion is detected and
robot will give an alarm to the remote user. [6]
8 JESHWIN DSOUZA PRABHU

4.2 Gun Firing Mechanism

Figure 5: Flow chart for gun firing action.

Decision is taken by the remote user. The decision will be shoot the target or not, move or
not etc. If the detected motion is by any object or authorized person, the remote user can be
leave the action.
4.3 Battery Charging Mechanism [7]

Figure 6 : Battery charging mechanism using buck boost converter.

DESIGN AND IMPLEMENTATION OF AN AUTONOMOUS… 9

5. HARDWARE PLATFORM OVERVIEW

The interfacing unit in this project is one that binds the control unit and robot. The signals
that received and decoded by the receiver unit are given to the microcontroller through
interfacing unit, which essentially comprises of max 232 chip. It can also allow the hardware
of this project to be connected to computer. It has the RS 232 cable which is connected to the
RF transmitter specification drivers and the receivers that make it compatible with any of the
standard devices. And namely PIC16F877A, sensors, servo motors, wireless and the PC with
matlab for control. This robot is programmed to detect sunlight by using Light Dependent
Resistors (LDR). Servo motor aligns the solar panel to receive maximum light. The robot is
programmed using MPLAB.
In this project the robot using the GUI created from where the robot can be
controlled using the PC. The user can speak into the MIC of the PC and this is then
processed and sends to the robot using the RF TX. At the other end the RF RX will get the
data and then will drive the robot accordingly. While the solar does its job of charging using
the sensors. The camera records the video tracking and the gun moves accordingly.
A. q-r co-ordinates
In order to understand the working mechanism in this project we need to go the q-r diagram
shown in figure [Link] frame is divided in 16 frames in the q-r co-ordinate.

Figure 7 : q-r co-ordinates

10 JESHWIN DSOUZA PRABHU

The q-r co-ordinates is used to indicate the position of the gun, the gun is set to position (1,4)
is the q-r co-ordinate when the gun fires during motion detected the gun goes to a new
position the new position acts as the gunsposition now.
 The power supply unit will be on board and will be used to power the entire unit.
 The LDR sensors are used to absorb the solar energy and to rotate the solar panel
respectively on the direction of the sunlight during the day time, noon and evening,
powerful motor connected to the solar panel helps the rotation of solar panel and at
night the solar panel rotates back to normal state.
 In the prototype the solar panel is fit and the voltage generated by the solar panel is
given to the battery source to charge it as the robot will be using the power from the
DC battery and also generating the respective voltages as required by the on-board
circuit. [8][9]
 Robot can be controlled manually or automatically. In automatic control we need to
press the start/stop button of the graphical user interface window.
 Initial position of gun is set to q=0,r=4 position according to the q-r co-ordinates.
 Whenever automatic control is asserted the robotic vehicle waits for 20 seconds. If it
detects any difference frame then the gun moves according to co-ordinates of q-r
diagram. If no difference image is found then the robot starts moving in the forward
direction
 A matrix is divided into 16 frames according to the image resolution 720*480 of the
camera.
 A square path is set and after 20 seconds it turns to the right and again waits 20
seconds checking for the difference frame, it keeps on doing this until it completes the
square path
 If any obstacle is detected in the motion of the robot, the infrared sensors attached to
the robot sense the obstacle and sends an alarming signal to the operator who is
operating. Then operator has the provision to move the vehicle manually so that the
obstacle doesn’t affect the vehicle motion.
 The 2D continuous image a(x,y) is divided into n rows and m columns. The
intersection of a row and a column is termed a pixel. The value assigned to the integer
DESIGN AND IMPLEMENTATION OF AN AUTONOMOUS… 11

coordinates [m,n] with {m=0,1,2,…,m-1} and {n=0,1,2,…,n-1} is a[m,n]. In fact, in

most cases a(x, y).
 A threshold value of 5 in this scenario. If any movement occurs the gun fires based on
the movement in the particular pixel of the frame and also indicates the location and
the threshold level on the screen, it also shows whether the gun has moved right, left
up or down.
 Due to the wheel belt it is possible for the robot to move in any terrain which is the
biggest advantage.
 If obstacle detected, the robot waits for a few seconds for it to move, if it does not
move the robot reverses back, moves right and after few seconds continuous its path.
 In case of detection of mines, when the robot comes in the terrain of mines the metal
sensor detects the mines and alerts the [Link] user comes to know the location of the
mines and need to take necessary action.
 The biggest advantage is the robot can work even at night because night vision camera
is used the camera switches to it IR light at night automatically.
 In addition, events recorded by a robot’s camera can provide evidence for future
forensic purposes.

6. TESTING AND RESULT

6.1 The Military Robot (Predator)
In the design of our military robot we have named it as “predator” the gun is mounted on
top, the camera in front the IR sensors ,metal sensor ,LDR sensor and solar panel are clearly
[Link] pro is connected for serial communication of data. To communicate with the base
station, from where all the parameters are observed based on the feedback given by the
camera.
6.2 Image acquisition tool
This project is tested using wireless night vision camera under many conditions, which
showed good result. Figure 8.1 shows initial adjustment of the camera resolution to 720*480
pixels in matlab under image acquisition toolbox.
12 JESHWIN DSOUZA PRABHU

Figure 8 : Shows the Machine Vision robot “Predator”.

Figure 9 : image acquisition tool

6.3 GUI Window
Figure 10 shows the graphical user interface window (gui) that serves as front end to our
project of machine vision. Front end was created using matlab
DESIGN AND IMPLEMENTATION OF AN AUTONOMOUS… 13

Figure 10 : Simple GUI box for robot control mechanism.

The project includes GUI box that allows user to click and drag the graphical object with a
mouse instead of entering text at a command line. It consists of a push buttons for robot
movement, as shown in the fig10
6.4 No Motion detected
Under normal conditions the image quality was very high so the motion detection is free of
errors. Figure 11 shows that no obstacle and no motion are seen.

Figure 11 : no obstacle and motion detected.

14 JESHWIN DSOUZA PRABHU

6.5 Motion detected

Figure 12 shows that a movement has occurred or object detected

Figure 12 as soon as movement is seen the gun [Link] shows the calibration of the gun , 80%
correctness is achieved in our project

Figure 13 shows that a soon as the metal piece is moved the gun fires. This is seen with the
help of laser as shown in figure.
DESIGN AND IMPLEMENTATION OF AN AUTONOMOUS… 15

As the object is moved to the right the gun position moves to its right and fires the laser. In
this case we achieved approximately 90% accuracy in gun firing.

CONCLUSION
 This Robot has been credited with helping military personal to track the movements
of enemies still, even considering all of its positive applications, it is impossible to
ignore in many ways such a system could be misused.
 Ecofriendly, as battery supply and charging unit is used.
 Computers with their abilities will not only revolutionized automated security, but
will be highly useful tools to their users. Progress in this field will enable researches
to turn their attention to more advanced systems. Further chip away at the barrier
between computers and their users.
 This robot has been a great help and experience in
 Border issues as terrorists attack have been on a rise .
 This robot works even at night due to the night vision camera used

ACKNOWLEDGEMENT
I would like to acknowledge and thank BCS Innovations R&D for supplying me the
necessary hardware components and their guidance and advice throughout the lab research
work.

REFERENCES

[1] Research on Adaptive Path Tracking Algorithms of Machine Vision-Based Robot Huang Jing ;
Coll. of Inf. & Electron., Zhejiang Univ. of Sci. & Technol., Hangzhou, China ; Chen Hanwei ;
Hanbo, Zhang 2012 Third International Conference on Digital Manufacturing and Automation
(ICDMA), ISBN: 978-1-4673-2217-1
[2] Frank Vahid and Givargis Tony, Wiley John and Sons, 2002 a unified hardware/software
introduction isbn 0471386782.
[3] Rafael. C. gonsalez, Richard. E. Woods, digital image processing, sixth Indian reprint,
addisonwesley limited, 2000
[4] Goswami Y., Kreith F., and Kreder J. (1999), "Principles of Solar Engineering, Chapter 2:
Fundamentals of Solar Radiation," Taylor & Francis, Philadelphia.
[5] Development of a mobile surveillance robot Ki Sang Hwang ; Chungnam Nat. Univ., Daejeon ;
Kyu Jin Park ; Do Hyun Kim ; Sung-Soo Kim . International Conference on Control,
Automation and Systems, 2007. ICCAS '07
16 JESHWIN DSOUZA PRABHU

[6] Mooncheol Won, Sung-Soo Kim, "Design and Control of a Marine Satellite Antenna," Journal
of Mechanical Science and Technology, Vol. 19, No. 1, pp. 473-480, 2005.
[7] Abdallah S. and Nijmeh S., "Two-Axis Sun Tracking with PLC Control," in Energy Conversion
and Management, vol. 45, 2004. 1931-1939.
[8] Canova A., L. Giaccone, and Spertino F., "Sun Tracking for Capture Improvement," 22nd
European Photovoltaic Solar Energy Conference EUPVSEC, WIP Renewable Energies, Milano.
3053-3058, September 2007.
[9] Solar Tracker Robot using microcontroller Afarulrazi, A.B. ; Dept. of Electr. Power Eng., Univ.
Tun Hussein Onn Malaysia, Parit Raja, Malaysia ; Utomo, W.M. ; Liew, K.L. ; Zarafi, M.
International Conference on Business, Engineering and Industrial Applications (ICBEIA), 2011
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 17-25

CROSS COMPILING RASPBERRY PI REMOTELY

TO A WINDOWS MACHINE

SINDHU B. TERADAL1, NELSON PEREIRA2 AND ROOPA B. HEGDE3

1
[Link] student, Dept. of E&C NMAMIT, Nitte and Project-Intern,
Expert Vision Labs Pvt. Ltd, NITK-STEP, Mangalore, Karnataka, India
2
Technical Director and Founder, Expert Vision Labs Pvt. Ltd.,
NITK-STEP, Mangalore, Karnataka, India
3
Assistant Professor, Dept. of E & C, NMAMIT, Nitte, Karkala,
Udupi District, Karnataka, India

Abstract
Among the recent technological advancements Raspberry Pi Single Development board is one, which
is going to change the way in which computers running on linux can be used. An asset of using Linux
over Windows on Raspberry Pi is that it is not commercial. Owing to the enormous applications that
Raspberry Pi has, it is limited by the RAM resources in certain applications. In order to overcome this
stricture we adapt a technique called as Cross compiling. Cross compiling is a process effectuated by
using a compiler capable of creating executable for a platform other than the one on which the
compiler is running. To ease the process carried out during cross compiling a SSH server is being
established on a Windows Machine.

----------------------------------
Keywords : Raspberry Pi, Linux, Cross Compiling, SSH server, Xming
© http: //[Link]
18 SINDHU B. TERADAL, NELSON PEREIRA AND ROOPA B. HEGDE

1. INTRODUCTION
Almost everywhere we can find Technology – in workplace, Offices, Schools, Colleges,
Libraries, Hospitals and even at our homes. We have become so much dependent on
technology that it has almost dominated us. We can’t even think of living our lives without
the modernly available gadgets. In the olden days computers were huge in size adding to the
bulkiness and also expensive. But now with the latest fabrication and compression techniques
things are getting miniaturized and it’s also becoming feasible for even a common man to
purchase them.
In analogous to a normal computer Raspberry Pi on booting the GUI contains file manager
present in the taskbar, which easily manages files, allows the user to view files, play audio
and navigate through the file system. In addition the GUI comprises of
 The LX terminal, which is the terminal window where the commands can be typed,
 Midori, a lightweight web browser
 Scratch, Programming teaching tool
 IDLE, the integrated development environment for Python scripting etc.
1.1 Raspberry Pi Board
Raspberry Pi is credit-card sized single board computer [4] similar to other single
development boards like Beagle Bone Black or Panda board. There are two models: Model a
costing $25 and Model B costing $35. This less expensive feature of Raspberry Pi earmarks
it from the other development boards. Being developed in UK by the Raspberry Pi
foundation, it has many distributors like Egoman, RS components, Element14 etc. which sell
the Raspberry Pi online. The one available in China has PCB which is red in color unlike the
green coloured ones which are available to the rest to the world.
The system on chip (SOC) associated with Raspberry Pi is Broadcom BCM2835, which in
addition VideoCore GPU also includes an ARM1176JZF-S 700 MHz processor following
ARM-7 architecture. The alpha version (i.e. Model A) was originally shipped with 256MB
RAM which was later upgraded to the beta version (Model B) having RAM of 512MB. The
various Linux flavors compatible with the Raspberry Pi are RISC OS, Arch Linux, Raspbian,
Fedora etc. I’m currently working on Raspbian operating system as it is the recommended
OS for beginners.
CROSS COMPILING RASPBERRY PI REMOTELY… 19

1.2 Hardware
A network time server should be used since the Raspberry Pi does not have real time clock.
However, a real time clock can be made available with battery backup through the I2C
interface which is done with the aid of GPIO pins. There are many features differentiating
the alpha and beta version of the Raspberry Pi like RAM capability as mentioned before.
Firstly, Model A doesn’t possess an 8P8C (RJ45) Ethernet port but it can be connected to the
internet by using a user- supplied USB Ethernet or Wi-Fi adapter inserted in the USB 2.0
slot. Secondly, Model A contains only one USB port while the Model B contains two.

Ethernet Port HDMI Out Micro USB Power Port

SDHC
card

Dual
USB
2.0
Ports Broadcom
Audio Out RCA Video BCM2835
Out SOC

Figure 1 : Raspberry Pi board

The Raspberry Pi board contains a processor and graphics chip, program memory (RAM) and
various interfaces and connectors for external device. Some of the devices are indispensible,
others are elective. Raspberry Pi abbreviated as RPi functions in the same way as a standard
PC, necessitating a keyboard for command entry, a monitor screen, a USB mouse and power
supply. As is typical of modern computers, generic keyboard and mouse are compatible with
the RPi. The hard disc of the type found in an archetypal PC is not really in keeping with the
miniature size of the RPi. Instead a SDHC flash memory card normally seen in digital
cameras can be used, which is the place where the OS is installed as well as it acts as a
storage device. RPi will ‘boot’ (load the operating system into RAM) from JZF into this card
in analogous to how a normal PC ‘boots up’ Windows from its hard disk.
20 SINDHU B. TERADAL, NELSON PEREIRA AND ROOPA B. HEGDE

1.3 SSH Server

Acronymically called as secure shell, the SSH is an authenticated network protocol for
remote command-line login, secure data communication, remote command execution and
other secure network services between two networked computers that connects through a
secure channel over an insecure network, a server and a client.
On one hand it can be considered as one of the per-eminent application for OS having Unix-
like environment, and on the hand it uses public-key cryptography to authenticate the remote
computer and allows it to authenticate the user. It’s a supersede to the then existing Telnet
and other insecure remote shell protocol such as Berkeley rsh. The latter protocols were
entitled to send information notably passwords, in plaintext, making it susceptible to
interception and disclosure using packet analysis.
1.4 PuTTY
Capable of Mimicking the functions of a one computer system in the host, with the latter
being different from the former, PuTTY is a terminal Emulator application that acts as a
client to various computing protocols primarily being SSH and some others like Telnet,
rlogin etc. The last three letters in the word PuTTY are capitalized as they are the acronym
for teletype. Officially it was written for Microsoft Windows and later extended to be ported
to Unix-like environment.

Figure 2 : PuTTY running a session on Windows Machine

CROSS COMPILING RASPBERRY PI REMOTELY… 21

1.5 Xming
For operating system like Microsoft Windows Xming is an implementation of X windowing
System. Also referred as X11, X windowing system is a type of GUI. Being superfluous to
third-party emulation software, it runs natively on Windows.

2. METHODOLOGY

Start

Windows Machine Raspberry Pi Board (RPi)

Connections done, booted

up and obtained the GUI

Remote Connection through the SSH server

Using Eclipse IDE to cross compile RPi to Windows Machine

Stop

Figure 3 : Proposed model

The proposed methodology is as shown in Fig. 3 where in after the SSH server is established
Cross Compiling is carried out.
2.1 Cross-Compiling
A common native compiler like gcc, g++ used for C or C++ coding, normally found on
archetypal PC is a compiler that runs on the Intel machine, as well as creates binaries for the
same. Meaning it creates binary files for the same type of machine that it runs on. Besides
running on an ARM machine the GCC tool also creates binaries for an ARM machine.
‘Arm-unknown-linux-gnueabi’ is one of cross compiler that runs on the Intel machine but
creates binaries for ARM machine like Raspberry Pi. Strictly speaking, it runs on one
architecture and creates binaries for the other. This allows us to develop and compile our
22 SINDHU B. TERADAL, NELSON PEREIRA AND ROOPA B. HEGDE

libraries/programs on our Windows machine, but when it comes to deploying the

libraries/binaries we deploy and run them on the Raspberry Pi.
On Raspberry Pi, IDE like ‘geany’ can be installed and C codes can be compiled then and
there itself which superfluous the use of a cross compiler. But the main intention of using
cross compilation over native compilation is that it tends to be faster. The RPi board may be
expeditious to the microcontroller but because of the constrained RAM resources it is
dawdling to an average Desktop computer.

3. EXPERIMENTAL RESULTS
With all the connections plugged in, booted up the small credit-size computer and started
working with the board. There is no hard disc as such for the RPi so an SD (secure disc) card
is used. I purchased a SDHC card preloaded with Linux-Debian flavored Raspbian operating
system. After booting up, I was able to obtain a Desktop-interface which is analogous to a
Windows machine containing several icons as mentioned before.

Figure 4 : Raspberry Pi’s graphical user interface (GUI)

In order to set up the SSH server I made use of two softwares i.e. Xming, which is used for
displaying the desktop interface of Raspberry Pi on the Windows Machine and PuTTY, a
terminal used to remotely connect to the Raspberry Pi
To enable SSH on the Raspberry Pi the following steps can be used:
 Open the LX Terminal and type ‘sudo raspi-config’
CROSS COMPILING RASPBERRY PI REMOTELY… 23

 Go to SSH  enable  OK  Finish

Figure 5 : Figure illustrating settings done in PuTTY configuration

By using SSH the need for using a separate monitor, keyboard and mouse is eliminated. But
both laptop/PC running any flavor of Windows and Raspberry Pi need to be connected to the
same router.

Figure 6 : Working on the Raspberry Pi’s terminal in Eclipse installed on the Windows
Machine
24 SINDHU B. TERADAL, NELSON PEREIRA AND ROOPA B. HEGDE

Though the cross compiling procedure was tedious I managed to successfully cross compile C
code from Windows machine to Raspberry Pi board using Eclipse IDE installed on Windows
machine.
Fig. 6 illustrates a simple C code of swapping two numbers. Zooming only the terminal part
(which is present at the bottom right) of the Fig. 6 and illustrating the further steps involved in
cross compiling we get

Figure 7 : The terminal asking the user to enter the values of x and y

Figure 8 : Entering the values of x and y

CROSS COMPILING RASPBERRY PI REMOTELY… 25

Figure 9 : Output obtained

REFERENCES

[1] Adafruit’s Raspberry Pi Lesson6. Using SSH [Online]. Available:[Link]

downloads/ pdf/[Link]

[2] Cross compiler [Online]. Available: [Link]

[3] Development environment for the Raspberry Pi using a croos compiler toolchain and Eclipse.
[Online]. Available: [Link]
cross-compiler/

[4] Getting started with Raspberry Pi. [Online]. Available: [Link]

raspberry-pi-getting-started-guide-how- to/

[5] Remote cross debugging-Windows to RaspberryPi [Online]. Available://[Link]/wordpress/

techstuff/development/remote-debugging-raspberrypi

[6] Remotely accessing the Raspberry Pi via SSH server-console mode [Online]. Available:
[Link]

[7] RPi hardware [Online]. Available: [Link]

[8] RPi hardware basic setup [Online]. Available: [Link]

[9] Secure shell [Online]. Available: [Link]

[10] Setting up windows Eclipse programming of the RPi [Online]. Available: [Link]
proj..com/pi/programming-in-c/compilers-and-ides/eclipse/programming-the-rpi-in-windows-
using-eclipse
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 27-38

ANALYSIS OF CRACKS AND DEFECTS ON CONCRETE

SURFACE USING IMAGE PROCESSING TECHNIQUES

SADAF M. MUJAWAR AND D. A. KULKARNI

Department of Computer Science and Engineering,
KLS Gogte Institue of Technology, Belgaum, Karnataka, India

Abstract
Concrete is nowadays one of the most required materials used in construction and for this purpose its
properties have been more and more studied so that it can be used at its maximum performance. In
this paper, aim is to concentrate on the quality of the concrete by considering the color changes
occurring in the curing process. For a curing test it requires various types of tests such as tensile
strength test, temperature test etc to be checked. But in this proposed idea digital image processing
can be used for determining the color of the concrete for each age of the specimen used. Cracking can
invite sudden failures of concrete structures. The objective of this paper is to develop an integrated
model based on digital image processing in developing the numerical representation of defects .The
integration model consists of crack quantification, change detection, neural networks, and classifiers
to visualize the defects in such a way that it mimics the on-site visual inspections. Quantifying the
surface color changes by digital image analysis to assess the quality of the concrete mixture is
investigated. Based on these results the concrete strength can be determined. This type of set up can
be done for the industries producing the large amount of cement products so that no manual inspection
of the things can be done. We also detect the cracks automatically by using the images as the data and
based on the information the type of concrete quality can be determined.

---------------------------------
Keywords : Concrete mixture, Tensile strength test, temperature test, crack detection.
© http: //[Link]
28 SADAF M. MUJAWAR AND D. A. KULKARNI

1. INTRODUCTION
There is increasing awareness of the need to inspect concrete structures, such as old
buildings and other concrete structures, because they may be affected by aging, natural
disasters or shoddy workmanship. To maintain and manage these concrete structures
properly, it is necessary to inspect and diagnose the state of their deterioration accurately, to
detect cracks on their surfaces at an early stage, and to repair detected defects or apply
reinforcement. Contact-less remote-sensing crack detection and quantification
methodologies are described in this paper. The systems and methodologies include depth
perception for detecting the cracks and also quantify the cracks. Proposed method includes
the crack detection, strength calculation, inspection and monitoring of the structure and
classification using neural network. In addition, in the field of product manufacturing, it is
extremely important to maintain high quality control by inspecting products for defects
accurately, thereby raising productivity. This is why increasing attention is being drawn to
technology for non-destructive testing of concrete structures, which enables unskilled
operators to inspect and diagnose aging structures or defective products quickly and
accurately without damaging them at all. It is an inevitable factor that materials forming the
concrete should have good quality, proper material rates should be used. In this paper we are
going to use image processing technique for three different quantities of cement proportions
mixtures added along with sand and aggregates. The sand and aggregates proportion is kept
same and varying the cement proportion to these mixtures the concrete cubes of size
15*15*15cm specimen are prepared. The images of various curing stages for various cement
proportion mixtures are collected for 3 days, 7 days, 14 days and 28 days by digital camera.
We calculate the color changes occurring in different stages of the cure period and based on
this result we can detect how well is the concrete cured.
Concrete surface defects e. g, cracks, and air pockets, are always produced in concrete
construction projects due to insufficient vibrations, inappropriate curing or other incorrect
operations. Therefore, it is necessary to inspect concrete surfaces and evaluate their quality
in any concrete construction project. This work is now mainly performed manually by
qualified inspectors. There is no doubt that manual inspection is a powerful inspection tool
.However, several limitations were found in manual inspection.
First, the results of manual inspection are subjective and not always reliable due to its heavy
reliance on inspectors personal experience and knowledge to make evaluation [1].
ANALYSIS OF CRACKS AND DEFECTS ON CONCRETE… 29

Besides that, it is time-consuming especially when a complex structure needs to be

inspected. Also, the requirement of experienced inspectors for manual inspection poses as a
challenge because of the pressing shortage of experienced and highly trained inspectors in
construction industry. In order to overcome the limitations of manual inspection ,automated
inspection using visual sensing was introduced as an advanced infrastructure inspection
methodology can be developed by using digital image processing techniques.
In the methodology, the defects at concrete surfaces are detected and assessed from the
analysis of digital images based on their distinct visual characteristics and/or shapes. This
way, direct personal observations are avoided. The limitations of manual inspection are
alleviated while the flexibility of performing inspection work is still kept. Following the idea
of automated defect detection, a large number of methods were developed for the detection
of cracks _Abdel-Qader et al. 2003; Sinha and Fieguth 2006; Fujia et al. 2006 and air
pockets -Suwwanakarn et al. 2007; Zhu and Brilakis [Link] the developed methods
can successfully detect defects at concrete surfaces, they have no ability of assessing the
impact of these defects on the quality of concrete surfaces.
A crack is the separation of an object or material into two, or more, pieces under the action
of stress. Depending on the substance which is cracked, the crack reduces the strength of the
materials in most cases, e.g. building walls, roads, etc. At the beginning cracks are used to be
detected manually. However, detecting a crack manually is a very intricate and time
consuming process. With the advance of science and technology, automated systems were
used to detect cracks instead of humans. By using the automated systems, less time is
consumed and the cost for detecting the cracks reduced [2].
The stability of structure is an important factor for the construction phase and in the
maintenance phase. The importance of safety is increased as the construction of high-rise
buildings, super long span bridges, and asymmetric buildings became popular. For this
reason, construction safety management system is actively under development these days
[1]. In addition, the interest in the automation of the construction management system has
been increased, because the civil infrastructure under the management is large in scale, the
regular evaluations are needed to guarantee the continuity in service, running cost is very
expensive, and the safety of workers should be ensured. Development of an automated and
effective infrastructure life-cycle management system, it is possible to secure the stability of
the facility, and also reduce the efforts of the inspectors, inspection, time, and maintenance
30 SADAF M. MUJAWAR AND D. A. KULKARNI

cost. Moreover, we can judge the condition of the structural health objectively by acquiring
and processing the data [2].
The proposed idea used here is to detect crack on the concrete surface after appropriate
images are captured during image acquisition phase. During Image acquisition, a tripod
stand with 40cm distance from the concrete specimen to the digital camera, images were
captured for various cured days such as 1 day, 3 days, 7 days, 14 days and 28 days. These
images were collected in RGB format as saved in jpeg format. Likewise images with
different cement proportion were collected for the same above mentioned days.
The resolution of all the tested images is 5152*3684. These images were captured under
normal room temperature conditions. During the test, each image is given as input into the
crack code, where surface defects i.e. cracks are detected and and result is shown as image
with crack.
The comparison of effectiveness of edge detection techniques Canny edge detector, Sobel,
Roberts and prewitt edge detector were used for detecting cracks and then came up with a
simple algorithm by using subplotting to detect crack on a a given image by pre-processing it
and then using the subplotting algorithm.
Recently, some methods for crack detection by means of visual inspection have been
proposed. Abdel-Qader et al. suggested a comparison of the effectiveness of crack detection
in the images of a bridge surface by using the wavelet transform, Fourier transform, Sobel
filter, and Canny filter [2]. They concluded that the wavelet transform is significantly more
reliable than the other methods. Hutchinson et al. used a canny filter and the wavelet
transform for crack detection and estimated the parameters using the receiver operating
characteristic (ROC) analysis [3]. We have also proposed an automatic visual inspection
system using images captured by a digital camera [3]. This system can extract and analyze
cracks on the concrete surface by combining several image processing techniques including
the wavelet transform, shading correction, and binarization. Kawamura et al. proposed a
method using a genetic algorithm for the semiautomatic optimization of image processing
parameters for precise crack detection. However, these methods do not consider the essential
characteristics of cracks such as its connectivity. Also, these methods use global image
processing methods such as the wavelet transform by focusing on the characteristics of the
entire image. The most appropriate can be used to detect cracks.
ANALYSIS OF CRACKS AND DEFECTS ON CONCRETE… 31

2. OVERVIEW OF THE SYSTEM DESIGN

2.1 Image Acquisition
Concrete samples whose sizes were 15×15×15 cm and which were waited in curing for 1
day, 3 days, 7 days, 14 days and 28 days were captured. Images of the concrete samples
which were dry ie uncured were also acquired.
A mechanism comprising of tripod, Nikon Coolpix digital camera, was built for getting
digital images of section surfaces (Fig 2). After the surfaces of the samples were dried, their
jpg spanned images having 5152×3864 pixel sizes, which were in RGB color mode, were
obtained. All of the imaging shoots on the concrete samples were performed in the same
height and light environment on the sample surface.
The arrangement of the image acquisition is as shown in fig.1. The images for various
concrete cubes are taken, ie the images for various proportions are collected by making an
experimental setup as shown below. The images are captured as a training set of data from
each variously cured days. The image of each sample is captured using digital camera (latest
configured Nikon Coolpix) and saved in a RGB colour with JPEG format. The camera is to
be setup on a tripod (40 cm height) in a constant position and lens aperture under control
lighting environment as shown below .These images are further used for the color and
texture detection as well as crack detection present if any by using image processing color
feature technique.
The configuration of the digital camera used is 20.1 Megapixel, 3.0 LCD and 5X Zooming
capacity. The camera was mounted on a tripod stand with distance from lens to the specimen
as 40 cm in height. The camera was zoomed by 2X to acquire the image of concrete surface.
Likewise, images of various surfaces for one particular cube were captured as shown below.
The image acquisition procedure was carried out in a dark room with light focusing on the
concrete specimen and under the flash light of the camera. The below figures illustrates the
idea as to how images were acquired by using digital camera for all the different proportion
cubes.
32 SADAF M. MUJAWAR AND D. A. KULKARNI

Figure 1 : Image Acquisition of concrete surfaces of the samples.

2.2 Image Processing Techniques Used for Crack or Defect Detection

After the image acquisition was done a total of 185 images were collected and all the images
stored in RGB format , images were to be pre-processed in order to meet some requirements.
The pre-processing step involved many trial and error methods where the final resulting
method used was converting the RGB image to HSV color space and varying the brightness
and contrast feature of the image. This image was converted to binary image and crack
detection code was applied. This method proved the accurate result for detecting the cracks.
The other trial and error methods for image pre-processing are using Histogram Equalization
on the original image as well as gray scale image to extract the crack part from the image.
This method was not satisfactory in providing clear cracks.
The next method tried was using a mask of 40*40 and applying the median filter to this
mask values so that the color intensity would be maintained. This method also proved to be
unsatisfactory because the crack was not detected and images without crack were detected as
images with crack.
The next method tried was using the imadjust (contrast stretching) function to change the
brightness and contrast of the original image to detect cracks. The method did not present
any good result because the part with air void and crack color intensity was changing and
this resulted in inaccuracy.
The next method tried was converting the RGB image to CIELab image and using imtool
function in matlab to know the thresholding value and detecting the crack. This method also
ANALYSIS OF CRACKS AND DEFECTS ON CONCRETE… 33

failed as the crack could not be detected and images without crack were detected as images
with crack and viceversa.
The next method tried was converting RGB image to HSV image colorspace and applying
the crack detection algorithm,this method was also unsatisfactory as it could not detect
cracks accurately.
Hence the final method which was used and that provided the accurate result was using the
HSV image and varying the intensity (Brightness) value and the hue (contrast) value and
making the saturation to 0(zero).This image was then Binarized and crack detection
algorithm applied, the resulting output was accurate by detecting images with “crack” as
“cracked images” and without crack as “images without crack”.
As described above about image acquisition the acquired images were resized and enhanced
by changing the brightness and contrast of the images. The following block diagram depicts
the flow of processes to detect crack. The processed images are given as input to the crack
detection algorithm and cracks if any are detected.

Figure 2 : Block Diagram to depict the crack detection process

After the image acquisition is done and images saved in jpeg format these images are used as
input to detect the defects if found any. From fig 4 we can observe that pre-processing of the
images is to be done in order to get appropriate results .In this study we have done pre-
processing by converting the image from RGB to HSV colorspace and varying the
34 SADAF M. MUJAWAR AND D. A. KULKARNI

brightness and contrast. This image is now converted by to RGB from HSV and then
converted to gray image.
2.2.1 Converting images to colour spaces
The RGB concrete sample images were converted to HSV colour space. These images were
then transformed into gray scale images. Gray scale image is an image which only carries
intensity information. It is composed of shades of gray, varying from black to white. The
converting process was done by using MATLAB software.
2.2.2 RGB to HSV conversion
HSV is a non-linear transformation of the RGB colour space which separates the luminance
component from the chrominance information and a device dependent colour space. This
conversion is done in MATLAB software and the intensity value is increased to get
appropriate detection of defects if found any. The saturation value was made zero and only
brightness and contrast value was changed. This is the first step in image pre-processing.
2.2.3 Morphological Operation
Morphology is a broad set of image processing operations that process images based on
shapes. Morphological operations apply a structuring element to an input image, creating an
output image of the same size. In a morphological operation, the value of each pixel in the
output image is based on a comparison of the corresponding pixel in the input image with its
neighbours. By choosing the size and shape of the neighbourhood, you can construct a
morphological operation that is sensitive to specific shapes in the input image.
The most basic morphological operations are dilation and erosion. Dilation adds pixels to the
boundaries of objects in an image, while erosion removes pixels on object boundaries. The
number of pixels added or removed from the objects in an image depends on the size and
shape of the structuring element used to process the image. In the morphological dilation and
erosion operations, the state of any given pixel in the output image is determined by applying
a rule to the corresponding pixel and its neighbours in the input image. The rule used to
process the pixels defines the operation as a dilation or an erosion.
The pre-processed images are now used to perform morphological operation. The
morphological operation used in this project study is erosion method. The erosion operation
is used on the black and white image using the structuring element. A structuring element is
a matrix consisting of only 0's and 1's that can have any arbitrary shape and size. The pixels
with values of 1 define the neighbourhood. disk is the type used for structuring element.
ANALYSIS OF CRACKS AND DEFECTS ON CONCRETE… 35

2.3. Crack or Defect Detection Algorithm

Input : A morphologically eroded image
Output : To find defect in given image
Process:
1. Accept eroded binary image with shape “disk” and size 5
2. Repeat the loop till whole size of the image is covered
3. Compute total number of black pixels in binary image
tpr=(sum(sum(~eroded))/(size of Image))*100
4. Check whether tpr value lies between threshold
5. Repeat the loop by extracting 10*10 pixels from the image
Set temp=eroded(10,10) pixels
Compute blackpixel=sum(sum(~temp))
Compute percentage of black pixels
pr=(blackpixel/size(temp))*100
6. Check whether pr lies between a specific range
7. if temp(10,10) is found with black pixels
8. Mark a “*” on black pixels in the binary image with different color
9. Maintain a Count of temp with black pixels
end loop /*Of computed pr values */
end loop /* Of computed tpr values */
10. Crop the image from eroded image until crack part is extracted
11. Compute Length of crack using distance formula
Crack length=sqrt(x2-x1)^2+(y2-y1)^2
12. Compute width of crack from the cropped image
13. Check whether size of row < size of column
14. Compute crack width=sum(~Icrop(:,1))
else
15. Compute crack width=sum(~Icrop(1,:))
16. Check whether count variable is > threshold value
17. Display ”Crack Detected with crack length and crack width values”
18. else
36 SADAF M. MUJAWAR AND D. A. KULKARNI

19. Display “Crack Not Found”

2.4 Color and Texture Feature Extraction
The surface images are to be tested with crack detection algorithm, if the images are found to
be without crack these images undergo the color and texture feature extraction to detect the
type of curing, number of days required to cure that particular specimen etc. The features
that are to be used are can be Haralick Features for texture feature extraction. Some of the
features tried which gave distinctive result was Mean of RGB image, Kurtosis of RGB,
Skewness of RGB, Contrast of RGB, Correlation of RGB and Entropy of RGB.
These extracted features are given as input to the classifier such as neural network.A set of
data is been trained and some of the sample images are to be tested by using the surface
images detected. A data set with all the known Strength from the civil laboratory can act as a
parameter to know the accuracy and type of curirng done.

3. FUTURE WORK
The project work carried out so far used the techniques such as pre-processing of the image,
morphological operation and detection of presence of crack in the given image.
The future work which is planned to be carried out is using Classification Technique where
one of the classification method either SVM or NN(Neural Network) most probably
Probability Neural Network (PNN) can be used. This classification technique is to be applied
for both training and testing the data set of images.
The Civil Laboratory test carried out on the concrete samples and the strength calculated for
all the cubes is maintained as a data set. This data set can be used to co-relate with the image
processing feature extraction techniques.
The Analysis part consists of graph which depicts the plotting of curing strength along with
the number of days used. This idea proposed is initially tested on concrete samples and
analysis work is been carried out. Some of the construction sites such as cement product
manufacturing companies, bricks, tile industries can use this method to maintain quality of
the products.
ANALYSIS OF CRACKS AND DEFECTS ON CONCRETE… 37

4. SNAPSHOTS
Image With Horizontal Crack

Image with diagonal Crack

38 SADAF M. MUJAWAR AND D. A. KULKARNI

CONCLUSION
In this study, we propose an efficient method for quality testing on the concrete mixture
based on the cement proportion added and the results are obtained by using image processing
techniques for color and texture detection based on the RGB and HUE properties. The
crack detection can be done by using the percolation processing test to detect the type of
cracks. The quality of concrete used can be determined by taking mixtures of different
amount of cement and acquiring images of such specimens and processing them .Hence this
study can help in designing a system setup for making the industries automated so that the
accuracy can be obtained as comaparable to the laboratory results.

REFERENCES

[1] Machine Vision-Based Concrete Surface Quality Assessment- Zhenhua Zhu and Loannis
Brilakis.
[2] Concrete Structure Analysis and Classification using Image Processing - Rashmi R.A, A.D
Mane, S.L Tade
[3] Hanbury and J. Serra. A 3D-polar coordinate colour representation suitable for image
analysis. Technical Report PRIP-TR-077, TU Wien, 2012.
[4] Image-Based Crack Detection for Real Concrete Surfaces Tomoyuki Yamaguchi, Shingo
Nakamura,Ryo Saegusa ,Shuji Hashimoto. Machine Vision and Applications 2010.
[5] A guide for checking curing process. Guide provided by NIST.
[6] Evaluation of texture methods for image analysis Mona sharma, Markos Markou, Sameer
Singh.
[7] Digital Image Processing Using Matlab - Haris Papasaika-Hanusch
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 39-45

A NOVEL EMBEDDED SYSTEM TO DETECT

THE PRESENCE OF LIFE IN LIMITED
ACCESSIBLE AREAS

PREM KUMAR E.1 AND THIPPESWAMY K. B.2

1
IEEE Graduate Student Member, Dept. of ECE
Acharya Institute of Technology, Bangalore, India
2
Dept. of ECE, Acharya Institute of Technology, Bangalore, India

Abstract
This paper discusses a novel embedded system which is used to detect and locate the presence of life
in limited accessible areas and obtain valuable information regarding the location and health condition
of the trapped life. The proposed robot based system with passive infra-red and heartbeat sensors is
interfaced for detection and the GPS module locates the position of the trapped life which helps in
reducing the time of operations that is otherwise wasted in just searching for the trapped life by the
rescue personnel. Hence it facilitates the operations of rescue teams involved in saving the lives of
people trapped in limited accessible area. A hardware prototype of the life detection system has been
developed and experimental results show that the proposed method is cost effective and efficient
which can be used to detect as well as locate the victims trapped in limited accessible areas without
risking life of rescue personnel.

-----------------------------------
Keywords: Infra-red, Heartbeat sensor, GPS, Life Detection, Hyper terminal

© http: //[Link]
40 PREM KUMAR E. AND THIPPESWAMY K. B.

1. INTRODUCTION
When life is trapped in limited accessible areas some of the existing methods used to detect
presence of life are the utilization of trained dogs, optical devices like binoculars, acoustic life
detectors and temperature sensor based rescue robots [1]. The dogs, risking their own life,
detect dead persons as well, which occupies the precious time that can be utilized to detect
only alive people. Optical devices, requiring expert operators, have limited degree of freedom
and cannot be relied on in the case of poor light. Acoustic life detectors require quiet working
environments, which is difficult to find in critical situations [6-7]. The sensor based rescue
robots, being very expensive, make use of the temperature sensors. Their behavior is very
vulnerable to the environment in which it is present which might eventually lead to false
alarms [2-5]. Due to the above drawbacks, there is a great demand to construct a sensitive life
detecting system which can be used to detect as well as locate the victims trapped in limited
accessible areas without risking life of rescue personnel. We have constructed an embedded
system to serve the above purpose.
Under the actual situation of the person being trapped in a limited accessible area,
information about the location of trapped person and details regarding the health condition
of the human subject would be of great value for the rescue personnel. This would help in
reducing the time of operation and thereby, help to save more lives. This system also aims
in saving lives without risking any more lives of people involved in saving them .Hence
taking all the requirements into consideration, a cost effective system is constructed which
can detect the presence of life in a limited accessible area most efficiently and in a shorter
span of time without risking anyone’s life.

2. PROPOSED LIFE DETECTING SYSTEM

The novel idea of this paper is to develop efficient and cost-effective life detection system.
The prototype consists of three sections:
 Input section or System controlling section.
 Processing section
 Output section.
A NOVEL EMBEDDED SYSTEM TO DETECT… 41

Figure 2.1 : Structural Outline of the Model

Figure 2.2 : Processing Section

42 PREM KUMAR E. AND THIPPESWAMY K. B.

The diagram shown in Fig 1 explains the structural outline of the model which shows the
overview of the system and the connection of different components used in the system. The
Fig 2 explains the processing section which involves the microcontroller 8051 and the
different components interfaced with it [9- 10].
2.1 Principle of working
In this module, the robot is controlled through laptop. When the robot is in the limited
accessible area life can be detected immediately through the camera as we can view the live
feed obtained from the camera on the LCD monitor. In the case of bad light/no proper image
reception the camera is not sufficient for the purpose then the slightest motion of the human
present in the area triggers the PIR and output will be sent to the laptop stating the message
as ‘Human Detected’.
If the detected human is not moving then the heartbeat sensor attached to the movable arms
is used to confirm that the detected human is alive or not [8, 11] which is known to us by the
message sent to the laptop as ‘Pulse Detected / Pulse Not Detected. Hence this system is
used to detect the presence of life trapped in limited accessible area. This system consists of
three sections which are explained below:
 System controlling section or Input section:
Through this section we control and interact with the robot wirelessly. It includes the
laptop, HyperTerminal software, zigbee module, power supply including step-down
transformer, 7805 voltage regulator & filter circuitry.
 Processing section:
Through this section we determine the presence of life in a limited accessible area. It
includes the chained robot, zigbee module, microcontroller, battery, relay board,
camera, heartbeat sensor, PIR sensor, GPS. This section interacts with the person
controlling the system, it takes orders from person with respect to the movement of the
robot and its arms and communicates the output to him.
 Output section:
In this section the processed output (which may be like human detected/ pulse detected/
pulse not detected) and GPS location of the detected human is made available to the
rescue personnel. This section even involves the laptop, HyperTerminal software and
LCD monitor to view the live video as seen from the camera mounted on the chained
robot.
A NOVEL EMBEDDED SYSTEM TO DETECT… 43

3. HARDWARE PROTOTYPE

Figure 3 : Hardware prototype

The hardware prototype is as shown in the fig 3. It consists of the chain robot which is
wirelessly controlled. The robot consists of the components like PIR, Camera, GPS,
Heartbeat detection sensor, Zigbee module, 4 channel relay board, Batteries, Controller
board, motors and much more mounted on it. The hardware prototype also includes the input
controlling section consisting of the circuitry required to convert the AC Voltage of 230V to
DC Voltage of 5V.

3. EXPERIMENTAL SETUP AND RESULTS

The zigbee module is powered through a USB - Serial Port Converter the laptop is connected
to the same zigbee module. A chain robot and its movable arms is constructed with the
camera being mounted on the arms of the chain robot such that it can have the complete
view of the area in front of the robot. The Heartbeat sensor is fixed firmly in between the
movable arms.
The PIR sensor and GPS module are placed on the robot. Battery powers different
components like the motor drivers, the controller board, PIR sensor, GPS, relay board and
zigbee module. All the components are mounted on the robot, which are interfaced with the
controller accordingly. The microcontroller is programmed using Keil software and written
44 PREM KUMAR E. AND THIPPESWAMY K. B.

code controls the robot movements, heartbeat sensor output, PIR output and every other
control through different ports.
Through the laptop we view the output sent from processing section. Along with it we use a
LCD monitor to view the live video from the camera. TV tuner is used for frequency
matching so that live feeding is received and viewed successfully. Once all the above phases
are completed, setup is ready to be tested.
Case 1: When Human is detected with No Pulse Confirmation

Case 2: When Human is detected with Pulse Confirmation

Figure 4 : Snap Shots of the Output

CONCLUSION
A novel embedded system which detects and locates the presence of life in limited accessible
areas obtaining valuable information regarding the location and health condition of the
trapped life has been developed. The proposed method does not risk life of rescue personnel
involved in saving the lives of trapped people and is cost effective as well as very efficient.
This system can be improvised in the future by automating the whole system with the robot
not being under manual control. The robot should be able to guide its path towards the
trapped life without any human assistance.
A NOVEL EMBEDDED SYSTEM TO DETECT… 45

REFERENCES

[1] Bäckström Carl-Johan, Christoffersson Niclas “Urban Search and Rescue- An evaluation of
technical search equipment and methods”, Department of Fire Safety Engineering, 2006.
[2] Colla V, Sabatini A M “A composite proximity sensor for target location” IMEKO sixth
international symposium on measurement and control in robotics Brussels, 1996.
[3] korba L Elgazzar S Welch T “Active infrared sensors for mobile robots,” IEEE transactions on
instrumentation and measurement vol 2(43), 1994.
[4] International Journal of Embedded Systems and Applications (IJESA) Vol.2, No.3, September
2012 DOI: 10.5121/ijesa.2012.2313 117: a new revolutionary infrared life detection system
USING atmega168 by Ashutosh Gupta and Betsy Thomas, 2012.
[5] Maamoun, M, Dept. of Electronics, Blida University, Algeria , Benbelkacem A , “Video signal
generation using a new interfacing technique for computer system “.
[6] Lin Jian-San, Chang Sheng-Fuh, Chang Chia-Chan, Hsu Chien-Kai and Hung Chia-Yin “An
Electromagnetic Vital Signal Detection Technique for Human Subjects Buried Under
Earthquake Rubbles”, 2005.
[7] Gilberto Grazzini, Massimiliano Pieraccini, Filippo Parrini, Alessandro Spinetti, Giovanni
Macaluso, Devis Dei and Carlo Atzeni, Dept. of Electronics and Telecommunications,
University of Florence, Italy “An Ultra-Wideband High-Dynamic Range GPR for detecting
buried people after collapse of buildings”, 2010.
[8] Obeid1 D. Sadek2 S., Zaharia1 G., and Zein G. El “Touch-less Heartbeat Detection and
Measurement-based Cardiopulmonary Modelling” , published in "Engineering in Medicine and
Biology Society, Buenos AIres : Argentine 2010.
[9] Mazidi M A, The 8051 Microcontroller and Embedded Systems using assembly and C, Second
edition, Pearson Publications, 2006.
[10] R Kamal Embedded Systems – Architecture, Programming and Design, Second Edition,
McGraw-Hill Publications, 2008.
[11] Everett H.R, Sensors for mobile robots, AK peters ltd, Wellesley MA, 1995.
International [Link] [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 47-55

LOSSLESS INDUCTOR CURRENT SENSING METHOD

WITH IMPROVED FREQUENCY RESPONSE

PRASEETHA A.1 AND SUBRAMANYA BHAT2

1
Student, 2Assistant Professor
St. Joseph Engineering College Vamanjoor, Mangalore, India.

Abstract
The lossless inductor current sensing method is well known and is frequently applied in the
measurement of output current in dc–dc converters due to its low cost and simplicity. This technique
uses a low-pass filter that is matched to the inductance (L) and winding resistance to compensate for
induced voltages due to the inductor. However, the waveform fidelity above the corner frequency of
the inductor is generally poor due to large production tolerances and thermal drift. In this letter, we
propose the use of a coupled sense winding, which increases the corner frequency of the sensing
circuit by more than two orders of magnitude. We show, as an example, that for a 3.85 μH inductor
the corner frequency of the measurement circuit can be increased from 36 Hz using the conventional
approach to 5.8 kHz using the coupled sense winding method. Above the new corner frequency, a
low-pass filter is still required but may now be constructed using a smaller capacitor and with
improved high-frequency response.

----------------------------------
Keywords : Current measurement, current sharing, dc–dc power converter, mutual coupling.

© http: //[Link]
48 PRASEETHA A. AND SUBRAMANYA BHAT

1. INTRODUCTION

Measuring the output current of dc–dc power converters requires the use of low-cost, low-
loss current sensors with reasonable accuracy at small size and satisfactory bandwidth. A
common technique is the lossless inductor current sensing method (further referred to as the
conventional method), which filters the voltage across the output inductor by means of a
low-pass filter in order to determine the current flowing through it[1]. This technique
measures the voltage drop across the winding resistance of the inductor, represented by
resistor R in Fig. 1. This can be achieved by matching the time constants of R1, C1 , and L,
R:

τ1 = τ → = (1)

If the compensation network is perfectly matched with the inductor, the corrected frequency
response of this current measurement method allows an accurate representation of the
current i(t).

Figure 1 : The winding resistance R of the output inductance L inside a power

converter can be used as a lossless measurement of the output current. A
lowpass circuit, whose time constant is matched with L andR, filters out the
induced voltages due to L.
Unfortunately, the match is rarely perfect in practice because of thermal drift of the resistors
R and R1 , capacitor C1 , and inductance L. In addition, the value of L is not a constant and
depends on the dc offset current flowing through the inductor due to variations in the
permeability of the core as a function of core magnetization. Finally, large production
LOSSLESS INDUCTOR CURRENT SENSING METHOD… 49

tolerances of 10% for the inductance and 5% for the low-pass filter make a calibration
procedure mandatory in order to achieve reasonable dynamic performance. Several self-
calibration techniques have been proposed that identify the actual values of L and R to tune
the values of R1 and C1 [2]. Another approach is to employ a self tuning digital current
estimator that accounts for variations in inductance and winding resistance. However, the
implementation of these solutions is not trivial and either requires a customized integrated
circuit or additional computation power.
In this letter, we demonstrate a simple solution that significantly improves the dynamic
performance of the lossless inductor current sensing method before the inclusion of any low-
pass filter and enables the use of filters with higher corner frequencies. However, the
technique does not account for the thermal drift of R, which causes a steady state current
measurement error. This thermal drift is identical to that encountered using the conventional
method, and for high-precision applications needs to be compensated, e.g., by means of a
temperature sensor.

2. THEORY
2.1 Conventional Method
Without a low-pass filter, the voltage across the inductor has a frequency response that is
determined by the inductance L and its winding resistance R.

Figure 2 : The standard inductor current sense method requires a low-pass filter with very low corner
frequency fc . Due to changes in R and L, the corner frequency changes, and an over- or under
compensation may exist, which deteriorates the resulting frequency response above the corner
50 PRASEETHA A. AND SUBRAMANYA BHAT

frequency. The proposed approach is advantageous in that it shifts the corner frequency of the
inductor by two decades and thus gives good waveform fidelity at higher frequencies.

Similar to a shunt resistor, we can determine the measurement bandwidth by using the 3-dB
corner frequency that is given by fc =

Naturally, the resulting measurement bandwidth is much lower than that of a shunt resistor
since the inductor is designed to minimize the value of R for a given inductance value L. As
an example, in our experimental setup we use an inductor with nominal values of 3.85 μH
for L and 0.87 mΩ for R. The resulting measurement bandwidth of this arrangement given by
(2) is only 36 Hz. Since the frequency response of this current measurement technique has
first-order high-pass behavior, we can theoretically achieve a perfect compensation by
applying a low-pass filter that has the same corner frequency as demonstrated in Fig. 1. This
approach has been used extensively over the last few years and is known as the lossless
inductor current sensing method. However, as mentioned in the introduction, it is difficult to
match the inductor corner frequency with that of the low-pass filter due to variations in the
component values resulting in a change of the corner frequency fc given by • Δfc . If the
low-pass filter corner frequency is higher than that of the inductor, under compensation
results, which leads to an overshoot during current transients. The other possible scenario is
that the corner frequency of the low-pass filter is lower than that of the inductor, which
results in an overcompensation. Both scenarios are depicted in Fig. 2. In practice, this means
that the resulting frequency response above 36 Hz deteriorates, and thus the waveform
fidelity is poor.

2.2 Proposed Method of Coupled Sense Winding

Our proposed coupled sense winding method is depicted in Fig. 3. We added a sense
winding with the same number of turns as the inductor L onto a common magnetic core. The
principle can be roughly explained by means of the transformer effect that induces a voltage
into the sense winding that opposes the primary voltage. As a result, only the voltage across
R contributes to the sense voltage vs . Since the current through the sense winding is small,
the wire size can be very small and therefore does not increase the size of the inductor. In
theory, this method allows it to exclusively measure the voltage drop across the inductor
winding resistance without the use of a matched low-pass filter. However, as we will see
LOSSLESS INDUCTOR CURRENT SENSING METHOD… 51

later, a small mutual inductance remains that makes it necessary to use the low-pass filter
(Fig. 3) in order to maintain high-frequency fidelity.

Figure 3 : A coupled sense winding automatically compensates the voltage induced

by inductance L so that, in theory, the sense voltage vs is exclusively determined
by the voltage drop across R.

Fig. 2 demonstrates that the resulting frequency response of the proposed coupled sense
winding technique has a much higher corner frequency (experimentally verified in Section
III) than the conventional method. Due to this improvement, any over- or under
compensation that might occur due to a mismatched low pass filter occurs at a much higher
frequency, and thus, allows excellent waveform fidelity up to much higher frequencies.

3. EXPERIMENTAL RESULTS
The configuration of a dc–dc buck converter circuit [3] is depicted in Table I. The current
through the inductor, L, takes the form of a triangular current waveform on a larger dc
current, shown in Fig. 7. Using the conventional method without a low-pass filter (R1 = 0
and C1 = 0), the sense voltage vs is given by the sum of Faraday’s induction law and voltage
drop across the winding resistance
vs = L. + iR. (3)
52 PRASEETHA A. AND SUBRAMANYA BHAT

Table 1 : Measurement Setup

Figure 7 : The current through the inductor of a dc–dc buck converter is a triangular
wave shape with a dc offset.

The measurement result for this configuration is shown in Fig. 8(a). A rectangular voltage of
substantial magnitude is observed due to the large Faraday induction component. Using the
configuration given in Table I, we calculate the Faraday induction component that gives 7 V
during the rising slope and −12Vduring the falling slope. The voltage contribution from iR is
negligibly small. This is in good agreement with the observed result.
LOSSLESS INDUCTOR CURRENT SENSING METHOD… 53

After the implementation of the coupled sense winding, the measured waveform in Fig. 8(b)
was observed. The induced voltage is, obviously, much smaller than before so that the iR
term of (3) results in a notable offset voltage. If we remove this offset voltage (iR = 30

A・0.87 mΩ = 26.1 mV), we obtain an induced voltage for the Faraday component of (3) of

45 mV during the rising slope and –75 mV during the falling period. Since the inductance
value is proportional to the induced voltage, the mutual inductance is –12 V/–0.075 V ≈ 160
times smaller than the nominal inductance value L. This yields a value for the mutual
inductance of 3.85 μH/160 = 24 nH. Equation (2) gives a corner frequency of 5.8 kHz for the
coupled sense winding method with 0.87 mΩ winding resistance.

Figure 8 : Measurement of the inductor voltage with a dc output current of 30 A.(a)

Conventional approach without compensation filter.

54 PRASEETHA A. AND SUBRAMANYA BHAT

Figure 9 : Measurement of the inductor voltage with a dc output current of 30 A

(b) Proposed approach using a sense winding.

Figure 10 : Measurement of the inductor voltage with a dc output current of 30 A

(c) Proposed approach combined with a low-pass filter having a cutoff frequency of
5.8 kHz.

This is a significant improvement over the 36 Hz corner frequency of the bare inductor.
In the measurement shown in Fig. 8(c), a low-pass filter has been applied with a cutoff
frequency of 5.8 kHz in order to compensate the mutual inductance. It should be noted that
the higher corner frequency allows for the use of smaller capacitors. Large capacitors
LOSSLESS INDUCTOR CURRENT SENSING METHOD… 55

generally come at higher cost and provide poorer high-frequency performance. As the
measurement confirms, the low-pass filter corrects the frequency response above 5.8 kHz,
reproducing the current waveform through the inductor. In addition, the coupled sense
winding method is not affected by variations in the performance of the inductor core material
and thus shows no dependence on temperature of the core or dc offset current, unlike the
conventional method.

CONCLUSION
We presented the coupled sense winding method to substantially enhance the waveform
fidelity of the conventional lossless inductor current sense technique by increasing the
apparent corner frequency of the inductor. The improvement is achieved with a sense
winding, having an equal number of turns to the main winding. The mutual inductance of the
sense winding is reduced to more than two orders of magnitudes smaller than the self-
inductance by means of minimizing the area enclosed by the sense wires. Other advantages
are as follows:
1) the mutual inductance is temperature-stable and independent of the dc offset current
through the inductor;
2) increased immunity against magnetic stray fields;
3) the low-pass filter requires a significant smaller capacitor and thus provides lower cost
and better high-frequency behavior;
4) no additional power losses;
5) large output voltage amplitude. In many applications, these advantages will justify the
additional complexity and cost of the coupled sense winding

REFERENCES

[1] Zhou X., Xu P., and Lee F. C., “A novel current-sharing control technique for low-voltage high-
current voltage regulatormodule applications,” IEEE Trans. Power Electron., vol. 15, no. 6, pp.
1153–1162, Nov. 2000.
[2] Garcea G., Saggini S., Zambotti D., and Ghioni M., “Digital autotuning system for inductor
current sensing in voltage regulation module applications,” IEEE Trans. Power Electron., vol.
23, no. 5, pp. 2500–2506, Sep. 2008.
[3] Power electronics, Daniel W Hart
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 57-71

A NOVEL APPROACH IN MODELING AND ANALYSIS

OF VECTOR CONTROLLED SYMMETRICAL
TWO-PHASE INDUCTION MACHINE FOR
SMART GRID APPLICATIONS

RAGHVENDRAPRASAD DESHPANDE, G. K. PURUSHOTHAMA

AND M. S. RAVIPRAKASHA
Department of Electrical and Electronics Engineering,
Malnad College of Engineering, Hassan - 573 202, Karnataka, India

Abstract
Many alternate energy sources are finding their application in the electric power sector to cope up
with the rapid/fluctuating demand of electric energy. Co-ordinated use of power electronics along
with sophisticated electric drives is becoming recent trend in the Renewable Energy Source (RES)
connected systems. Integration of RES into the existing power grid in order to have more reliable and
flexible system is expected to become one of the major components of Smart Grid. An unsymmetrical
Two-phase Induction motor (TPIM) associated with the vector control strategy can be employed as a
smart controlled Self Excited Induction Generator. In an unsymmetrical TPIM, vector control causes
extra coupling between the stator windings. This leads to the complexities in the implementation of
Field Oriented Controlled TPIMs. In this paper, a general model suitable for Indirect Rotor Field
Oriented Control (IRFOC) for TPIM is presented. Simulations are carried out to establish IRFOC
algorithm. The response of the drive for variations in torque and speed are obtained and reported.

----------------------------------
Keywords: Vector control, TPIM, IRFOC, Smart Grid
© http: //[Link]
58 RAGHVENDRAPRASAD DESHPANDE, G. K. PURUSHOTHAMA AND M. S. RAVIPRAKASHA

1. INTRODUCTION
The increasing concern about the environment and the rapidly increasing electrical power demand are
motivating the use of RES in the power sector. Among the renewable energy sources, the wind energy
conversion system is gaining popularity. The trend of using induction machine as a generator along
with the required power electronics control techniques is widely accepted method. The induction
generators can be of 3-phase or single-phase type depending the supply. This work presents the
necessary analytics in terms of speed dynamics to be performed for small rating wind farms where
highly reliable single/two-phase induction generators are used. The speed torque response of induction
machine is studied and analyzed in this work.
Small rating electric motors are produced in many varieties and sizes, and have penetrated to
applications in domestic society and highly automated world of industry in large scale. Single phase
induction motors (SPIMs), Brushless DC motors, and Permanent Magnet Synchronous Motors are
few examples for this category. SPIMs posses two windings and are therefore technically called as
two-phase induction motors (TPIMs). In the research and development field of electrical machine
design, there is a continuous debate on the “motor of the future” for low-power applications requiring
less than 2 kW. The researchers are focusing on special electrical machines (SEM) like Brushless DC,
permanent Magnet Synchronous or Electronically Commutated type Motors. The SEM’s have shown
significant improvements in efficiency, reliability, torque-per-volume/mass ratio, maintenance, and
service life. The vector control principles are based on the balanced electrical systems. Hence there is
a rapid growth in the symmetrical three-phase version of induction motor employing Field Oriented
Control (FOC) schemes. Although AC-drive technology for three-phase induction motors is mature
today, TPIMs still face difficulties when working under variable speed. The winding asymmetry in
TPIMs causes difficulty in the implementation of FOC scheme for TPIMs. The unsymmetrical TPIMs
has different winding turns and wire diameters in their main windings and auxiliary windings, which
leads to different winding resistances. This means that the parameters of the two windings are not
identical.
Although unsymmetrical TPIMs do have established mathematical models, there are no established
mathematical models defining the symmetrical TPIMs for the control schemes. This mathematical
analysis is a prerequisite to any modern control schemes. This work attempts to develop mathematical
model for a TPIM, and then to use this model for analysis. The result of the analysis can be used to
simulate the motor characteristics and thereby to design and implement control schemes. The
controlled operation of induction motor as induction generator along with digital control techniques in
the smart grid environment is becoming a major topic to be addressed. A wind power fed induction
generator can be used for power upregulation [1]. The possibility of exploring wind energy conversion
systems in the implementation of future smart power grid is well investigated in [2]. The induction
A NOVEL APPROACH IN MODELING AND ANALYSIS… 59

generator phenomena for two winding machine is discussed in [3, 4]. Several studies have
concentrated on the mathematical model of unsymmetrical TPIMs [5-8]. The mathematical model for
asymmetrical TPIMs is discussed in [5]. C. Mdemli [Link] [6] investigates the
problem of efficiency optimization in capacitor-run single-phase induction motors.
A SPIM coupled with a TRIAC-based optimal controller could be an ideal solution for most of the
commercial and residential applications since considerable energy savings is achieved as presented in
[7]. A variable SPIM drive by utilizing a smart power module incorporating six switches is discussed
in [8]. The main and auxiliary windings of a single-phase motor are driven by a two-phase inverter
with unbalanced voltage/frequency control to produce smooth torque. Several studies have
highlighted field oriented control scheme for unsymmetrical TPIMs [9-14]. The vector control
of a single-phase induction motor drive to implement low cost systems for low power applications is
discussed in [9]. The vector controlled unsymmetrical TPIM drive for low-power applications is
discussed in [10]. A mathematical model of Field Oriented Control of TPIM using double-sequence
controller to derive control laws is discussed in [11]. However, winding asymmetry in TPIMs causes
extra coupling between two stator windings and results in unbalanced machine operation with torque
pulsations. To overcome the problems of unbalanced operation, an indirect vector control system is
discussed [12] for unsymmetrical TPIMs including a simple and effective decoupling scheme. The
performance of unsymmetrical TPIM drive under nominal load in a closed-loop vector-controlled
system is discussed in [13]. A modulation strategy is developed to synthesize unbalanced two-phase
output voltage waveforms with open-loop V/f control. The additional decoupling signal concept
discussed in [12] and synthesize of unbalanced voltage discussed in [13] leads to the complex
implementation of the drive.
The works reported in [5-13] have mainly concentrated on the development and implementation of
mathematical model of unsymmetrical TPIMS. Also, the methods discussed in the literature does not
highlight on the aspect of elimination of unbalanced operation due to winding asymmetry. Thus, it is
inferred that the vector control strategies for SPIMs are complex and are difficult to implement. To
overcome the limitations of the vector controlled SPIM drive, a symmetrical TPIM model is presented
and is used for the development of control scheme.

2. IRFOC TPIM MODEL

The induction motor is described by different mathematical models, according to the control method
employed. In the vector control strategy, the induction motor operation can be analyzed in a way
similar to a DC motor.
60 RAGHVENDRAPRASAD DESHPANDE, G. K. PURUSHOTHAMA AND M. S. RAVIPRAKASHA

Rotating reference frame is used for the development of mathematical model of symmetrical TPIM.
When a machine is analyzed from a rotating reference frame, stator windings dse & qse and rotor
windings dre & qre referred to the reference frame represents the characteristics of the machine. These
windings are not stationary but they are rotating at synchronous speed. If d e-axis of the reference
frame is oriented along with the rotor flux vector ψ r , following set of equations can be written :
ψdre = ψr , an dψqre = 0 (1)

In (1) ψqre is zero because there is no component of the flux along the q-axis. This concept is very
similar to that of DC machine; in a fully compensated DC machine, field winding is along d-axis.
Even though armature winding is oriented along q-axis, the armature reaction MMF is cancelled by
the compensating winding and it results a very similar equation as that is mentioned in (1).
2.1 Equations for Torque Component

Te= (P/2) × (Lm2/ Lr) × ( iqse idse) (2)

Torque expression for DC motor can be written as:

Te=K ia if (3)
e
On comparing (2) & (3) it is inferred that ia and iqs & if and idse are analogues to each other, i.e. iqse is
the torque component of current and idse is the flux component of current. Vector control scheme for
induction motors mainly attempts in the development of control strategy as that of a DC motor. The
vector control scheme and decoupling concept is given by (2). This is called as rotor flux oriented
control scheme because reference frame is oriented along the rotor flux. The two currents iqse and idse
in (2) are called as controlling parameters. Practically, precise speed regulation and fast torque
response is needed. To achieve good response the hypothetical currents must be transformed into
motor currents ia and ib. The phase current equations in terms of the stationary reference frame must be
transformed to that in terms of the synchronous reference frame in the vector control. The
transformation required for current and voltage components is given by:
id cos θ sin θ 𝑖a (4)
[i ] = [ ][ ]
q cos θ −sin θ 𝑖b

𝑉d cos θ sin θ 𝑉a (5)

[𝑉 ] = [ ][ ]
q cos θ −sin θ 𝑉b
The abc→dq transformation and inverse transformation in the indirect vector-controlled three-phase
ac machine are omitted in the implementation of the vector controlled TPIM as shown by (4) and (5).
Further, the involvement of single current sensor makes vector controlled TPIM relatively more
simple and economical.
A NOVEL APPROACH IN MODELING AND ANALYSIS… 61

2.2 Evaluation of Rotor Flux Linkage

The currents supplied to the machine need to be oriented in-phase and in-quadrature for the rotor flux.
This can be accomplished by choosing ωe with the rotor flux entirely in the d-axis, resulting in the
following expressions:
ψqre = 0; ψdre =constant (6)
Te = (p/2) × (Lm/Lr) × ( iqse ψdre ) (7)

ψdre = Lm idse (8)

The stator direct axis reference current ids* is obtained from rotor flux reference input ψr *, from (8)
stator direct axis reference current can be written as:
idse*= ψr */ Lm (9)

The estimated rotor flux linkage given by:

ψr est=(Lm idse*) / (1+ τ r) (10)

As θe is evaluated indirectly, the control scheme is known as ‘Indirect Rotor Field Oriented Control
Scheme’, popularly known as IRFOC. The rotor flux linkage angle can be obtained as:
θe=∫( ωm+ ωsl) dt (11)
The mathematical model of symmetrical TPIM for vector control scheme is developed as depicted in
Fig. 1. Currents in the stator are measured and are converted into equivalent values in d-q rotating
reference frame. Reference speed is compared with the actual rotor speed and output of the
comparator is fed to PI controller. Corresponding d-q reference values are generated. Rotor flux
vector angle θe is calculated by using (11). Generated reference d-q values are converted into actual
currents ia and ib reference values. These reference currents are then compared with the actual motor
currents using a hysteresis controller and corresponding signals are generated to feed the gate drive
circuit. The transformation employed is the Park’s transformation. It has the unique property of
eliminating all time-varying inductances which occur due to electric circuits in motion and electric
circuits with varying magnetic reluctance.
Among all the PWM methods mentioned, the current regulated pulse width modulation technique
(CRPWM) is popular. Implementation of CRPWM is easy compared to other PWM techniques; hence
it is beneficial to adopt CRPWM technique to generate pulses. However, if a VSI is operated in a
current control mode then switching frequency is variable.
62 RAGHVENDRAPRASAD DESHPANDE, G. K. PURUSHOTHAMA AND M. S. RAVIPRAKASHA

Figure 1 : Control block diagram of IRFOC scheme.

In order to operate a VSI at a constant switching frequency, the current signals i d and iq have to be
converted into voltage signals. The voltage equations for Vd and Vq from generated id and iq reference
values can be obtained as:

Vdse = (rs +σ Ls p) idse+ (Lm/ Lr) ψdre- ωe (σ Ls iqse) (12)

Vqse = (rs +σ Ls p) iqse+ ωe ( Lm/ Lr) ψdre+ ωe (σ Ls idse) (13)

The torque expression given by (2) shows that torque is directly proportional to i d and iq. These
hypothetical currents can be used for control purpose, however for better performance these currents
are converted into voltages by using (12) and (13).

2.3 Control of Torque Component

Torque response of a normal TPIM is given by:
Te = [(P/2)×Lm(iqs idr –ids iqr )] (14)
Above equation can be further simplified and written as:
Te = [(P/2) × (Lm/Lr) × (iqs ψdr –ids ψqr )] (15)
Torque response as given by (15) is a general expression for torque which is also true for a drive in
synchronous reference frame, hence in a synchronous reference frame (15) can be written as:
Te = [(P/2)*(Lm/Lr)*(iqse Ψdre –idse Ψqre )] (16)
In (16), superscript ‘e’ indicates that torque equation is in synchronous reference frame. In a vector
controlled TPIM, the d-q axis are rotating with synchronous reference frame as shown in Fig.2, and d-
axis is aligned with the rotor flux vector arriving at, ψqre=0 and Ψdre=Ψr.
A NOVEL APPROACH IN MODELING AND ANALYSIS… 63

Figure 2 : Orientation of rotor flux in q-axes.

The torque expression can be again simplified and written as:
Te = [(P/2)*(Lm/Lr)*(iqse Ψdre )] (17)
Te = [(P/2)*(Lm2/Lr)*(iqse idse )] (18)
The torque expression obtained in (17) is very similar to (2). Hence, it is inferred that ia and iqse & if
and idse are analogues to each other, i.e. iqse is the torque component of current and idse is the flux
component of current. Torque in vector controlled drive can be controlled as that of a separately
excited DC motor. Further, by varying torque component iqse torque response will be instantaneous.
2.4 Response of the Symmetrical TPIM
To achieve fast/instantaneous response, it is insisted to change the torque component in the above
equation. Generally, in a vector controlled drive; i dse =ψdre=ψr is kept constant. This is mainly done to
avoid sluggish response. Induction motor is fairly a complex machine exhibiting complex control
techniques. Recent developments have led to FOC algorithm to control induction motor like
separately excited induction motor. Torque response of induction motor under the variation of rotor
flux is sluggish. It is because flux is associated with its own inductance value and time constant, these
together offers a considerable delay in the torque response. The response of the induction motor for
variable rotor flux is analyzed in this section. As rotor flux is variable quantity, the equations for flux
linkage are written as:
Ψdre = [( Lm/1+ τ r p) ×ids e ] (19)
From (19) it is inferred that if flux component Ψdre is varied to control torque, then the torque response
of the motor will be sluggish.

3. THE IRFOC ALGORITHM

The trans-vector algorithm used for the decoupling of flux and torque components is as follows:
 Measure the main winding current ia=ids in the d-axis by using a current sensor. If the measured
current in d-axis is Im cos θe by one current sensor, then ib=iqs in the q-axis is Im sin θe.
64 RAGHVENDRAPRASAD DESHPANDE, G. K. PURUSHOTHAMA AND M. S. RAVIPRAKASHA

 Calculate the rotor flux and its orientation.

 Rotate the two-axis coordinate system such that it is in alignment with the flux.
 Using the calculated transformation angle transform stator co-ordinates into a rotating
reference frame.
 This conversion provides the ide and iqe variables from ia and ib. This step is commonly known
as ‘Park Transformation’. This conversion transforms the set of currents which are in
stationary reference frame to that in terms of the synchronous reference frame.
 Generate Flux error signal using reference flux and estimated flux value.
 A PI controller is then used to calculate id*.
 id* & iq* are converted to a set of currents ia* & ib*.
 ia*, ib*, and ia, ib are compared using hysteresis comparator to generate inverter gate signals.

4. RESULTS AND DISCUSSIONS

Using the mathematical models developed, digital simulations were carried out to simulate the torque
responses of the IRFOC scheme using Matlab/Simulink. The simulations are made such that the drive
system is tested for various conditions such as low and high values of a step reference speed
command, with and without load. A symmetrical capacitor-less two winding motor is used for the
development of Simulink model. Response of torque and speed of a symmetrical TPIM are analyzed.
The specifications of the symmetrical TPIM used for the simulation are:
 Nominal power, Voltage, Frequency: 0.5HP, 220V, 50HZ.
 Stator Main winding Impedance/Stator Auxiliary winding impedance: 2.2Ω+ 51e -6 H( Rs+Ls)
 Main winding mutual inductance:43e-3 H(Lm)
The induction motor is modeled in MATLAB/Simulink using the modeling equations (1) to (19). The
Vd and Vq voltages are obtained from equations (12) and (13). These two voltage signals are used to
generate gating signals to the switching elements. Simulations are carried out to test the precise speed
regulation and the results are obtained for:
 Reference speed of 100 rad/sec.
 Reference speed of 100 to 150 rad/sec in the step time of 1.5 seconds.
 Reference speed of -80 to 120 rad/sec in the step time of 1.5 seconds.
Fig. 3 and Fig.4 shows the simulated values of the rotor speed and torque. A reference speed of 100
rad/sec is applied and motor speed reaches to the reference value, however torque command is kept
equal to zero to validate no load condition. The drive system is tested at speed near the rated speed by
applying a step reference speed command value of 100 rad/sec. Fig. 3 shows the rotor speed which
tracks the reference speed at steady state. The 2-phase current response for no-load steady state
A NOVEL APPROACH IN MODELING AND ANALYSIS… 65

condition is shown in Fig. 5. During stating, the currents are having higher values, later they will settle
down to steady state value of around 4 A. Stator currents in winding A and in winding B with small
no-load value at steady state is shown in Fig. 5. Fig. 6 shows the pulses generated by the IRFOC
algorithm; these pulses are fed to switching devices. Motor speed and torque response when the
reference speed changes from 100 to 150 rad/sec are shown in Fig. 7 and Fig.8. The actual value of
speed precisely tracks the reference command. A step-time of 1.5 seconds is set for a step reference
command in the Simulink model; it is inferred from the simulation results that motor responds quickly
to the reference command. The torque response for the change in reference speed is shown in Fig. 8.
Motor speed and torque response when the reference speed changes from -80 to 120 rad/sec are
obtained as shown in Fig. 9 and Fig.10. The reference command is set such that speed changes in the
opposite direction from −80 to 120 rad/s slowly under the no-load condition. A step-time of 1.5
seconds is set for a step reference command in the Simulink model; it is inferred from the simulation
results obtained as shown in below figure that motor responds quickly to the reference command. At
1.5 second speed gradually increases as a response to the change in reference command and it is seen
that at 4 seconds the actual speed reaches to the final value of 120 rad/sec. The torque response for the
change in reference speed is shown in Fig. 10. The variation in torque during speed reversal reaches a
value of high value of 30 N-m for few micro seconds.
Simulations are carried out to test the response of the machine for various reference values of torque
and the results are obtained for:
 Reference torque of 0 to 10 N-m in the step time of 3.5 seconds.
 Reference torque of 10 to 0 N-m in the step time of 2.5 seconds

Figure 3 : Speed response for reference speed of 100 rad/sec.

66 RAGHVENDRAPRASAD DESHPANDE, G. K. PURUSHOTHAMA AND M. S. RAVIPRAKASHA

Figure 4 : Torque response for reference speed of 100 rad/sec

Figure 5 : Two phase stator currents under steady state

Figure 6 : Pulses generated by IRFOC scheme

A NOVEL APPROACH IN MODELING AND ANALYSIS… 67

Figure 7 : Speed response for change in reference speed value from 100 to 150 rad/sec.

Figure 8 : Torque response for change in reference speed value from 100 to 150 rad/sec.

Figure 9 : Speed response for change in reference speed value from -80 to 120 rad/sec
68 RAGHVENDRAPRASAD DESHPANDE, G. K. PURUSHOTHAMA AND M. S. RAVIPRAKASHA

Figure 10 : Torque response for change in reference speed value from -80 to120 rad/sec.

Simulation results for the vector controlled TPIM under variable load conditions are obtained as in
Fig. 11 to 14. A load torque of 0-10N-m is applied in a step time of 3.5 seconds and the response of
the motor is shown in Fig.11. Motor is operated under no load condition till 3.5 seconds, hence it is
seen that the actual torque response till 3.5 seconds is zero. However, as a response to the step
reference command, the motor torque increases up to 10 N-m. The two-phase current response for the
variation in load is shown in Fig. 12. As the motor is operating on no load condition, it is seen that a
small amount of no load current is flowing in the two windings. Sudden change in reference torque
influences current to increase as shown. The step command of 0-10 N-m is applied in the step time of
3.5 seconds, hence it is observed that current start to rise at 3.5 seconds and settles down to new value.
Simulation results are obtained for the drives employing load variations from full load to no load and
the response of the motor when load torque gradually decreases from 10 N-m to 0 N-m. A reference
torque of 10 to 0 N-m is applied in step time of 2.5 seconds and the results obtained are shown in
Fig.13. Motor is initially started with a load torque of 10 N-m. In a step time of 2.5 seconds motor is
operated on no load. It is inferred from the simulation results that the motor response is improved as it
follows the reference command. The two-phase current response for the variation in load is shown in
Fig. 14. As the motor is operating initially on full load, it is seen that a large amount of load current is
flowing in the two windings. Sudden change in reference torque from Full load to zero (10 to 0)
influences current to decrease as shown in Fig.14
A NOVEL APPROACH IN MODELING AND ANALYSIS… 69

Figure 11 : Torque response for variation of torque from 0-10 N-m.

Figure 12 : Two-phase current response under variation of torque from 0-10 N-m

Figure 13 : Torque response for variation of torque from 10-0 N-m.

70 RAGHVENDRAPRASAD DESHPANDE, G. K. PURUSHOTHAMA AND M. S. RAVIPRAKASHA

Figure 14 : Two-phase current response under variation of torque from 10-0 N-m.

CONCLUSIONS
Generally, the single-phase motors are used for domestic purposes. Presently they are used for only
constant speed operation. These motor drives with suitable control strategies can provide comparable
characteristics as that of a DC motor drive. This work is to investigate the use of 1-phase AC motors
instead of DC motors for variable speed applications. The TPIM drive system is useful in low-power
motor drive applications and in cases where only a single-phase supply source is available. Further the
performance of symmetrical TPIM can be improved by using vector control algorithm. The model of
a TPIM is developed and it is used in providing a controlled drive system using the IRFOC approach.
The simulation results shows the improvemnts in the performance of the drive system. This type of
control scheme can be employed in wind energy sources, where a high dynaimc perforamnce of
indcution genearator is needed.

REFERENCES

[1] Juelsgaard Morten, Bendtsen Jan and Wisniewski Rafael, “Utilization of Wind Turbines for
Upregulation of Power Grids”, IEEE Transactions on Industrial Electronics, Vol. 60, No. 7, July
2013, pp. 2852-2863.
[2] Arestova Anna and Sidorkin Yury, “The use of wind farms and virtual power plants for
emergency control in the future smart super grids”, the 6th International Forum on Strategic
Technology, August 22-24, 2011, pp. 437-442.
[3] Murthy S. S., Singh Bhim and Vuddanti Sandeep, “A Novel and Comprehensive Performance
Analysis of a Single-Phase Two-Winding Self-Excited Induction Generator”, IEEE Transactions
on Energy conversion, Vol. 27, No. 1, March 2012, pp. 117-127.
A NOVEL APPROACH IN MODELING AND ANALYSIS… 71

[4] Tiwari Arvind K. a,*, Murthy S.S. b, Singh B. b, Shridhar L., “Design-based performance
evaluation of two-winding capacitor self-excited single-phase Induction Generator”, Electric
Power Systems Research 67 (2003), pp. 89-/97.
[5] Lungeanu Blaabjerg, F., Skaug K., and Tonnes M., “Two phase induction motor drives”, IEEE
Transactions on Industrial Applications, Vol. 10, No. 4, Jul./Aug. 2004, pp. 24–32.
[6] Mdemlis C., Kioskeridis I., and Thodoulidis T., “Optimization of single phase induction
motors—Part I: Maximum energy efficiency control”, IEEE Transactions on Energy
Conversion, Vol. 20, No.1, March 2005, pp.187–195.
[7] Mdemlis C., Kioskeridis I., and Thodoulidis T., “Optimization of Single-Phase Induction
Motors—Part II: Magnetic and Torque Performance Under Optimal Control”, IEEE
Transactions on Energy Conversion, Vol. 20, No. 1, Mar. 2005, pp. 196–203.
[8] Rahman M. F. and Zhong L., “A current-forced reversible rectifier fed single-phase variable
speed induction motor drive,” in Proc. 27th IEEE PESC, June. 1996, pp.114–119.
[9] Correa M. B. R., Jacobina C. B., Lima A. M. N., and da Silva E. R. C., “Field oriented control of
a single-phase induction motor drive,” in Proc. IEEE PESC, May 1998, pp. 990–996.
[10] Correa M. B. R., Jacobina C. B., Lima A. M. N., and da Silva E. R. C., “Rotor-flux-oriented
control of a single -phase induction motor drives”, IEEE Transactions on Industrial Electronics,
Vol. 47, No. 4, Aug. 2000, pp. 832–841.
[11] Correa M. B. R., Jacobina C. B., and da Silva E. R. C., “Vector control strategies for single-
induction motor drive system”, IEEE Transactions on Industrial Electronics, Vol.51, No. 5, Oct.
2004, pp. 1051-1080.
[12] Zadeh S. V. and Harooni S., “Decoupling vector control of single phase induction motor drives,”
in Proc. IEEE PESC, 2005, pp. 733–738.
[13] Nied , J. Oliveira, F. R. L. Sa, R. Campos, and L. H. R. C. Stival, “Single-phase induction motor
indirect field oriented control under nominal load,” in Proc. IEEE PEDS, 2010, pp. 789–793.
[14] Krause P.C., Thomas C.H., “Simulation of Symmetrical Induction Machinery”, IEEE
Transactions on Power Apparatus and Systems, Vol. 84, No.11, November 1965, pp. 1038-
1053.
[15] Krause Paul C., “Analysis of Electric machinery”, New York: McGraw-Hill, 1987.
International J. of Multidispl. Research & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 73-81

SIMULATION OF DIFFERENT MEMS PRESSURE SENSORS

VINAY SHETTAR, SNEHA B. KOTIN, KIRANKUMAR B. B.

AND B. G. SHEEPARAMATTI
Dept. of ECE, Basaveshwar Engineering College, Bagalkot 587103, India.

Abstract
MEMS(Micro-ElectroMechanical Systems) pressure sensors are designed to operate in linear range
and presently the most widely used devices. MEMS pressure sensors have gained popularity in
automotive, biomedical and industrial applications. In this paper, the design and simulation of
capacitive, piezoelectric and piezoresistive MEMS pressure sensors are proposed. The piezoelectric
sensor is composed of Lead Zirconate Titanate(PZT-5A), piezoresistive pressure sensor is composed
of n-Silicon(single-crystal, lightly doped) and p-Silicon(single-crystal, lightly doped), the capacitive
pressure sensor is composed of SiC(6H) these pressure sensors deflects due to applied pressure from 0
to 2MPa over it are accounted. The modelling and simulation is carried out for different MEMS
Pressure sensors using COMSOL Multiphysics a MEMS modelling and simulation software.

------------------------------------
Keywords : MEMS capacitive pressure sensor, piezoelectric pressure sensor, piezoresistive pressure
sensor, COMSOL Multiphysics. © http: //[Link]
74 VINAY SHETTAR, SNEHA B. KOTIN, KIRANKUMAR B. B. AND B. G. SHEEPARAMATTI

1. INTRODUCTION
Micro Electro Mechanical Systems (MEMS)[2] are the integration of mechanical elements,
sensors, actuators and electronics on a common substrate using integrated circuit process
sequences. Sensors are the devices that measure information from a surrounding
environment and provides an electrical output signal in response to the parameter it
measured. A sensor detects the quantity that is being measured. Sensors are made from
Silicon to make them low cost. Pressure sensors are the devices that read changes in
pressure and relay data to recorders or switches. Usually, stated as force per unit area.
MEMS based pressure sensors [3] are mechanically similar to traditional sensors with the
exception that these are Silicon based and on micrometer scale. Micro pressure sensors are
widely applied in automotive, biomedical, space, military and various industrial
applications. Pressure sensors fabricated by MEMS technology have the benefits of small
size, low cost and high performance [4][5]. MEMS are systems of small size, light weight,
enhanced performance and reliability finding widest of applications. A typical MEMS
Sensor is at least one order of magnitude smaller than the traditional sensors used to
measure instantaneous flow quantities like Pressure and Velocity. MEMS sensor measure
the pressure in term of deflection of sensing plate.
Nowadays it has become common for scientist and engineers working in micro-electro
mechanical system (MEMS) area to simulate the structures using MEMS simulation
software like COMSOL Multiphysics[1] before actual fabrication. The software helps to
build the structure, mesh it and then simulate it. The main intension in using this software is
to optimize the structure dimensions to get the required output which saves time and money
during fabrication.

2. LITERATURE SURVEY
The literature survey is carried out to find the existing models of pressure sensors. The
survey showed that the Piezoresistivity is the change of resistance of a material when it is
submitted to stress. A piezoresistor pressure sensor [6] measures the applied pressure on one
side of the diaphragm. The stress change in the diaphragm causes the resistance to change.
The capacitive pressure sensor [7] consists of two electrodes namely top electrode and
bottom electrode. Due to applied pressure on the top electrode it deflects towards the bottom
SIMULATION OF DIFFERENT MEMS PRESSURE SENSORS 75

electrode which will results in the change of the capacitance of the system and piezoelectric
pressure sensor [8] measures the applied pressure on one side of the diaphragm. The stress
change in the diaphragm will produce the voltage. That is Piezoelectric effect will produce
the voltage when it is submitted to stress.

3. PRESSURE SENSOR DESIGN

MEMS Capacitive pressure sensor and Piezoelectric pressure sensor are designed with
circular radius of 180µm, the capacitive pressure sensor is composed of SiC(6H) the
piezoelectric sensor is composed of Lead Zirconate Titanate(PZT-5A) the Piezoresistive
pressure sensor is designed with side length of 11µm piezoresistive pressure sensor is
composed of n-Silicon(single-crystal, lightly doped) and p-Silicon(single-crystal, lightly
doped).

4. CAPACITIVE PRESSURE SENSOR

A capacitor is formed by two plates which can store an electric charge. The charge generates
a potential difference which may be maintained using an external voltage. A capacitive
pressure sensor measures a pressure by detecting an electrostatic capacitance change; at
least one electrode of the capacitor is on a moving structure. Capacitive sensors have the
advantage over the piezoresistive type in that they consume less power, but have a non-
linear output signal and are more sensitive to electromagnetic interference. Capacitive
sensors are compatible with most mechanical structures, and they have high sensitivity and
low temperature drift. Figure 1 shows Capacitive pressure sensors

Figure 1 : Capacitive pressure sensor

76 VINAY SHETTAR, SNEHA B. KOTIN, KIRANKUMAR B. B. AND B. G. SHEEPARAMATTI

5. PIEZOELECTRIC PRESSURE SENSOR

Some substances, such as barium titanate, single-crystal quartz, and lead Zirconate-Titanate
(PZT) can generate an electrical charge and an associated potential difference when they are
subjected to mechanical stress or strain. When pressure (stress) is applied to a material it
creates a strain or deformation in the material. In a piezoelectric material this strain creates
an electrical potential difference, a voltage. Figure 2 shows Piezoelectric pressure sensors

Figure 2 : Piezoelectric pressure sensor

6. PIEZORESISTIVE PRESSURE SENSOR

Piezoresistivity is the change of resistance of a material when it is submitted to stress. A
piezoresistor pressure sensor[6] measures the applied pressure on one side of the diaphragm.
The stress change in the diaphragm causes the resistance to change. Figure 3 shows
Piezoresistive pressure sensors

Figure 3 : Piezoresistive pressure sensor

SIMULATION OF DIFFERENT MEMS PRESSURE SENSORS 77

7. SIMULATION AND RESULTS

COMSOL Multiphysics software is used for simulating of MEMS capacitive pressure
sensor to optimize the design, improve the performance and reduce the time of fabricating
process of the device. The objectives of analysis is to verify the deflection of the diaphragm
due to the applied pressure between the diaphragm and the substrate.
Figures 4,6,8 shows simulation setup of capacitive, piezoelectric and piezoresistive pressure
[Link] displacement of diaphragms are achieved for applied pressure ranging from 0 to
2MPa as 0.9065μm, 63.45μm, 1.548410^-7µm for capacitive,piezoelectric and
piezoresistive pressure sensor [Link] the displacement for piezoelectric pressure
sensor is 63.45µm it has high sensitivity.
Figure 5 shows graph of Applied pressure(Pa) Vs Capacitance(pF). As the applied pressure
increases there is linear increase in capacitance. Figure 7 shows graph of Applied
pressure(Pa) Vs Electric potential(V). As the applied pressure increases there is linear
increase in Voltage. Figure 9 shows the distribution of voltage for applied pressure.

Figure 4 : The model of Capacitive pressure sensor

78 VINAY SHETTAR, SNEHA B. KOTIN, KIRANKUMAR B. B. AND B. G. SHEEPARAMATTI

Figure 5 : Graph of Applied pressure(Pa) Vs Capacitance(pF)

Figure 6 : The model of piezoelectric pressure sensor

SIMULATION OF DIFFERENT MEMS PRESSURE SENSORS 79

Figure 7 : Graph of Applied pressure(Pa) Vs Electric potential(V)

Figure 8 . The model of piezoresistive pressure sensor

80 VINAY SHETTAR, SNEHA B. KOTIN, KIRANKUMAR B. B. AND B. G. SHEEPARAMATTI

Figure 9 : The distribution of voltage for applied pressure

CONCLUSION
The MEMS capacitive, piezoelectric and piezoresistive pressure sensors are modeled and
simulated using COMSOL Multi physics. By observing the outputs we conclude that the
piezoelectric pressure sensor is advantageous as it shows the maximum displacement and
hence has the high sensitivity. The fabrication of capacitive pressure sensor is simple and
has the low cost with highest accuracy as compared to piezoelectric and peizoresistive
pressure sensor.

REFERENCES

[1] [Link]

[2] Book on MEMS by Thai ran tsu, edition 5.

[3] Hezarjaribi Yadollah, et al, “Evaluation for diaphragm’s deflection” Arab journal of information
technology, vol. 8, no.2,april,2011
SIMULATION OF DIFFERENT MEMS PRESSURE SENSORS 81

[4] Rijks Th. G. S. M., et al., "Microelectromechanical tunable capacitors for reconfigurable rf
architectures," journal of micromechanics and microengineering 16 (3), pp. 601-611, February,
2006.

[5] Arshak K.L., Morris D., Arshak A., Korostynska O., and Jafer e., "Development of a wireless
pressure measurement system using interdigitated capacitors," IEEE sensors journal, vol.7, no.I,
January, 2007.

[6] Chaurasia Saloni, Chaurasia B.S. “Design and Simulation of Low Pressure Piezoresistive MEMS
Sensor using Analytical Models” 978-1-4673-0455-9/12 ©2012 IEEE

[7] Chang S-P. and Allen M.G., "Capacitive pressure sensors with stainless steel diaphragm and
substrate "Journal of micromechanics and microengineering 14 (4), pp. 612-618,2004

[8] Kutiš Vladimír, Stančík Stanislav, Dzuba Jaroslav, Murín Justín, Lalinský Tibor, Vank Gabriel,
Vallo Martin “Modeling and Simulation of Piezoelectric MEMS Pressure Sensor” modelling of
mechanical and mechatronic systems 2011, september 20 – 22, 2011 herľany, slovak republic
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 83-95

LOAD FLOW ANALYSIS IN AN INTERCONNECTED

SYSTEM WITH OPTIMAL SHUNT COMPENSATION

MEGHA FERNANDES AND SINCY GEORGE

Fellow IEI, Member, IEEE, Department of Electrical Engineering,
Fr. [Link] Institute of Technology, Vashi, Navi-Mumbai, India

Abstract
This paper determines the optimal location of a shunt compensator and its control to improve voltage
profile and power system loadability under steady state condition. A program is developed in
MATLAB platform to find the optimal location and amount of reactive power support by carrying out
load flow analysis to enhance the ability of the system to feed increasing power demand, maintaining
the voltage within limits in addition to avoiding overloading of the transmission lines. The validity of
the program is tested on a five bus system and results obtained are presented in this paper.

-----------------------------------
Keywords : Newton-Raphson load flow, Shunt compensation, power system loadability.

© http: //[Link]
84 MEGHA FERNANDES AND SINCY GEORGE

1. INTRODUCTION
The load flow analysis of the power system are carried out to estimate the magnitude and
phase angle of voltage at various parts of an interconnected power system and the power
flowing in each transmission line. This information is useful for future planning, operation,
economic scheduling, load shedding, and exchange of power between utilities, transient
stability and contingency analysis [1]. The power system is assumed to be balanced in steady
state so as to analyse the system by its single line diagram in per unit basis. Methods
developed to perform load flow analysis are categorized into conventional techniques that are
iterative and intelligent methods that are also called as stochastic or heuristic methods [2].
Gauss-Siedel method, Newton-Raphson method and fast decoupled methods fall in the
category of iterative methods. Fuzzy logic, Genetic Algorithm, Particle Swarm method, and
artificial neural networks are few of the intelligent methods for load flow analysis [3]. The
computational time required by conventional methods is more than the intelligent methods
but it provides accurate results. For a higher order bus system, intelligent methods are
preferred over conventional methods, but it may converge to local optimal solution instead of
global optimal. As the load flow studies are off-line studies, conventional method is selected
for load flow studies. In this paper the load flow analysis is performed by Newton Raphson
method as it gives accurate and reliable results.
The magnitudes of the bus voltages are to be kept within limits so as to ensure good power
delivery to the consumers under any loading conditions. In an interconnected system, the
power flow through the transmission lines depends upon the impedance of the transmission
lines, load connected to the system and the voltage at the buses. The transmission lines are
designed considering maximum power that is required to be transmitted. The p- model of the
transmission is selected for the program. As the power demand is increasing day by day,
overloading of certain transmission lines takes place. Overloading of the lines should be
avoided and the power should be redirected to the consumers through other under loaded
transmission lines using compensating equipment.
Supporting the system with reactive power enhances the ability of the system to deliver good
quality power to the consumers during under loaded and overloaded condition. Classical
methods to improve the voltage profile are by using capacitor banks, reactors and tap
changing transformer, while the active power flow through the line is controlled by phase
shifting transformers. Power electronics based modern compensation device grouped under
LOAD FLOW ANALYSIS IN AN INTERCONNECTED SYSTEM… 85

Flexible AC transmission Systems (FACTS) also improves the system performance and has
the advantages of fast response, better life of switches, less wear and tear etc. But as far as the
load flow analysis for a steady state balanced system is concerned, the devices used in
classical methods and FACTS both are to be modeled in similar way.
Improving steady state system performance is one of the benefits of using FACTS devices
[4]. Series and shunt FACTS devices and their combination are modeled in the load flow in
Matpower package and the optimal location is obtained by using Genetic algorithms [3, 5].
Optimum loadability of the system can also be achieved by using both series and shunt
FACTS devices. Shunt FACTS devices improve voltage of the system and hence enhance
system loadability [6, 8]. Series FACTS devices changes lines flows and improves system
loadability by avoiding overloading or under loading of the transmission lines [7]. FACTS
devices can be located by heuristic methods [3, 5, 7], or analytical methods [8]. This Paper
deals with NRLF analysis with and without shunt compensation. It gives best location for
single shunt compensation in a five bus system so that voltage and loadability are within the
limits.
This paper is organized as follows: section II gives the load flow results of the five bus
system[8] with increase in load, section III gives the methodology to select the location for
reactive power injection section and result of load flow with compensation, section IV
analyses the results of load flow, and section V ends with the conclusion.

2. LOAD FLOW SOLUTIONS USING NRLF WITHOUT COMPENSATION

The load flow analysis of the system is carried out using NRLF (Newton Raphson Load
Flow) to check whether the power transmission network is capable to deliver power to the
consumers during generator outages, increase or decrease in demand, addition of new
generators, unavailability of a transmission line and addition of new lines etc.
The single line diagram of p–section model of a transmission line connecting ith and jth bus is
shown in Fig. 1. Rij, Xij and B are the resistance, inductive reactance and susceptance of a
transmission line. Total susceptance is halved and connected to each bus. Here, shunt
conductance is neglected.
86 MEGHA FERNANDES AND SINCY GEORGE

Bus ‘i’ Bus ‘j’

Xij Rij
Vidj Vjdj

B/2 B/2

Figure 1 : Single line diagram of p – section model of a transmission line

The current flow through the lines connecting ith and jth bus is given by:

The active power (Pij) and and reactive power (Qij ) flows of the transmission line are given
as:

where,
Vi and Vj are the magnitudes at bus ‘i’ and bus ‘j’ respectively.
dj and dj are the phase angle of voltage at bus ‘i’ and bus ‘j’ respectively.
‘a’ is transformer tap ratio between bus ‘i’ and bus ‘j’
Consider a five bus system [9] as shown in the Fig. 2. ‘l’ represents transmission line
connecting the two buses. There are seven transmission lines connecting various buses. The
details of the line data of the system is given in Table I.
LOAD FLOW ANALYSIS IN AN INTERCONNECTED SYSTEM… 87

1 3 4
l-2 l-6

l-1 l-3 l-4 l-7

l-5
2 5

Figure 2 . Single line diagram of an interconnected five bus system [9]

Table 1 : Line Data Of Five Bus System

From bus(i) To bus(j) Rij(p.u.) Xij(p.u.) B/2(p.u.)

1 2 0.02 0.06 0.030
1 3 0.08 0.24 0.025
2 3 0.06 0.18 0.020
2 4 0.06 0.18 0.020
2 5 0.04 0.12 0.015
3 4 0.01 0.03 0.010
4 5 0.08 0.24 0.025

Bus data of the five bus system is shown in Table II. Bus code 0, Bus code 1 and Bus code 2
represents slack bus, generator bus and load bus respectively.

Table 2 : Bus Data of Five Bus System

Voltage Load Generation
Bus Bus
Mag Angle
No Code MW MVAR MW MVAR
( p.u.) (Deg)
1 1 1.06 0 0 0 0 0
2 2 1 0 20 10 40 30
3 0 1 0 45 15 0 0
4 0 1 0 40 5 0 0
88 MEGHA FERNANDES AND SINCY GEORGE

Voltage Load Generation

Bus Bus
Mag Angle
No Code MW MVAR MW MVAR
( p.u.) (Deg)
5 0 1 0 60 10 0 0

Load flow analysis of the five bus system is carried out in MATLAB platform, by Newton
Raphson method. Table III shows bus voltages and the generation at generator bus and slack
bus after load flow analysis.

Table 3 : Voltage at Buses by NRLF

Voltage Load Generation
Bus
Mag Mag
No. MW MVAR MW MVAR
( p.u.) ( p.u.)
1 1.0600 0 0 0 130.1 49.455
2 1.0200 -2.374 20 10 40 -24.12
3 1.0028 -4.783 45 15 0 0
4 1.0008 -5.108 40 5 0 0
5 0.9912 -5.92 60 10 0 0
Total 165 40 170.1 25.327

The load bus voltages are within the limit of 5% Vp.u. The active power flow, reactive
power flow and the losses in the transmission lines are shown in Table IV.

Table 4 : Line flow and Losses

Line Power flow Power loss
From (Bus) To (Bus) MW MVAR MVA MW MVAR
1 2 88.834 39.231 97.11 1.728 -1.309
1 3 41.275 10.222 42.52 1.334 -1.322
2 1 -87.107 -40.53 96.07 1.728 -1.309
2 3 24.568 -0.042 24.56 0.35 -3.041
2 4 27.809 0.187 27.81 0.449 -2.737
2 5 54.73 6.276 55.08 1.175 0.491
LOAD FLOW ANALYSIS IN AN INTERCONNECTED SYSTEM… 89

3 1 -39.941 -11.54 41.57 1.334 -1.322

3 2 -24.217 -2.999 24.40 0.35 -3.041
3 4 19.159 -0.458 19.16 0.037 -1.898
4 2 -27.36 -2.924 27.51 0.449 -2.737
4 3 -19.122 -1.44 19.17 0.037 -1.898
4 5 6.482 -0.636 6.513 0.036 -4.851
5 2 -53.554 -5.785 53.866 1.175 0.491
5 4 -6.446 -4.215 7.702 0.036 -4.851
Total loss 5.109 -14.66

The total loss in the transmission lines is 5.109 MW. The maximum limit of the power flow
through the transmission lines connecting generation and load buses, and interconnected load
buses is considered as 100 MVA. It is seen that the power flow through the transmission
lines is less than 100 MVA.
Assuming a uniform increase (l) in load demand at all the buses and a proportionate
increase in generation, it is observed that under voltage condition exists at few buses and few
transmission lines are over loaded. For increase in l=1.5 to 1.8 the voltage at bus-5 drops
from 0.9498 p.u. to 0.9306 p.u, and bus-4 drops to 0.943 p.u. But, the voltage of the load
buses are to be kept within specified limit to ensure good quality of power delivery to the
consumers. The range of the voltage permitted is considered as  5% of the per unit voltages
[10]. The voltage at bus ‘i’ beyond the tolerance band, results into the error termed as bus
voltage violation (BVV) represented as BVVi. The maximum power transmission limit of
line ‘l’ is given by Slmax. To avoid the overloading of the transmission lines, the power
transferred through the line ‘l’, Sl should be maintained below Slmax. Power flow through the
branch ‘l’ more than Slmax results into error termed as branch overloading (BOL) represented
by BOLl.

3. LOAD FLOW ANALYSIS FOR OPTIMAL SHUNT COMPENSATION

The load flow analysis is carried out by increasing the load demand in steps. The under
voltage condition and overloading condition can be improved by supporting the system by
reactive power. A midpoint shunt compensation is implemented in this paper. A transmission
90 MEGHA FERNANDES AND SINCY GEORGE

line is divided into two equal halves and a new bus is created for reactive power injection.
The possible locations for single reactive power compensation are equal to number of
transmission lines. The best location and the optimum reactive power to be injected can be
obtained by load flow.
Flow chart to find optimal location of shunt compensation and value for enhancing loadability
of the power system is shown in Fig.3. The bus voltage violation (BVV), branch overloading
(BOL) and the error is calculated as per equation 5, 6 and 9 respectively.
For ith bus,

where,
∆Vbi = 1- Vbi, = Constant,Vbi =Voltage at bus ‘i’.

where,
= Maximum limit of the power flow through line ‘l’
= Power flow through line ‘l’ and = constant.
Error can be calculated as:

Hence total error is sum of BVV error and BOL error as:

If all the bus voltages are within the limits, BVVi for all the buses is equal to 1. Similarly, if
the power flow through the transmission lines is less than Slmax, BOLi for all the transmission
lines is equal to [Link] product of all the BVVi is equal to BVV error and similarly, the
product of all the BOLl is equal to BOL error. For example, when the bus voltages and power
flow through the transmission lines are within the limits the value of BVV and BOL will be
unity and hence the error is zero.
LOAD FLOW ANALYSIS IN AN INTERCONNECTED SYSTEM… 91

Start

Initialize l=1.0

Compute
PGi = l*PGi, PLi = l*PLi, QLi = l*QLi

Load Flow Analysis

without compensation
l=l + 0.1

Calculate BVV,
BOL and error

yes

if error = 0

no
Place shunt compensation in line l =1

Load Flow
Analysis Q=Q+∆Q

yes yes
if error  0 if Q < Qmax

no no

Place no
compensation in if l = last line
next line
yes

Select the location with zero error at

maximum l, and minimum value of Q

End

Figure 3 : A flow chart to find optimal location and amount of compensation

92 MEGHA FERNANDES AND SINCY GEORGE

The error for a particular l is calculated and if it is equal to zero, then l is increased in steps,
say 0.1, but if it is not equal to zero, the system is supported by a midpoint shunt
compensation starting with line-1 and the error checked. If after the compensation, the error is
not equal to zero then the value of shunt compensation is increased in steps up to a maximum
limit of 100 MVAR. While if after compensation the error is equal to zero, then the
compensation value is stored and the compensation is removed from line-1 and again the
system is supported with midpoint shunt compensation in next line i.e. line-2. The
compensation value for each line is evaluated to obtain error equal to zero. The minimum
value of reactive power compensation is selected from the list of locations that provides zero
error.

4. RESULT ANALYSIS

The interconnected five bus system shown in Fig. 2 is simulated using m-file in MATLAB. A
NRLF is carried out for increased in steps of 0.1 and error is evaluated. For increase in l =
1.4 the error is equal to zero. Thus bus voltages and the transmission line power flows are
within limits, till loadability l =1.4. For loadability more than l= 1.4, it is seen that the
voltage at bus -5 drops below the limit without compensation and with midpoint shunt
compensation the voltage and the power flow through the line is brought within the limits.
The variation in load bus voltages without compensation and with compensation at midpoint
of line-6 is shown in Fig.4. It is clear from Fig.4 that beyond l = 1.5 the voltage of bus-5
drops below the limit and with compensation in line 6 it not only improves voltage in bus-5
but also improves the other bus voltages.
As increase in loadability till l = 1.4 the error is equal to zero, thus no compensation is
required. The amount of compensation in line which results in zero error for single
compensation for different l is shown in Table V . For example, when l = 1.5, a single shunt
compensator at the midpoint of line-1 with rating 65.87 MVAR, or a compensator with 25.75
MVAR at midpoint of line-2 is required to maintain the voltage at various buses and
loadability in all the lines within the limit. Similarly, minimum reactive power requirement
can be computed for other lines. From the table 5, it is clear that optimal location of a single
shunt compensator is the midpoint of line-6. It is observed that in line-1 for l = 1.5, the
amount of shunt compensation required is 65.8 MVAR but for l = 1.6 and above even with
LOAD FLOW ANALYSIS IN AN INTERCONNECTED SYSTEM… 93

maximum compensation of 100 MVAR the error is not equal to zero. It is represented by ‘X’
in the table.

Figure 4 : Bus voltages with increase in loadability with and without compensation

Table 5 : Shunt Compensation in Lines

Compensation in MVAR
l
Line 1 Line 2 Line 3 Line 4 Line 5 Line 6 Line 7
1.5 65.87 25.75 10.03 10.51 21.37 5.02 10.14
1.6 X 57.37 59.11 46.01 23.84 5.61 14.00
1.7 X X X 92.72 26.42 6.20 22.33
1.8 X X X X 37.03 21.35 30.45
94 MEGHA FERNANDES AND SINCY GEORGE

Loadability of the power system verses the reactive power required to maintain the voltage
and loadability of the line within the limit is shown in Fig. 5. It is observed that the
compensation required at line-6 is minimum, for l=1.5 to l=1.8 and hence it is selected as
the best location for shunt compensation.

Figure 5 : Loadability verses reactive power rating of the single shunt compensation in each
line

CONCLUSIONS

In this paper an iterative method to find the optimal location and the rating of a single shunt
compensator in an interconnected system to maintain the voltage and loadability at different
parts of the power system within the specified limit is presented. It is known that in a radial
system, in order to improve the voltage profile of the system, shunt compensation is to be
placed at the midpoint of the line connecting buses which is having under voltage. However,
in an interconnected system, it is found that a line location other than the line connecting
under voltage buses may also give better results. A five bus system considered in the paper,
shows under voltage at bus-5 and bus-4 for l greater than 1.5. However, it is found that the
shunt compensation placed at the midpoint of the line connecting bus-4 and bus-5 (line-7)
requires reactive power of 30.45 MVAR to maintain the voltage and loadability of the line
within the limits. While, the compensation at the midpoint of line-6, requires less reactive
LOAD FLOW ANALYSIS IN AN INTERCONNECTED SYSTEM… 95

power of 21.35 MVAR. Thus, the optimum location for single reactive power compensation
in the system considered for study is in line-6. The MATLAB based iterative method for
obtaining optimal location of shunt compensation gives good results. The computation time
required by this iterative method will increase with higher number bus system and number of
lines, but the method provides accurate and optimal results. The advantage of the program is
that it provides satisfactory results for higher order systems. The program is tested for IEEE-
14 bus system and satisfactory results are obtained.

REFERENCES

[1] Sadaat Hadi, Power System Analysis, Tata Mc-Graw Hill, 2002
[2] Kothari [Link] Dhillon J.S., Power System Optimization, PHI Pvt. Ltd., 2004
[3] Ghahremani Esmaeil and Kamwa Innocent: “ Optimal Placement of Multiple-Type FACTS
Devices to Maximize Power System Loadability Using a Generic Graphical User Interface”,
IEEE Transactions on Power Systems, Vol. 28, No. 2, May 2013
[4] Hingorani N. G. and Gyugyi L., Understanding FACTS Concepts and Technology of Flexible
AC Transmission Systems. Piscataway: IEEE Press, 1999.
[5] Gerbex Stéphane, Cherkaoui Rachid, and Germond Alain J., “Optimal Location of Multi-Type
FACTS Devices in Power System by Means of Genetic Algorithms ” IEEE Transactions on
Power Systems, Vol. 16, No. 3, August 2001.
[6] Najafi S.R., Abedi M., hosseinian S.H.: “A Novel Approach to Optimal Allocation of SVC
using Genetic Algorithms and Continuation Power Flow”, First International Power and Energy
Conference PECon 2006, November 28-29, 2006, Putrajaya, Malaysia.
[7] Rashed G. I., Shaheen H. I., Cheng S. J.: “Optimal Location and Parameter Settings of Multiple
TCSCs for Increasing Power System Loadability Based on GA and PSO Techniques.” Third
International Conference on Natural Computation (ICNC 2007)
[8] Sharma Nikhilesh Kumar, Ghosh Arindam and Varma Rajiv Kumar, “A Novel Placement
Strategy for Facts Controllers” IEEE Transactions on power delivery, Vol 18., No. 3, July 2003.
[9] Stagg and El Abiad, Computer Methods in Power System Analysis, Mc-Graw Hill, 1968
[10] IEEE-STD-519-1992- ‘IEEE Recommended Practices and requirements for Harmonic Control
in Electrical Power System’
International J .of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 97-109

THE SIMPLE CONTROL SYSTEM TO TOLERATE

THE FAULT IN BLDC MOTOR DUE TO
HALL EFFECT SENSOR FAILURE

SWATHI K.1 AND N. MUNIYAPPA2

1
Student, 2Asst professor, Electrical and Electronics Department,
The Oxford College of Engineering, Bommanahalli, Bangalore, India.

Abstract
This paper gives a fault tolerant control system for Hall Effect sensor failure of Permanent magnet
brushless DC (BLDC) motor. The control system in this paper is capable to detect and identify the
Hall Effect sensor breakdown based on sensor signal. In this paper behaviour of BLDC motor is
studied for Hall Effect sensor breakdown through simulation model. BLDC motor simulation model
are validated first by experimental data under no fault condition. After analysing the simulation result
of for no fault condition, the control technique used to detect the fault in position sensor. The
simulation is carried out for sensor breakdown in phase A for two possible faults. The simple method
used here to generate the hall signal of faulty hall sensor to maintain the performance of the motor in
good condition. The simulation model shows the correct performance of fault tolerant control system
designed.

----------------------------------
Keywords: DSPIC30F2010, Cutoff switch
© http: //[Link]
98 SWATHI K. AND N. MUNIYAPPA

1. INTRODUCTION
The Brushless DC (BLDC) motor is the ideal choice for applications that require high
reliability, high efficiency, and high power-to-volume ratio. Generally speaking, a BLDC
motor is considered to be a high performance motor that is capable of providing large
amounts of torque over a vast speed range. BLDC motors are a derivative of the most
commonly used DC motor, the brushed DC motor, and they share the same torque and speed
performance curve characteristics. The major difference between the two is the use of
brushes. BLDC motors do not have brushes (hence the name "brushless DC") and must be
electronically commutated. Commutation is the act of changing the motor Phase currents at
the appropriate times to produce rotational torque. In a Brush DC motor, the motor assembly
contains a physical commutator which is moved by means of actual brushes in order to move
the rotor. With a BLDC motor, electrical current powers a permanent magnet that causes the
motor to move, so no physical commutator is necessary. A BLDC motor is highly reliable
since it does not have any brushes to wear out and replace. When operated in rated
conditions, the life expectancy is over 10,000 hours. For long term applications, this can be a
tremendous benefit. Whenever a motor breaks down or needs to be replaced, your project, or
part of it, must be shut down. This costs you time and money, perhaps a great deal depending
on how long it takes to replace the worn part or parts and get the application started again.
Although a BLDC motor may cost more than a brushless motor, it will often more than pay
for itself in the amount of work time saved. Brushless DC electric motor (BLDC motors, BL
motors) also known as electronically commutated motors (ECMs, EC motors)
are synchronous motor that are powered by a DC electric source via an
integrated inverter/switching power supply, which produces an AC electric signal to drive
the motor. In this context, AC, alternating current, does not imply a sinusoidal waveform,
but rather a bi-directional current with no restriction on waveform. Additional sensors and
electronics control the inverter output amplitude and waveform (and therefore percent of DC
bus usage/efficiency) and frequency (i.e. rotor speed). BLDC Motors have been used widely
in different industrial and commercial applications since 1970’s. It is a novel DC motor that
is commutated electronically due to absence of brushes. Half Effect sensors are mounted
inside the BLDC motor with electrical degree 120 phase differences to detect rotor position.
Output of each sensor is high (logic ‘1’) for 180 electrical degree and is low (logic ‘0’) for
the next 180 degree with respect to rotor position. In this paper position sensor failure of a
THE SIMPLE CONTROL SYSTEM TO TOLERATE… 99

three phase (star connection) permanent magnet BLDC motor with the inbuilt Hall Effect
sensor. Failure of Hall Effect sensors effect directly on the applied voltages to the BLDC
motor and degrade the performance of overall motor drive. Various possible faults may
happen in a Hall Effect sensor such as flaws in the core, changes in the bias current, change
in the magnetic properties of the ferrite core due to the temperature variations, changes in the
orientation of the induced magnetic field in the sensor. Any of these faults may result to
breakdown of the Hall Effect sensors in BLDC motor. Failure of Hall Effect sensor directly
affects the applied voltage to BLDC motor and degrades the performance of overall motor
drive. The proposed fault tolerant control system should be capable of do the following
tasks. I.e. fault detection, fault identification and remedial strategies. In this paper position
sensor failure of a three phase (star connection) permanent magnet BLDC motor with the
inbuilt Hall Effect sensors is discussed. BLDC motor and three phase VSI drive are shown in
Fig.1.
There are a few research works on Hall Effect sensor failure of BLDC motor specifically for
electric vehicle applications. Jeong et al. have presented a control strategy that provides fault
tolerance to the major sensor faults which may occur in an interior-permanent-magnet-motor
(IPMM)-based electric vehicle propulsion drive system.

Figure 1 : BLDC motor drive system

100 SWATHI K. AND N. MUNIYAPPA

Position sensors fault is detected through difference between the estimated rotor angle and
the actual measured one through a sensor less algorithm based on extended EMF in rotating
reference frame. In this approach reconfiguration to sensor less control scheme is introduced
to rectify the fault and maintain the proper operation of the motor after fault occurrence is
detected. Complexities of sensor less control scheme and transition algorithm to senseless
control are the main drawbacks of the proposed method. A three phase low voltage BLDC
motor is used as a practical test motor to validate simulation results. BLDC motor is
simulated in Simulink using SimpowerSystems library.. The three phase variable source
inverter drive of BLDC motor is simulated using MOSFET switches.

2. CONTROLLING METHODS FOR BLDC MOTOR

An electronic Brushless DC Controller (also known as a Driver, or Electronic Speed
Controller), replaces the mechanical commutation system utilized by a Brush DC Motor, and
is required by most Brushless DC Motors to operate. In a Brushless DC Motor controller,
either a Hall Effect Sensor or Back EMF (Electromotive Force) is used to identify the
position of the rotor. Understanding the orientation of the rotor is crucial to operating the
Brushless DC Motor. The Hall Effect uses three hall sensors within the Brushless DC Motor
to help detect the position of the rotor. This method is primarily used in speed detection,
positioning, current sensing, and proximity switching. The magnetic field changes in
response to the transducer that varies its output voltage. Feedback is created by directly
returning a voltage, because the sensor operates as an analogue transducer. The distance
between the Hall plate and a known magnetic field can be determined with a group of
sensors, and the relative position of the magnet can be deduced. A Hall sensor can act as an
on/off switch in a digital mode when combined with circuitry. Back EMF, also known as the
Counter-Electromotive Force, is caused by a changing electromagnetic field. In a Brushless
DC Motor, back EMF is a voltage that occurs where there is motion between the external
magnetic field and the armature of the motor. In other words, the voltage is developed in an
inductor by an alternating or pulsating current. The polarity of the voltage is constantly the
reverse of the input voltage. This method is commonly used to measure the position and
speed of the Brushless DC Motor indirectly, and due to the lack of Hall Sensors within the
controller, these are often referred to as sensor less [Link] Encoders can also be
added to the Brushless DC Motor, allowing both direction and speed to be determined. More
THE SIMPLE CONTROL SYSTEM TO TOLERATE… 101

precise applications may use Optical Encoders with a third index signal, to determine pulse
per revolution. A sensor motor will directly report to the controller, the current position of
the motor. Driving a sensor motor requires a look-up table. The current sensor position
directly correlates to a commutation pattern for the bridge circuits.

3. BLOCK DIAGRAM
The source to the BLDC motor is AC supply which is converted to DC through VSI. Here
six switch 3 legs inverter is used. MOSFET is used as switch in order to control the speed of
the motor .The speed of the motor is controlled through pot speed controller which is
interfaced to DSP. Starting of motor and controlling the direction of rotation of motor is
controlled using motor start key and direction control key. The output of hall sensors is
indicated through LED’s interfaced to controller. BLDC motor is connected to the controller
through BLDC motor driver i.e. voltage source inverter.

Figure 2 : Block diagram

102 SWATHI K. AND N. MUNIYAPPA

The controller used here is DSPIC30F2010, which is an 28 pin IC. A very good quality of
this controller compared to the other controller is, it has in build 10 bit ADC with six input
channels. Also it has 6 in built PWM channels. PWM signals are generated by the controller
in order to operate the switches in particular sequence. Hall sensor signals are given to the
controller through cut-off switches. The hall sensor signals are indicated through LED’s
interfaced to controller. If the hall sensor fails the cut-off switch will disconnect the
particular signal from sensor.

4. FAULT DIAGNOSIS
A three phase low voltage BLDC motor is used as a practical test motor to validate
simulation results. BLDC motor is simulated in Simulink using SimPower Systems Library.
If the position sensor fails, it causes the output signal either to be constant high (logic 1) or
constant low (logic 0). The output of the position sensor does not change according to the
change in the rotor position. Therefore simulation is carried out for both the cases i.e. for
both logic 0 and logic 1. Here simulation is done for the failure of Hall Effect position sensor
of phase A. The simulink model is shown in Fig.3.
Electronic commutation is done by decoding the position sensor signals. Decoding rules of
Hall Effect signals to choose a proper switching vector of VSI are shown in Table 1.
4.1 Simulink Model

Figure 3 : Simulink model

THE SIMPLE CONTROL SYSTEM TO TOLERATE… 103

Table-1: Decoding rules of Hall Effect signal

ROTORANGLE HALL A HALL B HALL C COND SWITCH

(ELEC DEGREE)

30-60 1 0 1 S1,S4

90-150 1 0 0 S1,S6

150-210 1 1 0 S3,S6

210-270 0 1 0 S3,S2

270-330 0 1 1 S5,S2

330-30 0 0 1 S5,S4

The addition of Hall signals introduced by (1) is a fault signature for Hall Effect sensors
breakdown. Maximum possible value of Hf is 2, where the minimum possible value is 1
(value of Hf should be 1 or 2) for each specific electrical angle section. If Hf value goes over
of these limits Hall Effect sensor failure is detected. Hall Effect sensors Fault Flag (HFF) is
introduced for sensor fault detection. HFF is set to ’1’ if Hf value is more than 2 (it means
that one of the position sensor signals is constant one), HFF is set to ’-1’ if Hf value is less
than 1 (it means that one of the position sensor signals is constant zero) and HFF is ’0’ in
case of no fault. Maximum fault detection time is the time of on electrical rotation of rotor
which is quite fast.
HF=HA+HB+HC

4.2 Case : 1 Hall Effect signal is constant zero

BLDC motor is tested for the case 1 where the hall signal output is constant 0. Hall Effect
sensor fault of phase A(HA=0) is applied to BLDC motor running at 3000 RPM speed. Hall
Signal at constant 0 is shown is Fig.4.
104 SWATHI K. AND N. MUNIYAPPA

Figure 4 : Hall signal at constant zero.

The controller designed in this paper detects the fault in sensor, and then it is capable of
generating the signal corresponding to the particular hall sensor as we have taken Hall sensor
of Phase A (Hall A).After the signal generated by controller, the hall signal is shown in
Fig.5.

Figure 5 : Hall sensor signal after fault is corrected

THE SIMPLE CONTROL SYSTEM TO TOLERATE… 105

The speed and torque curve for case 1 is shown in fig.6.

Figure 6 : Speed and Torque characteristics (HA=0)

Hall Effect sensor failure causes change of the switching signals of VSI and effect directly
on the applied line voltage o.
Effect of various Hall Effect sensors fault on the switching signals of VSI is given in Table
II. Switching signals S1 and S6 are constant zero (switches S1 and S6 remain open circuit)
for HA = 0 fault condition. Change of applied voltages cause variation of the stator phase
currents of the BLDC motor that effects directly on electrical torque of motor and increases
torque ripple.
Table 2 : Sensor fault effect of switching sequence
Fault Type Open Switches of VSI

HA=0 S1,S6

HA=1 S5,S5
106 SWATHI K. AND N. MUNIYAPPA

Fault Type Open Switches of VSI

HB=0 S3,S2

HB=1 S1,S4

HC=0 S5,S4

HC=1 S3,S6

4.3 Case 2 : Hall Effect signal is constant one

BLDC motor is tested for the case 2 where the hall signal output is constant 1. Hall Effect
sensor fault of phase A (HA=1) is applied to BLDC motor running at 3000 RPM speed. Hall
Signal at constant 1 is shown is Fig.7.

Figure 7 : Hall signal is constant one

It can be seen that speed and torque responses of BLDC motor for HA = 1 are quite similar
to the ones for HA = 0 fault condition. However by checking switching signals of VSI, it can
be observed that switches S2 and S5 are constant zero (switches S2 and S5 remain open
circuit) after fault occurrence which are not the same as previous fault condition. Therefore
applied voltages to the motor are completely different for HA = 1 fault condition. Hall sensor
signal after fault is detected is shown in Fig.8.
THE SIMPLE CONTROL SYSTEM TO TOLERATE… 107

Figure 8 : Hall sensor signal after fault is corrected

The speed and torque curve for case 1 is shown in Fig.9.

Figure 9 : Speed and torque characteristics (HA=1)

5. REMEDIAL STRATERGIES
After identification of the faulty Hall Effect sensor, corresponding sensor signal is
disconnected and is substituted with a generated commutation signal by DSP controller.
108 SWATHI K. AND N. MUNIYAPPA

Commutation signal is generated based on 120 electrical degree phase difference from the
other two Hall Effect signals.
5.1. Flow chart

CONCLUSION
Fault tolerant control system for Hall Effect position sensors failure of BLDC motor is
discussed in this paper. Behavior of BLDC motor is studied for position sensor failure
situations. BLDC motor is implemented on the verified simulation model. A knowledge
THE SIMPLE CONTROL SYSTEM TO TOLERATE… 109

based table is developed to identify the faulty sensor by analyzing the simulation [Link]
BLDC motor runs according to the signal generated from hall sensors. If any of the one
sensor fails then it directly affects on the performance of the motor. As BLDC motor runs
according to the signal generated from hall sensors. If any of the one sensor fails then it
directly affects on the performance of the motor. The controller in this project detects,
identifies the fault in hall sensors. After identifying the fault, controller will generate the
signal corresponding to the faulty sensor output so that it won’t affect the performance of the
motor. In real time applications, there will be chances of failure of hall sensors so that motor
can stop immediately which will create problems. There will be particular sequence in which
the switches should operate, so the particular hall signal is generated by controller which will
make the motor to run in same speed continuously even if the hall sensor fails and will
match the sequence of switches operation.

REFERENCES

[1] Tashakori A., Ektesabi M., and Hosseinzadeh N., “Characteristics of suitable drive train for
electric vehicle,” in International Conference on Instrumentation, Measurement, Circuits and
Systems (ICIMCS 2011), Vol. 2, pp. 51–57, ASME, 2011.
[2] Tashakori A. and Ektesabi M., “Comparison of different pwm switching modes of bldc motor as
drive train of electric vehicles,” World Academy of Science, Engineering and Technology, vol.
67, pp. 719–725, 2012.
[3] Balaban E., Saxena A., Bansal P., Goebel K., and Curran S., “Modeling, detection, and
disambiguation of sensor faults for aerospace applications,” IEEE Sensors Journal, vol. 9, no.
12, pp. 1907–1917, 2009.
[4] Samoylenko N. B., Han Q. C., and Jatskevich J., “Dynamic performance of brushless dc motors
with unbalanced hall sensors,” IEEE Transactions on Energy Conversion, vol. 23, no. 3, pp.
752–763, 2008.
[5] Jeong Y.-S., Sul S.-K., Schulz S., and Patel N., “Fault detection and fault-tolerant control of
interior permanent-magnet motor drive system for electric vehicle,” IEEE Transactions on
Industry Applications, vol. 41, no. 1, pp. 46–51, 2005.
[6] Wang L., Liu J., and Wu X., “Fault analysis on driving motors of lunar rover wheels,” (Beijing),
20-23 Aug 2011. International Conference on Electrical Machines and Systems, ICEMS 2011.
[7] Liu X.-Q., Zhang H.-Y., Liu J., and Yang J., “Fault detection and diagnosis of permanent-
magnet dc motor based on parameter estimation and neural network,” IEEE Transactions on
Industrial Electronics, vol. 47, no. 5,pp. 1021–1030, 2000.
[8] DSP based electromechanical motion control by Hamid A. Toliyat and Steven G. Campbell.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 111-124

CONVERSION OF SOUND ENERGY TO ELECTRICAL

ENERGY USING PIEZOELECTRIC CRYSTAL

APEKSHA LOKARE, POOJA PATIL AND SOUMYA ASANGIMATH

Abstract
The need for electricity is growing very rapidly. Energy is the basic necessity for life. We all know
energy provides us luxury and comfort. But the electricity produced is not able to satisfy the present
demands and it is very difficult to imagine life without electricity. Since there is a constant depletion
of energy sources we need to find some non conventional sources of energy to generate electricity. In
this Modern World there is lot of noise pollution on roads, airports, in industries etc. There are many
opportunities for converting vibrations into electrical power. Vehicular traffic produces intense
mechanical disturbances. We can think of converting this sound energy into electricity. Sound is a
vibration that propagates as a mechanical wave of pressure and displacement, through some medium
(such as air or water). Sometimes sound refers to only those vibrations with frequencies that are
within the range of hearing for human. The pressure created by the sound could be used to convert it
into electric energy. We know mechanical energy can be converted into electricity. Piezo material
converts mechanical strain into electric energy this property of piezo material could be used to make a
device which would be able to sustainably convert the sound energy to electric energy. Electricity can
be produced by the noise of the vehicles and can be used for lighting the street lights, from
conversations, mobile phones can be charged, and the electricity generated can also be used for small
household applications.

---------------------------------
Keywords: Energy Harvesting, Piezoelectricity, conversion, Piezoelectric crystals, vibrations, Sound
energy, Electricity. © http: //[Link]
112 APEKSHA LOKARE, POOJA PATIL AND SOUMYA ASANGIMATH

1. INTRODUCTION
As we all know the advancement in technology has reached to such an extent that life
becomes miserable without them. The day to day activities which we carry out with the
gadgets, appliances all require electricity. And due to rapid rise in the population and drastic
progress of mankind the need for electricity is increasing, but the production of electricity is
limited and is not meeting to that extent to satisfy the needs. Therefore there is scarcity of
electricity all over the world as development is taking place at a very high rate. It is
necessary to increase the supply of electric power for that it is very essential for us to find
other alternative methods to produce electric-energy.
The most common thing that we encounter around us the most is noise pollution and as it is
freely available we can convert this unwanted sound into electric energy and use it for some
productive purpose.
Sound is a form of energy that is transmitted by pressure variations which the human ear can
detect. The vibration can come from a tuning fork, a guitar string, the column of air in an
organ pipe, the diaphragm of a loudspeaker, the vocal cords, or virtually anything that
vibrates in a frequency range that is audible to a listener. As sound energy is a mechanical
energy it can be converted to electricity. When a device is subjected to vibration, an inertial
mass can be used to create movement. This movement can be converted to electrical energy
using piezoelectric crystals. Sound energy could be easily converted into heat energy which
could be easily converted into electricity but it is not highly efficient as the loss in
conversion will be more.
CONVERSION OF SOUND ENERGY TO ELECTRICAL… 113

Piezoelectricity is the ability of certain crystals to produce a voltage when subjected to

mechanical stress. The word is derived from the Greek word piezein, which means to
squeeze or press. The effect is reversible; piezoelectric crystals, subjected to an externally
applied voltage, can change shape by a small amount. Piezoelectric crystals convert
mechanical energy from pressure, vibrations or force into electricity. They are capable of
generating electrical charge when a mechanical load is applied on them. Due to their inherent
ability to detect vibrations, piezoelectric materials have become a viable energy scavenging
source. The piezo materials exhibit
[1] Direct piezo electric effect is the production of electricity when the crystals are
mechanically stressed .
D = d . T + εT . E (1)
[2] Converse piezo electric effect is the stress or strain in the crystals when an electric
potential is applied.
S = sE . T + dt . E (2)
Where D is the electric displacement vector, T is the stress vector, εT is the dielectric
permittivity matrix at constant mechanical stress, sE is the matrix of compliance coefficients
at constant electric field strength, S is the strain vector, d is the piezoelectric constant matrix,
and E is the electric field vector. The subscript t stands for transposition of a matrix.
This effect occurs naturally in quartz crystals, but can be induced in other materials, such as
specially formulated ceramics consisting mainly of lead, zirconium and Titanium (PZT).In
order to “activate” the piezo properties, the material is first heated to its Curie temperature.
114 APEKSHA LOKARE, POOJA PATIL AND SOUMYA ASANGIMATH

Piezoelectric crystals can be divided into “hard” and “soft” crystals.

Hard doped PZT materials can be exposed to high electrical and mechanical stresses. The
stability of their properties makes them ideal for high-power applications.
Soft doped piezoelectric crystals are distinguished by a comparatively high domain mobility
and thus “ferro electrically soft” behavior, i.e. relatively easy polarization. Quartz, topaz,
Rochelle salt are some of the other examples of piezoelectric crystals. The most common
crystals used is lead zirconate titanate crystal.
Naturally occurring crystals
Berlinite (AlPO4), Cane sugar, Quartz, Rochelle salt, Topaz, Tourmaline Group Minerals,
and dry bone (apatite crystals).
Man-made ceramics
Barium titanate (BaTiO3), Lead titanate (PbTiO3), Lead zirconate Computational
Intelligence Applications to Renewable Energy-2012 titanate (Pb[ZrxTi1-x]O3 0<x<1) -
More commonly known as PZT,
Potassium niobate (KNbO3), Lithium niobate (LiNbO3), Lithium tantalate (LiTaO3),
Sodium tungstate (NaxWO3), Ba2NaNb5O5, Pb2KNb5O15.
Polymer
Polyvinyledene fluoride (PVDF).
The piezoelectric materials that exist naturally as quartz do not have interesting properties
for the production of electricity, however artificial piezoelectric materials such as PZT (Lead
Zirconate Titanate) present advantageous characteristics. Piezoelectric materials belong to a
larger class of materials called ferroelectrics. One of the defining traits of a ferroelectric
material is that the molecular structure is oriented such that the material exhibits a local
charge separation, known as an electric dipole.
CONVERSION OF SOUND ENERGY TO ELECTRICAL… 115

Throughout the artificial piezoelectric material composition the electric dipoles are
orientated randomly, but when a very strong electric field is applied, the electric dipoles
reorient themselves relative to the electric field; this process is termed poling. Once the
electric field is extinguished, the dipoles maintain their orientation and the material is then
said to be poled. After the poling process is completed, the material will exhibit the
piezoelectric effect.
In a piezoelectric crystal, the positive and negative electrical charges are separated, but
symmetrically distributed, so that the crystal overall is electrically neutral. When a stress is
applied, this symmetry is disturbed, and the charge asymmetry generates a voltage.
Electricity can be generated directly from sound energy by piezoelectric effect. The way it
works is that the mechanical energy of sound is applied directly to a crystal (or possibly a
ceramic) with strong piezoelectric characteristics, and the crystal will generate a small
amount of voltage in response to the application of that mechanical energy (sound). What we
are doing is "squeezing" the crystal. A squeeze will generate a small voltage for the
duration of the squeeze. When the crystal is released, another small voltage will be generated
in the opposite polarity.
Advantages of Using Piezoelectric Materials
• Small size
• Broad frequency range
• 2-wire operation
• Ultra low noise
• Wide dynamic range
• Wide temperature range
• Simple signal conditioning
• Cost effective implementation

2. LITERATURE REVIEW
There are a number of applications in which piezoelectric crystals are used for generation of
electricity and the previous applications in which they have been used are as follows
2.1 Footsteps Power Generation
The piezoelectric crystal arrays are laid underneath the pavements, foot paths and other high
traffic areas like highways, speed breakers for maximum voltage generation. The voltage
116 APEKSHA LOKARE, POOJA PATIL AND SOUMYA ASANGIMATH

thus generated from the array can be used to charge the chargeable Lithium batteries,
capacitors etc. These batteries can be used as per the requirement.
2.2 Power Generating Shoes
In United States Defense Advance Research Project Agency (DARPA) initiated a innovative
project on Energy harvesting which attempts to power battlefield equipment by piezoelectric
generators embedded in soldiers' boots. However, these energy harvesting sources put an
impact on the body. DARPA's effort to harness 1-2 watts from continuous shoe impact
while walking were abandoned due to the discomfort from the additional energy expended
by a person wearing the shoes.
2.3 Generation from Vibrations
Researchers are also working on the idea of utilizing the vibrations caused from the
machines in the gym. At workplaces, while sitting on the chair, energy can be stored in the
batteries by laying piezoelectric crystals in the chair. Also, the studies are being carried out
to utilize the vibrations in a vehicle, like at clutches, gears, seats, shock-ups, foot rests.
2.4 Mobile Keypad and Keyboards
The piezoelectric crystals can be laid down under the keys of a mobile unit and keyboards.
With the press of every key, the vibrations created are used by piezoelectric crystal and
hence can be used for charging purpose.
CONVERSION OF SOUND ENERGY TO ELECTRICAL… 117

2.5 Floor Mats, Tiles and Carpets

A series of crystals can be laid below the floor mats, tiles and carpets which are frequently
used at public places.
2.6 People Powered Dance Clubs
In Europe, certain nightclubs have already begun to power their night clubs, strobes and
stereos by use of piezoelectric crystals. The crystals are laid underneath the dance floor.
When a bulk of people use this dance floor, enormous amount of voltage is generated which
can be used to power the equipments of the night club. A famous night club in London
exploited the principle of piezoelectricity in making its dance floor. The dance floor uses
piezoelectricity where crystal and ceramics create a charge to generate electricity. The
nightclub has a "bouncing" floor made of springs and a series of power generating blocks
which produce a small electrical current when squashed. As dancers move the floor up and
down to squeeze the blocks, the current is fed into nearby batteries which are constantly
recharged by the movement of the floor. The electricity created in this way is used to power
parts of the nightclub such as the sound and lighting.

2.7 Existing Approach

There is already an existing approach for converting sound to electricity .we could convert
sound energy to heat energy as sound wave travel by oscillating the particles of the medium
118 APEKSHA LOKARE, POOJA PATIL AND SOUMYA ASANGIMATH

so when sound energy travel through the medium it will disturbs the particle of the medium
these disturbances created by sound will be used to convert it into heat energy as when the
particles of the medium will be pushed by the sound wave it will collide with adjacent
particle of the medium this collision will result in production of heat energy the production
of heat energy will be more in the denser medium so for more heat production we will need a
material with very high density. This heat energy will be converted into electricity.
Production of heat energy will be more in the denser medium so for more heat production we
will need a material with very high density. This heat energy will be converted into
electricity. But this method is less efficient due to more energy losses taking place while
conversion of sound energy to heat energy and then heat energy to electrical energy.

3. OVERVIEW
The main objective of this paper is to find more precise method for directly converting sound
energy vibrations to electrical energy using the principle of piezoelectricity. As presented
above the existing approach first converts sound energy to heat energy and then to electricity
but since there are losses occurring during conversion the best method is to directly
convert sound vibrations to electrical energy. The Basic idea of this method is to make use of
a panel consisting of piezoelectric crystals connected in series.
CONVERSION OF SOUND ENERGY TO ELECTRICAL… 119

The various sources of sound vibrations that can be applied to the crystal panel are as shown in the
table 1.
Table 1

Sound sources (noise) Sound pressure

Examples with distance Level Lp dB SPL

Jet aircraft, 50 m away 140

Threshold of pain 130

Threshold of discomfort 120

Chainsaw, 1 m distance 110

Disco, 1 m from speaker 100

Diesel truck, 10 m away 90

Curbside of busy road, 5 m 80

Vacuum cleaner, distance 1 m 70

Conversational speech, 1 m 60

Average home 50
120 APEKSHA LOKARE, POOJA PATIL AND SOUMYA ASANGIMATH

Sound sources (noise) Sound pressure

Examples with distance Level Lp dB SPL

Quiet library 40

Quiet bedroom at night 30

Background in TV studio 20

Rustling leaves in the distance 10

Hearing threshold 0

Table 2

Sound pressure p N/m² = Sound intensity I W/m²

Pa Sound field quantity Sound energy quantity

200 100

63.2 10

20 1

6.3 0.1

2 0.01

0.63 0.001

0.2 0.000 1

0.063 0.000 01

0.02 0.000 001

0.006 3 0.000 000 1

0.002 0.000 000 01

0.00063 0.000 000 001

0.000 2 0.000 000 000 1

0.000063 0.000 000 000 01

0.00002 0.000 000 000 001
CONVERSION OF SOUND ENERGY TO ELECTRICAL… 121

Table 2 represents the sound intensity as well as the pressure created by the vibrations of
these sound sources. From the table we can observe that greater the sound pressure higher
will be the sound intensity level. The sound vibrations of high decibel values have to be
applied to the crystal panel to produce voltage.
To begin the mechanical vibrations of sound is directly applied to the piezoelectric crystal
panel due to this the crystal undergoes stress as a result of this it gets compressed and thus
produces voltage of one polarity and when the crystal is released voltage of another polarity
is produced. The voltage thus produced is in the AC form so it is rectified or converted to
DC using the rectifier or AC to DC converter the output voltage is in DC but since it consists
of ripples it is then filtered using a capacitor, the output voltage obtained then has The basic
block diagram involved in this method of conversion of sound vibrations to electricity is as
shown below
The block diagram consists of a sound source that may be woofer, noise on the roads or any
other source then the piezoelectric crystal panel, rectifier or AC to DC converter, capacitor,
regulator and load may be any application or a storage battery.

To begin the mechanical vibrations of sound is directly applied to the piezoelectric crystal
panel due to this the crystal undergoes stress as a result of this it gets compressed and thus
produces voltage of one polarity and when the crystal is released voltage of another polarity
is produced. The voltage thus produced is in the AC form so it is rectified or converted to
DC using the rectifier or AC to DC converter the output voltage is in DC but since it consists
of ripples it is then filtered using a capacitor, the output voltage obtained then has
122 APEKSHA LOKARE, POOJA PATIL AND SOUMYA ASANGIMATH

fluctuations which is then maintained at a constant level by a regulator and then it is stored
in a battery or can be used to power some small electronic devices
The main attribute that differentiates this approach from the existing approaches is that, the
losses incurred by this method are low as compared to the losses occurring in the existing
approaches. Hence the output and the efficiency obtained from this approach is better than
the existing ones. This paper intends to generate electrical energy sufficient enough to
overcome the present problems of energy crisis. With sound harvesting technology batteries
could be charged from the human voice, music and highway traffic noise. The output
expected from this method will be sufficient enough to power small electronic and wireless
devices. The generated electricity can also be used for high voltage applications, where the
voltage is boosted up.

4. MERITS
 By converting sound energy to electrical energy efficiently the scarcity of electricity is
reduced globally and also helps in reducing the CO2 level in the atmosphere as it is
one of the cleanest form of energy
 The noise pollution in the road would be converted into electric energy and light the
street lightning, signals and various other electrical appliances.
 The noise pollution in runway could be used to produce electricity.
 The noise pollution in industries could be used to produce electricity and work certain
low voltage machine its scope doesn’t end.

5. DEMERITS
• Lots of research is to be done in this field. Its efficiency is not that good so
Improvement is being required.
• This method is bit costlier as the initial cost of piezoelectric crystals is high and large
number of crystals are required.
• It cannot be used in the places where decibel of sound is very low.

6. APPLICATIONS
 With sound harvesting technology batteries could be charged from human voice,
music concerts etc.
CONVERSION OF SOUND ENERGY TO ELECTRICAL… 123

 Once successfully harvested, this energy can be used to eliminate batteries in wireless
sensors. Eg: Mobile phones could be charged during conversations.
 The noise pollution in the road can be converted into electric energy and light the
streets, signals and various other electrical appliances.

7. ITS SCOPE
The power that is stored in a battery is in DC form so it can be converted into AC using the
inversion circuit and can be used for various applications which include street lighting,
illuminating industries, etc.

CONCLUSION
Energy harvesting is an emerging technology combining both electrical and mechanical
fields. Out of the current popular methods of energy harvesting, sound energy everywhere in
the environment is the most promising for energy harvesting. From the overview, it is quite
clear that sound energy harvesting has great potential and some very important part of
applications is still in the research and development stage. As sound has enormous amount
of energy with it, it could be used by converting it into electrical energy for various
purposes. The possibility of overall dependence on ambient energy resources may remove
some constraints required by the limited reliability of standard batteries.

REFERENCES

[1] International Journal of Environmental Science and Development, Vol. 2, No. 6, December
2011.
[2] IEEE transactions on ultrasonics, ferroelectrics, and frequency control, vol. 51, no. 7, july 2004.
[3] BVICAM’s International Journal of Information Technology (BIJIT) Bharati Vidyapeeth’s
Institute of Computer Applications and Management (BVICAM), New Delhi
[4] International Journal of Emerging Technology and Advanced Engineering ISSN 2250-2459,
Volume 2, Issue 10, October 2012
[5] Pratibha Arun Vetal. Int. Journal of Engineering Research and Applications Vol. 3, Issue 5, Sep-
Oct 2013, pp.478-482
[6] ARPN Journal of Engineering and Applied Sciences VOL. 8, NO. 7, JULY 2013, ISSN 1819-
6608.
[7] Journal of the Korean Physical Society, Vol. 57, No. 4, October 2010, pp. 947_951.
124 APEKSHA LOKARE, POOJA PATIL AND SOUMYA ASANGIMATH

[8] Raju M., “Energy Harvesting, ULP meets energy harvesting: A gamechanging combination for
design engineers,” Texas Instrument White Paper, Nov. 2008.
[9] Vullers R.J.M., Leonov V., Sterken T., Schmitz A., “Energy Scavengers For Wireless Intelligent
Microsystems,” Special Report in Microsystems & Nanosystems, OnBoard Technology, June
2006.
[10] Imec, “Design for analog and RF technologies and systems,” Computational Intelligence
Applications to Renewable Energy-2012 63.
[11] Imec, “Micropower generation and storage,”.
[12] Whetten F., “Energy Harvesting Sensor Systems – A Proposed Application for 802.15.4f,”
DOC: IEEE802.15-09/0074-00-004f
[13] Cossio C., “Harvest energy using a piezoelectric buzzer,” EDN, pg.94-96, March 20, 2008.
[14] Singh U. K. and Middleton R. H., “Piezoelectric power scavenging of mechanical vibration
energy” Australian Mining Technology Conference, 2-4 Oct. 2007, pp. 111- 118.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 125-133

SINGLE PHASE POWER FACTOR CORRECTION

VINITA V. PARAB, V. N. SHET AND SONALI PANDIT

Department of Electrical and Electronics Engineering,
Goa College of Engineering
Ponda, Goa, India.

Abstract
Majority of the applications involving electronic circuitry require a regulated DC supply. Since AC
supplies are more commonly available, a suitable AC-DC converter becomes mandatory for such
applications. These AC-DC converters involve a number of non-linear devices which reduce the
system power factor and introduce harmonics in the power system leading to adverse effects. Hence it
is essential to use a suitable power factor correction technique to condition the supply current. The
objective of the paper has been in the direction of better understanding of AC/DC converter. Emphasis
of this paper has the design of 100W AC/DC converter with high input power factor and tight output
voltage regulation. Simulation studies have been carried out using MATLAB/SIMULINK by using
Average current mode control topology.

-----------------------------------
Keyword : PFC, Boost converter, Average current mode control, PI controller.
© http: //[Link]
126 VINITA V. PARAB, V. N. SHET AND SONALI PANDIT

1. INTRODUCTION
The rise in the industrial, commercial and residential applications of electronic equipment
has resulted in a huge variety of electronic devices requiring mains supply. Most electronic
equipment is supplied by 50-60 Hz utility power and more than 50% of this power is
processed through some kind of power converter. Usually Power converters use a diode
rectifier followed by a bulk capacitor to convert AC voltage to DC voltage. Since these
power converters absorb energy from the AC line only when the line voltage is higher than
the DC bus voltage, the input line current contains rich harmonics, which pollute the power
system and interfere with other electric equipment. These converters usually have a low
power factor. Since low power factor loads tend to draw higher currents for the same value
of power, an increase in electricity bills results and high harmonic distortion of the line
current causes EMI problems and cross interferences through the line impedance between
different systems connected to the same grid. These effects discussed above necessitate an
effective power factor correction method for improving the quality of power. [1]
This power factor correction can be implemented in two ways, namely active and passive
power factor Correction. Passive solutions can be used to achieve this objective for low
power applications. With a filter inductor connected in series with the input circuit, the
current conduction angle of the single-phase full-wave rectifier is increased leading to a
higher power factor of about 0.8 and lower input current distortion. With smaller values of
inductance, these achievements are degraded. However the large size and weight of these
elements, in addition to their inability to achieve unity power factor or lower current
distortion significantly, make passive power factor correction applicable for lower power
applications. Active PFC solutions are a more suitable option for achieving near unity power
factor and sinusoidal input current waveform with extremely low harmonic distortion. [2]

2. ACTIVE POWER FACTOR CORRECTION TOPOLOGIES

Many DC–DC converter circuits and topologies are used at the PFC applications.
Converters used for PFC applications are Boost converter, Buck converter , Buck-boost
converter, Flyback converter, Cuk converter etc. the most common topology used for power
factor correction, can operate in two modes continuous conduction mode and discontinuous
conduction mode. Different current control techniques are being used to achieve a high input
power factor they are discussed in section 3.
SINGLE PHASE POWER FACTOR CORRECTION 127

3. CURRENT CONTROL TECHNIQUES FOR PFC

Many PFCs based on the boost topology have been proposed in the literature. Various
control strategies have also been implemented. In the following, the most popular control
techniques are reviewed for PFC [3].
• Peak current control.
• Hysteresis control.
• Borderline control.
• Discontinuous PWM control.
• Average current control.
3.1 Average current mode control
These PFC applications generally have two control loops. The first loop of these is a voltage
control loop for regulated output voltage (outer loop). The second loop is a current control
loop for high PF value and active shaping the input current. This structure is named as
average current mode control. In basic form, boost converter operates as follows, switch
controls the boost converter circuit. When the switch is closed, current flows through the
inductance. After then, the switch is opened and the current is forced to flow through the
diode to the output. Multiple cycles of this switching cause the output capacitor voltage to
build due to the charge it stores from the inductor current. The result is a higher output
voltage than the source voltage. Generally, voltage control of boost converter can be fulfilled
via traditional control methods such as PI controller.
Fig 1 shows circuit diagram of PFC wherein the output voltage is sampled and compared
with the reference voltage. The error signal is input to the voltage controller. The difference
(or "error" signal) is then used to adjust some input to the process in order to bring the
measured process value back to its desired set point. Unlike simpler controllers, the PI can
adjust process outputs based on the history and rate of change of the error signal, which
gives more accurate and stable control. PI controllers do not require advanced mathematics
to design and can be easily adjusted (or "tuned") to the desired application, unlike more
complicated control algorithms based on optimal control theory. The PI output is fed to the
Multiplier. Other input to Multiplier is the input voltage (rectified/mains). Multiplier
generates the current reference. This reference current is then compared with the inductor
current. The error is input to the current controller (PI).The PI output then controls the
128 VINITA V. PARAB, V. N. SHET AND SONALI PANDIT

switching action of the MOSFET switch. Thus according to the fluctuations in output
voltage the width of the PWM pulses would be varied in order to achieve PFC. [4][5]

Figure 1 : PFC Implementation Based on average current method

3.2 Mathematical expressions for average current control method
This can be better explained with the following equations:
Because of low losses and filtering of the high level harmonics from inductance, input and
output powers can be accepted equal (Pi =PO) and following equations are written

(1)

(2)

Where is obtained from equation 1 and substituted in equation 2, hence inductor current
reference is obtained as follows

(3)

Inductor reference current of the boost converter depends on output and input voltages of
the circuit. For the PFC, firstly a reference is calculated for the source current at control
SINGLE PHASE POWER FACTOR CORRECTION 129

operation. For this, output voltage controller is compared with a sample of source voltage.
Hence we get the following expression:

(4)
After then output of the voltage controller providing the amplitude of the reference current
is multiplied with the sample of the full wave rectified voltage signal and written as
(5)
Obtained current reference is compared with the actual value of the current and founded
error from this comparison is used as input to the current controller and duty cycle is
founded as in equation 6.

(6)

At last the output of the current controller is compared with a carrier signal and the gate
signal of the switch is found which is given to the MOSFET switch.

4. SIMULATION MODEL
This section will focus on implementation of boost converter in open loop for PFC and
average current mode control method in MATLAB/SIMULINK. The Parameters of the
boost converter were calculated [6][7] and calculated values are tabulated in table 1 as
shown below.

Table 1 : AC-DC boost converter design Parameters for PFC

Output power 100W

Input voltage 230 +/- 30% Vac

Line frequency 50Hz

L 25mH

C 200uF

Output voltage 400Vdc

Switching frequency 20KHz

130 VINITA V. PARAB, V. N. SHET AND SONALI PANDIT

4.1. MATLAB/ SIMULINK of Boost Converter in Open Loop

Figure 2 : Schematic of Boost PFC using MATLAB

(a) Output voltage from simulated circuit

(b) Input voltage in phase with input current

Figure 3 : MATLAB/SIMULINK results for Boost Converter in Open loop.
SINGLE PHASE POWER FACTOR CORRECTION 131

4.2. MATLAB/SIMULINK of Boost Converter with Average Current Control method

The control algorithm of the average current control method have been developed and
implemented on the MATLAB/SIMULINK. The PFC Boost converter is designed with the
specification as mention in Table [Link] pulses for the switch in Boost converter are
generated using the above mention method. Average current control method was simulated
by PI controller as shown in fig 4. response of this method was shown in the figure 5 with
smaller harmonics content and input current in phase with input voltage .The power factor
with average current control method was found around [Link] THD which signifies the
amount of distortion present in the input current signal is found to be the least for this control
method it was found around 4.19%.

Figure 4 : Schematic of Boost Converter with average current control method

(a)Output current from simulation

132 VINITA V. PARAB, V. N. SHET AND SONALI PANDIT

(b)Output voltage from simulation

(c) Line current harmonic spectra THD=4.19%.

Figure 5 : MATLAB/SIMULINK results for PFC with Average current control method
SINGLE PHASE POWER FACTOR CORRECTION 133

CONCLUSION
In this Paper an Average current mode control method for boost PFC converter was
simulated using MATLAB/SIMULINK. The results obtained were analysed, from the
simulated result it is observed that output voltage is constant, input voltage and input current
is in phase and thus 0.99 Power factor was achieved and THD of the input current is 5%.
Thus to meet the limits and come up with growing AC/DC power supply market, The PFC
stage is currently required. The power factor correction is applicable for Industrial variable
speed drives, High Efficiency Servers, High efficiency Telecom AC-DC Converter, home
appliances-Washing machines, Refrigerators, Compressors, Fans, etc where AC induction
motors, Permanent magnet synchronous motors or Brushless DC motors are used.

REFERENCES

[1] Daniel ALBU, Nicolae DRAGHICIU, Gabriela TONT and Dan George TON
“Converters with Power Factor Correction “ACTA Electrotehnica, Volume 48, Number
1, 2007
[2] “Power factor correction handbook” HBD8531D, Rev. 3,sep.2007
[3] Rossetto L., Spiazzi G., Tenti P. “control technique for power factor correction
converters” University of Padova, ITALY
[4] Kurkute Sanjay L, Patil Pradeep M and Mohite K.C. “A Digital Power Factor
Correction using Floating Point Processor for Pulse Width Modulation Control in Boost
Converters” International Journal of Electronic Engineering Research ,ISSN 0975-
6450 Volume 1 Number 2 (2009).
[5] García Oscar, Member, Cobos José A., Prieto Roberto, Member, Alou Pedro, and
Uceda Javier, “Single Phase Power Factor Correction: A Survey” IEEE Transactions
On Power Electronics, Vol. 18, No. 3, May 2003.
[6] Todd Philip “UC3854 Controlled Power Factor Correction Circuit Design” unitrode
application note.
[7] Reddy Meghana R., Bhaskar B.N., Mohan H.R. “Single phase active PFC using boost
converter” International Journal of Advanced Computational Engineering and
Networking, ISSN: 2320-2106
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 135-145

HARNESSING ELECTRICAL ENERGY FROM SOUND

USING PIEZOELECTRIC FILM

NIKITA ARORA, POONAM MALI AND SOUMYA ASANGIMATH

S.G Balekundri Institute of Technology,
Belgaum-590010, India

Abstract
Energy crisis is the talk of the world today. There have been many inventions and discoveries but the
future of those innovations has been jeopardized due to the non-availability of resources. There are
many energy resources which have not been explored yet. One such viable option which has been
underused is the sound energy. Sound is a freely and vastly available renewable source of energy. As
sound is mechanical wave comprising of vibrations, sound energy can be converted into electrical
energy by making use of the piezoelectric film/sheet. This paper proposes to harness electricity from
sound vibrations and finally making power unavailability a thing of the past.

----------------------------------
Keywords : Energy crisis, sound, renewable source, vibrations, piezoelectric film/sheet.
© http: //[Link]
136 NIKITA ARORA, POONAM MALI AND SOUMYA ASANGIMATH

1. INTRODUCTION
In the recent years electricity has become a basic necessity for mankind. Due to the rapid
increase in load, the demand is not met by the supply which has led to energy crisis all
around the world. To bridge the gap between demand and supply researchers all over the
globe are working towards a common goal of finding new and innovative methods for
electricity production. One such unique attempt is generation of electricity from sound
energy by making use of a piezoelectric film/sheet.
Sound is a vibration that propagates as a mechanical wave of pressure and displacement
through some medium. The propagation of sound is generally affected by three things
 A relationship between pressure and density. This relationship, affected by temperature,
determines speed of sound within a medium.
 The propagation is also affected by the motion of the medium itself. For example, sound
moving through wind. Independent of the motion of sound through the medium, if the
medium is moving, the sound is further transported.
 The viscosity of the medium also affects the motion of sound waves. It determines the
rate at which sound is attenuated. For many media such as air, water etc attenuation due
to viscosity is neglected.
The mechanical vibrations that can be interpreted as sound are able to travel through all
forms of matter: gases, liquids, solids and plasmas. The matter that supports sound is called
the medium. Sound cannot travel through vacuum.
Sound waves are often simplified to a description in terms of sinusoidal plane waves, which
are characterized by the following properties:
 Frequency
 Wavelength
 Amplitude
 Wave-number
 Sound pressure
 Sound intensity
 Speed of sound
 Direction
137

Sound that is perceptible by humans has frequencies from 20hz to 20khz. In air at standard
temperatures and pressure, the corresponding wavelengths of sound waves range from 17m
to17mm.
Piezoelectricity is the electric charge that accumulates in certain solid materials (such as
crystals, sensors, films etc) in response to applied mechanical stress. The word
piezoelectricity means electricity resulting from pressure. It is derived from the Greek word
piezo or piezein which means to squeeze or press, and electric or electron which stands for
amber, an ancient source of electric charge. Piezoelectricity was discovered in 1880 by
French physicists Jacques and Pierre Curie.
The piezoelectric effect is understood as the linear electromechanical interaction between the
mechanical and electrical state in crystalline materials with no inversion symmetry. The
piezoelectric effect is the reversible process. The materials which exhibit the direct
piezoelectric effect (the internal generation of electric charge resulting from an applied
mechanical force) also exhibit the reverse/converse piezoelectric effect (the internal
generation mechanical strain resulting from an applied electric field).

The first demonstration of piezoelectric effect was in 1880 by brothers Pierre curie and
Jacques curie. They combined the knowledge of pyroelectricity with their understanding of
the underlying crystal structures that gave rise piezoelectricity. Quartz and Rochelle salt
exhibited the most piezoelectricity.
Piezoelectricity is the combined effect of the electrical behavior of the material.
𝐷 = 𝜀𝐸
Where 𝐷 is the electric charge density displacement, 𝜀 is the permitivity and 𝐸 is the electric
field.
138 NIKITA ARORA, POONAM MALI AND SOUMYA ASANGIMATH

Some of the naturally occurring piezoelectric materials are quartz, berlinite, sucrose,
Rochelle salt, topaz etc. the synthetic piezoelectric materials available are barium titanate,
lead titanate, lead zirconate titanate, zinc oxide etc.

2. LITERATURE REVIEW
There are a number of applications in which the piezoelectricity principle is used for
generation of electricity. Some of them are as mentioned below.
2.1 Footsteps Power Generation
The piezoelectric crystal arrays are laid underneath the pavements, foot paths and other high
traffic areas like highways, speed breakers for maximum voltage generation. The voltage
thus generated from the array can be used to charge the chargeable Lithium batteries,
capacitors etc. These batteries can be used as per the requirement.
2.2 Power Generating Shoes
In United States Defense Advance Research Project Agency (DARPA) initiated an
innovative project on Energy harvesting which attempts to power battlefield equipment by
piezoelectric generators embedded in soldiers' boots. However, these energy harvesting
sources put an impact on the body. DARPA's effort to harness 1-2 watts from continuous
shoe impact while walking were abandoned due to the discomfort from the additional energy
expended by a person wearing the shoes.
2.3 Generation from Vibrations
Researchers are also working on the idea of utilizing the vibrations caused from the
machines in the gym. At workplaces, while sitting on the chair, energy can be stored in the
batteries by laying piezoelectric crystals in the chair. Also, the studies are being carried out
to utilize the vibrations in a vehicle, like at clutches, gears, seats, shock-ups, foot rests.
2.4 Mobile Keypad and Keyboards
The piezoelectric crystals can be laid down under the keys of a mobile unit and keyboards.
With the press of every key, the vibrations created are used by piezoelectric crystal and
hence can be used for charging purpose.
139

2.5 Floor Mats, Tiles and Carpets

A series of crystals can be laid below the floor mats, tiles and carpets which are frequently
used at public places.
2.6 People Powered Dance Clubs
In Europe, certain nightclubs have already begun to power their night clubs, strobes and
stereos by use of piezoelectric crystals. The crystals are laid underneath the dance floor.
When a bulk of people use this dance floor, enormous amount of voltage is generated which
can be used to power the equipments of the night club. A famous night club in London
exploited the principle of piezoelectricity in making its dance floor. The dance floor uses
piezoelectricity where crystal and ceramics create a charge to generate electricity. The
nightclub has a "bouncing" floor made of springs and a series of power generating blocks
which produce a small electrical current when squashed. As dancers move the floor up and
down to squeeze the blocks, the current is fed into nearby batteries which are constantly
recharged by the movement of the floor. The electricity created in this way is used to power
parts of the nightclub such as the sound and lighting
140 NIKITA ARORA, POONAM MALI AND SOUMYA ASANGIMATH

3. EXISTING APPROACH
In this method we could convert sound energy to heat energy as sound wave travel by
oscillating the particles of the medium so when sound energy travel through the medium it
will disturbs the particle of the medium these disturbance created by sound will be used to
convert it into heat energy as when the particles of the medium will be pushed by the sound
wave it will collides with adjacent particle of the medium this collision will result in
production of heat energy the production of heat energy will be more in the denser medium
so for more heat production we will need a material with very high density. This heat energy
will be into electrical energy

This method is less efficient due to more energy losses taking place while conversion of
sound energy to heat energy and then heat energy to electric energy then the other methods.
141

As here the conversion is done two times .you may think that according to Newtown’s law
energy can’t be created nor destroyed it could be only converted from one form to
another…so how there could be an energy loss. So while converting sound energy to heat
energy there will be some loss of sound energy as some of the sound energy would be
converted into another form also while converting from heat energy to electric energy all
energy wouldn’t be converted into electric energy some of the heat energy would get
converted into another form of energy. And here as our main focus is to convert sound
energy to electric energy so conversion into other form of energy is loss of energy for us.
By making use of a new approach using a piezoelectric film the sound energy could be
directly converted into electrical energy thus reducing the losses and increasing the
efficiency.

4. PIEZOELECTRIC FILM
The PVDF piezoelectric material is a relative new class of piezoelectric sensors. The PVDF
material is a thin plastic polymer sheet that has a thin electrically conductive nickel-copper
alloy deposited on each side. This material is called a piezo film. The material resembles
aluminum foil; with the exception that it is more resilient (doesn’t crinkle as easily).
142 NIKITA ARORA, POONAM MALI AND SOUMYA ASANGIMATH

To use piezoelectric film one makes an electrical connection to the electrically conductive
coating on each side of the PVDF material. Most piezo film have wire connections. If one
needs to attach electrical connections to raw piezo film this is accomplished using either
copper tape or a conductive epoxy. Soldering directly to the film is not possible because the
heat would destroy the underlying plastic PVDF material.

PVDF is also pyroelectric, meaning that it produces an electrical charge in response to a

change in temperature. The material is sensitive to infrared energy between 7-20 micrometer.
This makes it an excellent sensor for heat detection of the human body. Film sensors are
commonly employed n motion sensors, where with the use of a Fresnel lens, can detect
human body within fifty feet.

5. PIEZO-FILM PROPERTIES
Piezo-film has an extremely high output voltage, about 10 times larger than ceramic
materials. Since the material is thin light weight flexible film, it can be glued into shaped
designs. The material has high mechanical strength and it’s impact resistant. Other attributes
include
 Wide frequency range (0.001hz-10^9hz)
143

 Low acoustical impedance (close to water and human tissue, efficient for sonar
imaging)
 High dielectric strength
 Good mechanical strength
 Moisture resistant, inert to many chemicals
One major disadvantage of the piezo-film is that it is a weak electromechanical transmitter
when compared to piezoelectric ceramics.

6. SOUND HARVESTING TECHNOLOGY

The block diagram representation for harnessing of electrical energy from sound energy is as
shown in the figure below.

A sound wave comprises of vibrations. When the vibrations of sound are applied to a piezo-
film, the piezo-film undergoes stress. As a result of the applied pressure the film generates a
small voltage in response to the applied sound vibration.
The voltage generated is in AC form. To store the voltage it is first converted into DC by
making use of a simple bridge rectifier circuit. The rectified output consists of ripples which
can be filtered out from the voltage by making use of a capacitor of appropriate range. Once
this is done the ripple free voltage can be stored in the battery or can be utilized instantly for
small electronic devices.
The stored voltage can also be converted into AC by making use of an inversion circuit,
boosted up and then be utilized for AC powered appliances.
144 NIKITA ARORA, POONAM MALI AND SOUMYA ASANGIMATH

7. MERITS
The various merits of this technology are as follows:
 It is the cleanest form of energy as it does not harm the environment in any way.
 The noise pollution in the surroundings could be used for something useful.
 This technology can reduce the use of polluting non renewable sources of energy and
still meet the load demands as sound is freely and vastly available.

8. DEMERITS
As every innovation even this technology has a few demerits. A few are mentioned below.
 As the piezoelectric film is expensive the overall cost of this technology is very high.
 There is a lot research needed in this field as it is still a new technology.
 The piezo-film is sensitive to sounds having high decibel ranges hence this technology
could not be applicable in all situation

9. APPLICATIONS
Generation of electricity from sound has unlimited applications. Some of them are
mentioned below.
 Electricity could be generated from the noise produced in the factories and could be
used for the lighting of industries.
 Noise pollution due to heavy traffic could be utilized to power street lights.
 Music in concerts and movie theaters could be utilized to power various electronic
appliances in theaters and auditoriums
 The takeoff sound of jets could help contribute to the large electricity demand at
airports.

10. FUTURE SCOPE

 This technology could completely eradicate the use of batteries in small electronic
devices.
 It could allow a person to charge his/her phone while talking.
 This technology could be used defense fields by soldiers who are in locations where it is
impossible for them to charge their communication devices such as walkie-talkie.
145

CONCLUSION
Since the source for electricity generation in sound harvesting technology is vastly available,
renewable and free of cost, this technology could help us solve all our energy crisis problems
and help produce sufficient electricity in an eco-friendly way.

REFERENCES

[1] International Journal of Environmental Science and Development, Vol. 2, No. 6, December
2011.
[2] IEEE transactions on ultrasonic’s, ferroelectrics, and frequency control, vol. 51, no. 7, July
2004.
[3] BVICAM’s International Journal of Information Technology (BIJIT) Bharati Vidyapeeth’s
Institute of Computer Applications and Management (BVICAM), New Delhi
[4] International Journal of Emerging Technology and Advanced Engineering ISSN 2250-2459,
Volume 2, Issue 10, October 2012
[5] Vetal Pratibha Arun. Int. Journal of Engineering Research and Applications Vol. 3, Issue 5, Sep-
Oct 2013, pp.478-482
[6] ARPN Journal of Engineering and Applied Sciences VOL. 8, NO. 7, JULY 2013, ISSN 1819-
6608.
[7] Journal of the Korean Physical Society, Vol. 57, No. 4, October 2010, pp. 947_951.
[8] Raju M., “Energy Harvesting, ULP meets energy harvesting: A gamechanging combination for
design engineers,” Texas Instrument White Paper, Nov. 2008.
[9] Vullers R.J.M., Leonov V., Sterken T., Schmitz A., “Energy Scavengers for Wireless Intelligent
Microsystems,” Special Report in Microsystems & Nanosystems, OnBoard Technology, June
2006.
[10] Imec, “Design for analog and RF technologies and systems,” Computational Intelligence
Applications to Renewable Energy-2012 63.
[11] Imec, “Micro power generation and storage,”.
[12] Whetten F., “Energy Harvesting Sensor Systems – A Proposed Application for 802.15.4f, ”
DOC: IEEE802.15-09/0074-00-004f
[13] Cossio C., “Harvest energy using a piezoelectric buzzer,” EDN, pg.94-96, March 20, 2008.
[14] Singh U. K. and Middleton R. H., “Piezoelectric power scavenging of mechanical vibration
energy” Australian Mining Technology Conference, 2-4 Oct. 2007, pp. 111- 118.

Number 2 - December 2012

International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 147-159

IMPROVEMENT OF POWER QUALITY BY REDUCING

HARMONICS PRODUCED BY NON-LINEAR LOADS

DEEPAK S1 AND B. V. SUMANGALA2

1
[Link] (Power Electronics) Dept. EEE, 2 Head, Dept. EEE,
Dr. Ambedkar Institute of Technology, Bengaluru, India

Abstract
The main aim of this project is to explain the effects of Harmonics in the Power System and steps to
reduce the effects of Harmonics. This project will also explain how Harmonic distortion is one of the
most important problems associated with power quality and creates several disturbances to the Power
System. It includes the Harmonic reduction techniques to improve the power quality for that purpose
Active power filters are used to improve power quality by compensating harmonics and reactive
power required by a non linear load. A Series active and passive shunt filter is simulated on matlab
simulink platform and the control strategy based on instantaneous reactive power theory so that the
voltage waveform injected by the active filter is able to compensate the reactive power.

----------------------------------
Keywords : Active power filters, harmonics, hybrid filters, instantaneous reactive power, power
quality. © http: //[Link]
148 DEEPAK S AND B. V. SUMANGALA

1. INTRODUCTION
Harmonics are the by-products of modern electronics. They occur frequently when there are
large numbers of personal computers (single phase loads), uninterruptible power supplies
(UPSs), variable frequency drives (AC and DC) or any electronic device using solid state
power switching supplies to convert incoming AC to DC. Non-linear loads create harmonics
by drawing current in abrupt short pulses, rather than in a smooth sinusoidal manner the
terms “linear” and “non-linear” define the relationship of current to the voltage waveform..
A solution to power quality problems are offered by active power filters. There are also three
types of active power filter such as series active power filter, shunt active power filter&
hybrid active power filter. In that the cost of shunt active power filter is relatively high and
they are not preferable for a large scale system. The series active power filters works as a
kind of harmonic isolator rather than a harmonic voltage generator. Series active power
filters are used to eliminate voltage harmonics, to balance and regulate the terminal voltage
of the load or line. Also, it has been used to reduce negative-sequence voltage.
A new control algorithm is hybrid filter, which is a combination of an active series filter and
passive shunt filter is quite popular because of the solid-state devices are used in the active
series part can be of reduced size and cost. Also the major part of the hybrid filter is made of
the passive shunt LC filters are used to eliminate lower order harmonics. It has the capability
of reducing voltage and current harmonics at a reasonable cost, and allows the use of active
power applications at a relatively low cost. Moreover, compensation characteristics of
passive filters can be significantly improved by connecting a series active power filter at its
terminals, giving more flexibility to the compensation scheme and avoid possibility of the
generation of series and parallel resonance.

2. HARMONICS
Harmonics have been around a long time. Since the first AC generator went online more than
100 years ago, electrical systems have experienced harmonics. The harmonics at that time
were minor and had no detrimental effects.
When a sinusoidal voltage is applied to a certain type of load, the current drawn by the load is
proportional to the voltage and impedance and follows the envelope of the voltage waveform.
These loads are referred to as linear loads (loads where the voltage and current follow one
IMPROVEMENT OF POWER QUALITY BY REDUCING… 149

another without any distortion to their pure sine waves). Examples of linear loads are resistive
heaters, incandescent lamps, and constant speed induction and synchronous motors.
In contrast, some loads cause the current to vary disproportionately with the voltage during
each half cycle. These loads are classified as nonlinear loads, and the current and voltage have
waveforms that are non-sinusoidal, containing distortions, whereby the 50-Hz waveform has
numerous additional waveforms superimposed upon it, creating multiple frequencies within
the normal 50-Hz sine wave. The multiple frequencies are harmonics of the fundamental
frequency.
In our industry other large contributors are:
 Variable Speed Drives (VSD)
 Battery Chargers
 Computer Power Units (CPU)
 Discharge Lighting (fluorescent, mercury, sodium, etc.)
 Personal Computers (PC)
 Rectifiers
 Uninterrupted Power Supplies (UPS)
The actual problems of any system will vary, depending on the types and number of installed
harmonic producing loads. Most system can withstand nonlinear loads of up to 15% of the
total electrical system capacity without concern, but, when the nonlinear loads exceed 15%
some non-apparent negative consequences can be expected. For systems that have nonlinear
loading of more than 25%, particular problems can be become apparent. The following is a
short summary of most, but not all of the problems caused by harmonics:
 Capacitor Failure - Harmonic Resonance
 Circuit Breakers Tripping - Inductive Heating and Overload
 Computer Malfunction or Lockup - Voltage Distortion
 Conductor Failure - Inductive Heating
 Electronic Equipment Shutting down - Voltage Distortion
 Flickering of Fluorescent Lights - Transformer Ballast Saturation
 Fuses Blowing for No Apparent Reason - Inductive Heating and Overload
 Motor Failures (overheating) - Voltage Drop
 Neutral Conductor and Terminal Failures - Additive Currents
 Overheating of Metal Enclosures - Inductive Heating
150 DEEPAK S AND B. V. SUMANGALA

 Power Interference on Voice Communication - Harmonic Noise

 Transformer Failures - Inductive Heating
The heating effects of harmonic currents can cause destruction of equipment, conductors, and
fires. The results can be unpredictable, legal and financial ramifications. Voltage distortions
can lead to overheating of equipment, electronic equipment failure, expensive downtime, and
maintenance difficulties. Harmonic currents and voltage distortion are becoming the most
severe and complex electrical challenge for the electrical industry. The problems associated
with nonlinear loads were once limited to isolated devices and computer rooms, but now the
problem can appear throughout the power and utility system. [5]

3. FILTERS
Harmonic filters are used to eliminate the harmonic distortion caused by nonlinear loads.
Specifically, harmonic filters are designed to attenuate or in some filters eliminate the
potentially dangerous effects of harmonic currents active within the power distribution
system. Filters can be designed to trap these currents and, through the use of a series of
capacitors, coils, and resistors, shunt them to ground. A filter may contain several of these
elements, each designed to compensate a particular frequency or an array of frequencies.
3.1 Types of harmonic filters involved in harmonic compensation
Filters are often the most common solution that is used to reduce harmonics. Unlike other
solutions, filters offer a simpler inexpensive alternative with high benefits. There are three
different types of filters each offering their own unique solution to reduce and eliminate
harmonics. These harmonic filters are broadly classified into
i. Passive
ii. Active
iii. Hybrid structures.
The choice of filter used is dependent upon the nature of the problem and the economic cost
associated with implementation.
A passive filter is composed of only passive elements such as inductors, capacitors and
resistors thus not requiring any operational amplifiers. Passive filters are inexpensive
compared with most other mitigating its structure may be either of the series or parallel type.
Active filters use active components such as IGBT transistors to inject negative harmonics
into the network effectively replacing a portion of the distorted current wave coming from the
IMPROVEMENT OF POWER QUALITY BY REDUCING… 151

load. This is achieved by producing harmonic components of equal amplitude but opposite
phase shift, which cancel the harmonic components of the non-linear loads. Hybrid filters
combine an active filter and a passive filter. Its structure may be either of the series or parallel
type. The passive filter carries out basic filtering (5th order, for example) and the active filter,
through precise control, covers higher harmonics.[4]
3.2 Series active power filter and parallel passive filter
Series active filter and parallel passive filter topology shown in fig. 1, An active power filter
is connected in series to the ac source impedance it is possible to improve the compensation
characteristics of the passive filters in parallel connection. In order to allow current harmonic
compensation, a parallel LC filter must be connected between the nonlinear loads and the
series transformers. It is well known that series active power filters compensate current
system distortion caused by non-linear loads by imposing a high impedance path to the
current harmonics which forces the high frequency currents to flow through the LC passive
filter connected in parallel to the load.
The high impedance imposed by the series active power filter is created by generating a
voltage of the same frequency that the current harmonic component that needs to be reduced.
Current harmonic and voltage unbalance compensation are achieved by generating the
appropriate voltage waveforms with the three phase voltage-source inverter. Active power
filter is used with passive filters to improve the compensation characteristic of passive filter,
also it avoid the possibility of generation of series or parallel resonance The active filter
generates a voltage which compensates the passive filter and load reactive power, so it allows
the current harmonics to be reduced. The calculation algorithm is based on the instantaneous
reactive power theory the control target is to achieve constant power in the source side. [1]

4. INSTANTANEOUS POWER THEORY

Proposed theory based on instantaneous values in three-phase power systems with or without
neutral wire, and is valid for steady-state or transitory operations, as well as for generic
voltage and current waveforms called as Instantaneous Power Theory or Active- Reactive (p-
q) theory which consists of an algebraic transformation (Clarke transformation) of the three-
phase voltages in the a-b-c coordinates to the α-β-0 coordinates, followed by the calculation
of the p-q theory instantaneous power components.
In a three-phase system voltage and current vectors can be defined by
152 DEEPAK S AND B. V. SUMANGALA

𝑣 = [𝑣𝑎 𝑣𝑏 𝑣𝑐 ]𝑇 (1)

𝑖 = [𝑖𝑎 𝑖𝑏 𝑖𝑐 ]𝑇 (2)
The vector transformations from the phase reference system a-b-c to α- β-0 coordinates can
be obtained,
1 1 1
𝑉0 √2 √2 √2 𝑉𝑎
2 1 1
[𝑉𝛼 ] = √3 1 −2 −2 [𝑉𝑏 ] (3)
𝑉𝛽 √3 √3 𝑉𝑐
[0 2
− ]
2

The instantaneous real power in the a-b-c to α- β-0 frame is calculated as follows:
p3ϕ (t) = Vα iα + Vβ iβ + V0 i0 (4)
This power can be written as
p3ϕ (t) = p + p0 (5)
Where p is the instantaneous real power without zero sequence component and given by
p = Vα iα + Vβ iβ (6)
It can be written in vectorial form by means of dot product
p = Vαβ (7)
Where
𝑖𝛼𝛽 = [𝑖𝛼 𝑖𝛽 ]𝑇 (8)
𝑉𝛼𝛽 = [𝑉𝛼 𝑉𝛽 ]𝑇 (9)
𝑇
𝑝0 Is the zero sequence instantaneous power, 𝑖𝛼𝛽 calculated as follows In a three-wire
system there are no zero sequence current components,
𝑝0 = 𝑣0 𝑖0 (10)
The imaginary instantaneous power is defined by the equation
q = Vα iβ − Vβ iα (11)
This can be expressed by means of the dot product
𝑇
q = Vαβ 𝑖𝛼𝛽⊥ (12)
Where
𝑖𝛼𝛽⊥ = [𝑖𝛽 −𝑖𝛼 ]𝑇 (13)
Both power variables previously defined can be expressed as
IMPROVEMENT OF POWER QUALITY BY REDUCING… 153

𝑇
𝑝 𝑖𝛼𝛽
[𝑞 ] = [ 𝑇 ] 𝑉𝛼𝛽 (14)
𝑖 𝛼𝛽⊥

By means of the current vectors and the real and imaginary instantaneous power, the voltage
vector can be calculated as
𝑝 𝑞
𝑉𝛼𝛽 = 𝑖2 𝑖𝛼𝛽 + 𝑖2 𝑖𝛼𝛽⊥ (15)
𝛼𝛽 𝛼𝛽

Electric companies try to generate electrical power as sinusoidal and balanced voltages so it
has been obtained as a reference condition in the supply. Due to this fact, the compensation
target is based on an ideal reference load which must be resistive, balanced and linear
V= Re I (16)
Re is the equivalent resistance V is the load voltage vector and i the load current vector.
The average power supplied by the source will be
𝑝𝑠 = I12 R e (17)
Where I12 is the square rms value of the fundamental harmonics of the source current vector
when voltage is sinusoidal and balanced, only the current fundamental component transports
the power consumed by the load.
Compensator instantaneous power is the difference between the total real instantaneous
power required by the load and the instantaneous power supplied by the source
pc (t) = pL (t) − ps (t) (18)
In this equation, the average power exchanged by the compensator has to be null, that is
1
pc (t) = T ∫ pL (t) dt = 0 (19)

When average values are calculated in (18), and (17) and (19) are taken into account
1
0 = T ∫ pL (t) dt − I12 R e (20)

Therefore, the equivalent resistance can be calculated as

pL
Re = I21
(21)

Where pL is the load average power, defined as

1
pL (t) = T ∫ pL (t) dt (22)
𝑝𝐿
𝑉𝑝𝑐𝑐𝛼𝛽 = 𝑖
𝐼12 𝛼𝛽
(23)

𝑖𝛼𝛽 Is the source current in 0-α-β coordinates In this equation, the restriction of null average
power exchanged by the active filter is imposed.
154 DEEPAK S AND B. V. SUMANGALA

The load voltage is given according to (15) by

𝑝 𝑞
𝑉𝐿𝛼𝛽 = 𝑖2𝐿 𝑖𝛼𝛽 + 𝑖2𝐿 𝑖𝛼𝛽⊥ (24)
𝛼𝛽 𝛼𝛽

Where 𝑝𝐿 is the real instantaneous power and 𝑞𝐿 is the load imaginary instantaneous power
The reference signal for the output voltage of the active filter is
∗
𝑉𝑐𝛼𝛽 = 𝑉𝑝𝑐𝑐𝛼𝛽 − 𝑉𝐿𝛼𝛽 (25)
Considering (23) and (24)The compensation voltage is
∗ 𝑝𝐿 𝑝𝐿 𝑞𝐿
𝑉𝑐𝛼𝛽 =( − 2 ) 𝑖𝛼𝛽 − 2 𝑖𝛼𝛽⊥ (26)
𝐼12 𝑖𝛼𝛽 𝑖𝛼𝛽

Fig1 shows the system with series active filter, parallel passive filter and unbalanced and
non-sinusoidal load.

Figure 1 : Series active and shunt passive filter topolog

5. MATLAB SIMULATION and RESULTS

The system shown in Fig.1 has to be simulated in the Matlab-Simulink platform. Each power
device has been modeled using the SimPower System toolbox library. The system shown in
Fig. 2 has been simulated in the Matlab- Simulink platform to verify the proposed control.
Each power device has been modeled using the Sim PowerSystem toolbox library.
The power circuit is a three-phase system supplied by a sinusoidal balanced three-phase 100-
V source with a source inductance of 5.8 mH and a source resistance of 3.6Ω. The passive
filter is constituted by two LC branches tuned to the fifth and seventh harmonics. Each
element value is included in Table 1.
IMPROVEMENT OF POWER QUALITY BY REDUCING… 155

Table 1
Table Head Passive element values
Passive Filter for 5th Passive Filter for 7th
Source
harmonics harmonics

Resisance Rs=3.6Ω --- ---

Inductance Ls=5.8mH L5=13.5mH L7=6.75mH

Capacitance ---- C5=30µF C7=30µF

The nonlinear load consists of an uncontrolled three-phase rectifier with an inductance of 55

mH and a 25 resistor connected in series on the dc side. Fig. 4.2 shows the phase “a” load
current Total harmonics distortion (THD) is 18.6% and the power factor 0.947, when the
system is not compensated
The 5th and 7th harmonics are the most important in the current waveform. They are 16.3%
and 8.4% of the fundamental harmonic respectively Two LC branches were connected to
mitigate the fifth and seventh harmonics. The source current waveform with the passive filter
connected is shown in Fig,2 the THD falls to 4.7%.The 5th and 7th harmonics decrease to
3.6% and0.9%, respectively. The passive filter was designed only to compensate the source
current harmonics; the reactive power was not considered. The power factor of the set load
and passive filter is 0.97. To simulate the active filter series connection with a passive filter,
it is necessary to determine the APF reference signal. The control scheme used to calculate
the active filter compensation voltage as calculated in equation (26).

Figure 2 : Circuit realization in Matlab without filters

156 DEEPAK S AND B. V. SUMANGALA

Fig 3 Simulated waveform of Load current of phase A without filters the load current

Fig 4 FFT analysis of the waveform in fig3

Fig 5 Circuit realization in matlab consisting passive filters

IMPROVEMENT OF POWER QUALITY BY REDUCING… 157

Fig 6 Simulated waveform of Source current of phase A with passive filters

Fig 7 FFT analysis of the waveform in fig6

Fig 8source current when source impedance is varied

158 DEEPAK S AND B. V. SUMANGALA

Fig 9 source current when resistance in DC side is varied

Fig 10 Expected source current waveform of phase A, when active power filter is connected [2]

CONCLUSION
Harmonics plays an major role in determining power quality it can be reduced by using
filters of different topology, which improves total harmonic distortion ,power factor of
power system,hence improves power quality,even though passive filter was successful in
reducing THD to ieee standards for a set load and source impedance, but it’s operation is
limited by the variations of load and source impedance, thus it requires other equipment to
suceesfully reduce the THD of the system within the IEEE standards(<5%)
IMPROVEMENT OF POWER QUALITY BY REDUCING… 159

REFERENCES

[1] Metkari Archana Subhash, Prof. Pawar S.H., IOSR Journal of Electronics and Communication
Engineering (IOSR-JECE) ISSN : 2278-2834, ISBN : 2278-8735, PP : 37-40
[Link] Second.
[2] Salmerón P. and Litrán S. P., Improvement of the Electric Power Quality Using Series Active
and Shunt Passive Filters IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 25, NO. 2,
APRIL 2010.
[3] Power System Harmonics Causes and Effects of Variable Frequency Drives Relative to the
IEEE 519-1992 Standard Bulletin No. 8803PD9402 August, 1994 Raleigh, NC, U.S.A.
[4] Nalini Kiran C., Dash Subhransu Sekhar, Latha [Link], A Few Aspects of Power Quality
ImprovementUsing Shunt Active Power Filter International Journal of Scientific & Engineering
Research Volume 2, Issue 5, May-2011
[5] Arrillaga J., Watson N. R.: Power System Harmonics. New York: John Wiley, c2003
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 161-174

AC-AC CONVERTER USING SWITCHED

CAPACITOR PRINCIPLE

SAI DARSHAN S, RAMESHA S S AND SACHIN ANGADI

Department of Electrical & Electronics Engineering,
B. V. B College of Engineering & Technology,
HUBLI- 580031, Karnataka, India.

Abstract
This paper proposes a new ac static powerconverter by using switches and capacitors, intended to
replace the conventional autotransformer in commercial and residential applications. The operating
principle, the converter analyses are described in this paper. The main advantages of this proposed ac-
ac converter is the absence of magnetic elements, the stress voltages in all components being equal to
half of the high-side voltage, the common reference between input and output voltages, , the ability to
be bidirectional, the high efficiency, and the high power density. In order to demonstrate the
performance of this converter, the configuration is simulated by using saber simulation software. The
relevant simulated results are reported herein.

-----------------------------------
Keywords : AC–AC converter, bidirectional, switchedcapacitor (SC).
© http: //[Link]
162 SAI DARSHAN S, RAMESHA S S AND SACHIN ANGADI

1. INTRODUCTION
Throughout the world, it is common to find commercial and domestic appliances that work
with different voltages from those available on the electrical grid and the most popular
solution is the employment of a low-power/ low-voltage autotransformer. However, the
autotransformer efficiency is poor and it produces audible noise and there is a utilization of
copper in the auto transformer for winding purpose which leads to losses.
This power converter composed only of capacitors and switches, without magnetic devices,
they can achieve significant size reduction in comparison with any other converters generally
which are used in switched-mode power supplies. Further, the behavior of these circuits can
be described by simple equivalent circuits and it is possible to fabricate this converter on a
single semiconductor IC chip.
A recent publication has published this power converter to ac–ac static conversion based on
the principle of capacitors and switches where a brief analysis and experimental results for a
step-down/step-up converter with rated power of 600W, high-side voltage of 220Vrms , low-
side voltage of 110 Vrms , line frequency of 60 Hz, and switching frequency of 50 kHz were
presented. The important characteristicsof the converter are the stress voltages in all
components being equal to half of the high-side voltage, the differential output voltage, the
employment of two SC legs, and eight unidirectional switches. The resultsobtained in that
paper is to show that, the employment of this ac–ac converter can contribute with new and
efficient solutions for the ac–ac area, for example, in low power/low-voltage systems aimed
at replacing the conventional autotransformer in commercial and domestic applications.
The purpose of this paper is to introduce a new ac–ac static power converter topology based
on the principle of capacitor and switches. The important characteristics of this proposed
power converter are the absence of magnetic elements and the stress voltages on the switches
being equal to half of the high-side voltage. In addition to those characteristics, there are also
the new characteristics of having a common reference between input and output voltages, no
dc component in the capacitor voltages, and the employment of a single SCleg and four
bidirectional [Link] analysis of the converter, equivalent circuits, and simulation
results are reported herein.
AC-AC CONVERTER USING SWITCHED CAPACITOR PRINCIPLE 163

2. PROPOSED SC AC–AC CONVERTER

The proposed SC ac–ac converter is presented in Fig. 1. Thisconverter operates as a step-
down converter when configured as shown in Fig. 1(a) or as a step-up converter when
configured according to Fig. 1(b). The only difference is that the points where the source and
the load are connected have to be interchanged. The circuit has four bidirectional switches
represented as S1, S2, S3, and S4 and three capacitors represented as C1, C2, and C3. Fig.
1(c) and (d) shows the proposed gate signals and the practical implementation of one
bidirectional switch by making use of two MOSFETs, respectively.
In this topology, one half of the high-side voltage is applied to C2 and C3. Capacitor C1,
being connected to C2 and after this to C3 in each switching period, equalizes the input and
output voltages.

(a)
164 SAI DARSHAN S, RAMESHA S S AND SACHIN ANGADI

(b)

Figure 1 : Proposed SC ac–ac converter: (a) step-down configuration, (b) step-up configuration,
(c) gate drive signals, and (d) bidirectional switch model and its practical implementation using
two MOSFETs.
AC-AC CONVERTER USING SWITCHED CAPACITOR PRINCIPLE 165

Figure 2 : Theoretical waveforms of the proposed ac–ac converter: (a) input and output
voltages; (b) voltages across the capacitors; and (c) voltage across the switches.

3. PRINCIPLE OF OPERATION
This section presents the principle of operation of the proposedSCac– ac converter in two
steps, i.e. first analyzes the main characteristics of the converter at low frequency (frequency
of the input voltage) and second, gives the information about the operation in different
stages of the converter of the switching period (high-frequency analysis).
3.1 Low-Frequency Analysis
The low-frequency analysis presents the principle of operation, the main waveforms, and
voltage stresses on the components of the proposed converter in a period of the input source.
The SC C1 in the ac–ac converter shown in Fig. 1(a) and (b) ensures the voltage balance
between the capacitors C2 and C3. Consequently, it operates as a step-down or step-up
converter and the voltage stresses on the capacitors and on the switches are one-half of high
166 SAI DARSHAN S, RAMESHA S S AND SACHIN ANGADI

voltage vH. For both operation modes, the expected waveforms of the proposed converter
are illustrated in Fig. 2, where the input and output voltages are in Fig. 2(a), the voltages
across the capacitors are in Fig. 2(b), and the voltages across the switches are in Fig. 2(c),
Vkp being the peak value ofthe high voltage vH .
3.2 High-Frequency Analysis
An SC converter can operate in three modes, which are defined by the charge of the SC.
These modes are complete charge (CC), partial charge (PC), and no charge (NC). The best
operation regions are in PC and NC modes. The NC mode usually requires a high operation
frequency and large capacitances, and the advantages in relation to PC mode are not
significant; thus, the PC mode is considered for the proposed converter. Therefore, the
waveforms described in this section, in the operation stages, relate to the converter operating
in PC mode. The proposed ac–ac converter operating as a step-down converter presents two
operation stages per switching period. During the positive half of the grid voltage, with the
converter in the step-n configuration, these stages can be described as follows.
First stage starts when switches S1 and S3 are turned ON. Capacitor C2 discharges and
capacitor C3 charges during the first part of this stage (Δt1A). When their currents reach
zero, C2 starts to charge and C3 starts to discharge until the end of the stage (Δt1B).
Capacitor C1 charges throughout this stage and the power source vH delivers energy to the
circuit. Switches S1 and S3 are turned OFF at the end of the first stage. This topological
stage is shown in Fig. 4(a).
Second stage starts when switches S2 and S4 are turned ON. Initially, the power source
receives energy from the circuit, capacitor C2 discharges, and capacitor C3 charges until
their currents reach zero (Δt2A). After this, the power source delivers energy to the circuit,
capacitor C2 charges and capacitor C3 discharges until the end of the stage (Δt2A).
Capacitor C1 discharges throughout this stage. Switches S2 and S4 are turned OFF at the end
of the second stage. This topological stage is shown in Fig. 4(b).
After the second stage, another switching period starts from the first stage.
In the negative half-cycle of the grid, the converter has similar operation stages with
different current directions, as can be seen in Fig. 3(c) and (d). The main theoretical
waveforms for the positive and the negative half-cycles of the grid are shown in Fig. 3(a)
and (b), respectively.
AC-AC CONVERTER USING SWITCHED CAPACITOR PRINCIPLE 167

Figure 3 : Main theoretical high-frequency waveforms. (a) Positive half-cycle of the grid voltage.
(b) Negative half-cycle of the grid voltage.
168 SAI DARSHAN S, RAMESHA S S AND SACHIN ANGADI
AC-AC CONVERTER USING SWITCHED CAPACITOR PRINCIPLE 169

Figure 4 : Topological stages for step-down configuration. Positive half-cycle of the grid voltage:
(a) First stage. (b) Second stage. Negative half-cycle of the grid voltage: (c) First stage.
(d) Second stage.
170 SAI DARSHAN S, RAMESHA S S AND SACHIN ANGADI

Simulation of proposed converter

The proposed converter simulated by using SABER simulation software. Two MOSFET’s
are connected as shown in fig.1(d) to make bidirectional ac switch , like wise we have used
8 MOSFET’s to make 4 switches out of it and the pulses which are required to drive the
MOSFET,s are generated by the pulse drive obtained in the simulation software.
Simulation of converter in step down mode
The simulated circuit configuration for step down mode is shown in fig and the waveforms
are plotted according to the configuration.
The main specifications and components used in the simulated circuit are presented in the
Table I.
Fig.5. (a) shows the input and output voltage waveforms. The output voltage follows the
shape of the input voltage with half the magnitude and also the waveforms in the fig.5.
(b)represents voltage across the switches and the following waveforms in fig.5. (c)
represents the gate pulses given to drive the switches. The voltage across the switches is also
half of the input voltage. Thus stress voltage in all switches being equal to half of the input
voltage

Table I
Description Specification Quantity
Input voltage 40V(max) -
Output voltage 20V(max) -
Input voltage frequency 50Hz -
Switching frequency 1KHz -
Capacitors(C1,C2,C3) 10uF 3
MOSFETs for bidirectional switches Ideal 8
Diodes Ideal 8
Load resistance 1000 ohm 1
AC-AC CONVERTER USING SWITCHED CAPACITOR PRINCIPLE 171

(a)

(b)
172 SAI DARSHAN S, RAMESHA S S AND SACHIN ANGADI

(c)
Fig. 5. Simulation of the step down configuration mode (a) Step down configuration circuit. (b)
Voltage waveforms across switches and gate pulse. (c)Waveform of input and output voltage
and source current.
Simulation of converter in step up mode
The simulated circuit configuration for step up mode is shown in fig.6. (a), and the
waveforms are plotted according to the configuration which is shown in fig.6. (b).
The main specifications and components used in the simulated circuit are presented in the
TableII.
Fig.6. (b) shows the input and output voltage waveforms. The output voltage has the shape
of the input voltage with half the magnitude. Thus stress voltage in all switches being equal
to half of the input voltage
Table Ii
Description Specification Quantity
Input voltage 40V(max) -
Output voltage 80V(max) -
Input voltage frequency 50Hz -
Switching frequency 10KHz -
Capacitors(C1,C2,C3) 10uF 3
AC-AC CONVERTER USING SWITCHED CAPACITOR PRINCIPLE 173

Description Specification Quantity

MOSFETs for bidirectional switches Ideal 8
Diodes Ideal 8
Load resistance 100 ohm 1

(a)

++-
(b)
Fig. 6. Simulation of the step up configuration mode (a) Step up configuration circuit. (b)
Waveforms of input and output voltage
174 SAI DARSHAN S, RAMESHA S S AND SACHIN ANGADI

CONCLUSIONS
The following conclusions can be drawn from the theoretical analysis and the experimental
results reported in this paper:
1) The proposed ac–ac static power converter employs only capacitors and switches;
2) The converter employs a single SC and has a common reference between output and
input voltages;
3) The converter operating in open loop with a constant duty cycle of 0.5 presents better
voltage regulation than its autotransformer counterpart with the same Specifications;
4) The topology does not require complex control algorithms, the conversion efficiency is
high,
5) The circuit can operate as a step-down converter (static gain of 0.5) or a step-up
converter (static gain of 2), in both modes, only one-half of the high-side voltage is
applied to the switches and the capacitors;
6) The theoretical analysis results were corroborated by the simulated results;
7) The proposed converter is a potential candidate for substituting the conventional
autotransformer in low-power applications.

REFERENCES

[1] Lazzarin T. B., Andersen R. L., Martins G. B., and Barbi I. , “A 600 W switched-capacitor ac-
ac converter for 220 V/110 V and 110 V/220 V applications,” IEEE Trans. Power Electron., vol.
27, no. 12, pp. 4821–4826, Dec. 2012.
[2] Ioinovici A., “Switched-capacitor power electronics circuits,” IEEE Circuits Syst. Mag., vol. 1,
no. 3, pp. 37–42, Third Quarter 2001.
[3] Seeman M. D. and Sanders S. R., “Analysis and optimization of - DC converters,” IEEE Trans.
Power Electron., vol. 23, no. 2, pp. 841–851, Mar. 2008.
[4] Kimball J. W. and Krein P. T., “Analysis and design of switched capacitor converters,” in Proc.
IEEE Appl. Power Electron. Conf., Mar. 2005, pp. 1473–1477.
[5] Kimball J. W., Krein P. T., and Cahill K. R., “Modeling of capacitor impedance in switching
converters,” IEEE Power Electron. Lett, vol. 3, no. 4, pp. 136–140, Dec. 2005.
[6] On-Cheong M. and Ioinovici A., “Switched-capacitor inverter with high power density and
enhanced regulation capability,” IEEE Trans. Circuits Syst. I: Fundam. Theory Appl., vol. 45,
no. 4, pp. 336–347, Apr. 1998.
International J .of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 175-185

BUCK – BOOST CONVERTER FOR EFFICIENT

UTILIZATION OF SOLAR PHOTO VOLTAIC
ENERGY FOR MULTIPLE UTILITIES

LALITHA DARBHA1, SNEHA K2, VIDYA H3,

SHRADHA M 4 AND SHRIVATSAV M5
1.
Assistant Professor, Department of Electrical & Electronics,
S. G. Balekundri Institute of Technology, Belgaum, Karnataka, India
2,3,4 &5 th
8 Semester UG Students of B.E, Department of Electrical & Electronics,
S.G.B.I.T, Belgaum, Karnataka, India

Abstract
In the present world, nearly 1.5 billion people i.e. more than 1/4th of world’s population have no access to the
electricity. About 85% of the world’s population lives in the villages. As said by the international energy agency
the need for energy would increase by 50% more than the present. India being highly populated, the requirement
of power is increasing continuously. India is being expected to be the 2 nd largest energy consumer by 2025. At
present the source for 70% of the power generation in India is [Link] of economy of any country is
indicated by the percapita utilization of electrical energy. Conventional sources like coal and oil are causing
global warming in addition to scarcity and increase in cost. Renewable sources like hydal and wind vary
seasonally. In a tropical country like India, exploring solar Photo Voltaic based applications gives advantage of
availability of 8 hours a day for more than 9 months in a year. Output of Photo Voltaic cells is dc and stored in
battery. As the conventional electric supply system is ac based, Inverter & Step up converters are used. At the
utilization level, most of these loads require rectification and step down of voltage levels. Increase in the number
of stages of conversion increases complexity, cost and reduces overall efficiency. This Paper proposes to
improve the efficient utilization of renewable energy by a single stage converter known as Buck –Boost
converter. This converter can provide a wide range of voltage that can be utilized for huge number of
applications. Thus thereby reducing the transmission and distribution costs.

---------------------------------------
Keywords : Photo Voltaic, Buck Boost converter, PWM, Microcontroller
© http: //[Link]
176 LALITHA DARBHA, SNEHA K, VIDYA H, SHRADHA M AND SHRIVATSAV M

1. INTRODUCTION
The demand for Solar PV systems is growing worldwide because of rapid developments in
technology resulting in lowering payback periods. Solar PV cells convert light energy from
sun directly to electrical energy. As energy from sun is not available continuously
throughout the day, it has to be stored in battery. To protect batteries from overcharging or
deep discharge, charge controllers are used. The batteries for Solar PV applications are
usually of lead acid type for its low cost, long life, maintenance free [1]. Motors working on
Battery are generally used in Railways, Material handling equipments, Forklifts, Stackers,
Pallets, Hydraulic power packs & Bus [Link] power obtained from the solar
panel is dc that depends on the size and rating of the panel.. This available power can be
utilized for many dc applications with the variation in the voltages as per the device
requirements. With the improvement in capabilities and the reduction in the cost of power
electronic components like IGBT’s, BJT’s, MOSFET’s etc it has been possible to collect
energy, store using battery, protect using the charge controllers [2][3][4]. A buck-boost
converter can work as a boost or buck converter depending on the input and the output
conditions.
Conventionally AC has the advantage of transformer, which facilitates step up or step down
of voltage levels. In dc for low power applications buck- boost converter provides the
variable voltages as per the device requirements. As the dc power is directly being used for
the dc appliances the conversion of power from ac to dc is avoided. The transmission losses
are eliminated as generated power need not be transmitted over longer distances by grid.
This increases the efficiency of the system. Application of for Low voltage High currentDC-
DC Converters for Hydrogen production were discussed by Peter [Link][5]. For Renewable
energy sources like wind, Input voltage varies widely above and below the rated voltage.
Single phase buck boost converters are designed for wind based systems. [6]. Studies with
the variable input voltage for PV based applications were done.[7] . Most of the study was
focussed on input voltage control charging a 12V battery with buck boost converter for input
voltage variation from 5V to 20 V was studied. [8]. Applications of Buck Boost converter
for improvement of power factor was discussed. [9] [10]. Application of buck boost
converter for USB charging at 5V was patented. Applications for direct dc-dc utilization for
Light, DC fan, Mobile, Radio were explored using voltage regulators. [11][12]. The
disadvantage with the voltage regulator say LM7805 to charge a mobile rated 4.5 volts from
BUCK – BOOST CONVERTER FOR EFFICIENT UTILIZATION… 177

a battery of 12V will be to dissipate the drop of 7 V continuously as heat. Similarly few
application require DC of 24V, 32V etc. Conventional transformers cannot be used in DC for
step up. This paper discusses output voltage variation for multiple utilities using buck boost
converter.

2. DESCRIPTION
DC-DC converters are of 4 types, Buck (Output voltage is less than input), Boost (Output
voltage is greater than the input voltage), Buck Boost, Cuk Converters.
2.1 Buck Converter
Buck Converter is one in which the output voltage is less than the input/supply voltage.
Buck is a non isolated power topology also called as step down power storage. The basic
operation of the buck converter has the current in an inductor controlled by two switches, a
transistor and a diode. Specifically, the switch and the diode have negligible voltage drop
when in on state and zero current flow when off state and the inductor has zero series
resistance. There are two modes of operation: Mode 1: It begins when switch
(IGBT/MOSFET) is switched on at t=0, the input current rises and flows through the filter
inductor L, filter capacitor C and load resistor R. Mode 2: It begins when IGBT/MOSFET is
switched off at t=t1, the freewheeling diode D conducts due to energy stored in the inductor
and the inductor current continues to flow through L, C, load and diode D. Then with the
decrease in the magnitude of inductor current, mode 1 repeats and the cycle continues.
2.2 Simulated circuits for DC to DC buck converter

Figure 1: Buck circuit simulated in MATLAB for output of 6V for Radio.

178 LALITHA DARBHA, SNEHA K, VIDYA H, SHRADHA M AND SHRIVATSAV M

Figure 2 : Buck circuit simulated in MATLAB (application for cellphone with 4.5V/450mA)
When switch is ON for 'DT' seconds, it is supplying current to the load, initially current flow
to the load is restricted. Diode 'D' becomes reverse bias, as energy is also being stored in
inductor' L' and voltage across inductor 'Vdc-Vo' which is positive so that inductor current
'IL' increases linearly, voltage across switch becomes zero.
When switch is OFF for '(1-D)T' seconds. Diode 'D' becomes forward bias and voltage
across inductor '-VO' which is negative so that inductor current 'IL' decreases linearly,
voltage across switch is 'Vdc'.
According to KCL,
IL=IO+IC
Switch ON: IL > IO, Remaining current goes through capacitor that is capacitor starts
charging.
Switch OFF: IL<IO, Capacitor starts supplying the difference in current that is capacitor
discharging until the next cycle starts.
Ideal condition:
As all components are ideal, whatever the input power is given the same output appears.
INPUT POWER = OUTPUT POWER
Vdc*Idc = Vo*Io (If , Vo=D*Vdc)
Idc = D*Io
Therefore average Idc current < average Io current, similar to step down transformer
Estimation of inductance:
Vinmax = maximum input voltage,
Vinmin = minimum input voltage,
Vout = desired output voltage
BUCK – BOOST CONVERTER FOR EFFICIENT UTILIZATION… 179

Dbuck = minimum duty cycle for buck mode,

Dboost = maximum duty cycle for boost mode
η = estimated efficiency at calculated Vin, Vout, and Iout,
Fsw = switching frequency of the converter

L>

Kind = estimated coefficient that represents the amount of inductor ripple current relative to
the maximum output current.
A good estimation for the inductor ripple current is 20% to 40% of the output current, or
0.2 < Kind < 0.4.
2.3 Boost Converter
Boost Converter is one in which the output is more than the input/supply voltage. It is also
called as step up power storage. The input current to the boost power stage is continuous
/non-pulsating. The fig below indicates the block diagram of a boost converter. There are
two modes of operation:
Mode 1: It begins when IGBT/ MOSFET is switched on at t=0, the input current rises and
flows through inductor L and switch.
Mode 2: It begins when IGBT/ MOSFET is switched off at t=t1, the current flowing through
the switch would now flow through L, C, load and diode D. Then with the decrease in the
magnitude of inductor current, mode 1 repeats and the cycle continues.
2.4 Simulated circuits for DC to DC boost converter

Figure 3 : MATLAB simulated circuit for application in Dc motors rated at 24V

180 LALITHA DARBHA, SNEHA K, VIDYA H, SHRADHA M AND SHRIVATSAV M

2.5 Operation
Switch is closed for DT (0 to DT).when switch is closed diode is off, capacitor supplies the
power to load. when switch is open the stored energy in the inductor is supplied to the
[Link] Boost converter is the one where Inductor resistance is [Link]-Ideal is the one
where Inductor resistance is taken into [Link] Ideal case, voltage & currents can be
found by simply applying KVL & KCL [Link] Vo is constant & ripple free
i) When Switch is closed : VL = Vdc Current increases io= C (dVo/dt) – (Vo/R)
ii) When Switch is Open: VL = Vdc – Vo and Current decreases
Transfer Function is given by Vo= Vdc/ (1-D)
1 For ideal case : Vo=Vdc when D=0 Vo=V(infinity) when D=1
2 For non ideal case Vo=Vdc when D=0 Vo=V(infinity) when D=1
And even when duty cycle was set to 1 output observed was Vo=0 volts for non ideal
converter
2.6 Estimation of inductance

L>

2.7 Buck-boost converter

Buck-Boost is the one in which the buck as well as boost operation is possible i.e the input
voltage can be varied from minimum value to a maximum value. The system operates as a
buck converter for duty cycles between 0-50% and as a boost converter for duty cycles
between 50-100%.The fig (1)below indicates the block diagram of a buck-boost converter.
Working operation:
There are two modes of operation in buck boost converter as in the buck converter and the
boost converter.
Mode 1: In this mode the IGBT/ MOSFET is turned on and diode is reverse biased, the
input current rises and flows through the inductor and IGBT/ MOSFET . Mode 2: IGBT/
MOSFET is switched off and the current which was flowing through inductor would flow
through capacitor, diode and the load. The energy stored in the inductor will be transferred to
load and the inductor current would fall until IGBT/ MOSFET is switched on again in the
next cycle. The output obtained by the buck boost converter has the reverse polarity and
hence called as an inverting regulator. It has very high efficiency but can be used only for
low power applications. LC filters are used to reduce the ripple content of the output voltage.
BUCK – BOOST CONVERTER FOR EFFICIENT UTILIZATION… 181

The average output voltage in a buck-boost converter is given by,

Vo=k*Vin/(1-k),
where k is duty cycle, if the duty cycle is 25%, then Vo=(0.25*Vin)/0.75=Vin/3
if duty cycle is 75%, then Vo= Vo=(0.75*Vin)/0.25=3*Vin,
thus by varying the duty cycle average output voltage can be varied.

Figure 4 : Buck boost circuit simulated in MATLAB for output of 4.5V for Mobile Charging

Figure 5 : Simulated output for printer at 32V

The figure 4 above represents the circuit used for charging mobile with pulse width of 42%
proposed converter gave simulated output of 4.5 V for mobile charging. Fig 5 shows the
expected output of 32V for printer. A step up output for 220 V is obtained in Fig 6 which is a
modified buck boost converter with normal polarity, whereas the other circuits show
reversed polarity.
182 LALITHA DARBHA, SNEHA K, VIDYA H, SHRADHA M AND SHRIVATSAV M

Figure 6 : Simulated output for 220 V DC( For application in motors)

3. EXPECTED OUTPUT FROM THE SIMULATED CIRCUITS OF THE

CONVERTER

Output voltage L(in uH) Pulse Application

4.5V 1000 42% Mobile charger

6V 1000 54% Radio

24V 750 65% 24V dc motor

32V 100 75% Printer

48V 100 84% Telecommunication

215V 750 30% Dc motor

Simulated results show a wide range of control with variation in inductance and by using
pulse width modulation technique.
BUCK – BOOST CONVERTER FOR EFFICIENT UTILIZATION… 183

4. RESULTS

Figure 4 : Solar Panel of 75 W Figure 5 : Battery of 75AH

BUCK CONVERTER BOOST CONVERTER

VDC = 12 V,f = 4.37 kHz, D = 62.5 %,VO = 6.26 V Input -12VOutput of Boost converter :17.96V.
BUCK BOOST CONVERTER

Input [Link] of Buckboost converter 19V

184 LALITHA DARBHA, SNEHA K, VIDYA H, SHRADHA M AND SHRIVATSAV M

The Capacitor is not able to hold the charge for longer time. And the inductor was getting
burnt due to high current, With addition of parallel capacitor at the input that problem was
solved.
Duty cycle variation was done using Micro controller. Limitations for using micro 8051
controller triggering circuit are: Voltage generated by the micro controller is not sufficient
enough to trigger the switch hence amplification / drivers are required. When duty cycle is
changed , frequency is also getting changed .generation of large frequencies is not possible
without duty cycle variation

CONCLUSIONS
 Buck boost converter facilitates transformer like action by providing variable output
voltage as required by applications.
 Direct dc is fed to appliances hence ac to dc conversion is avoided.
 The losses occurring due to transmission and distribution are minimized.
 High efficiency & reliability.
 Minimum heat generation.
No magnetic circuit is required.

FUTURE SCOPE
1. As buck boost converter provides a reversed polarity at output, suitable converter
topology could be explored to avoid reversal.
2. Simple, low cost and Efficient PWM strategy could be explored.

ACKNOWLEDGMENT
We thank the members of Vision Group of Science and Technology, as the idea was partly
executed by utilizing resources procured under VGST/ TRIP scheme. We also thank
KSCST, Dept of Science & Technology, Govt. of Karntataka for selecting our project for
sponsorship under 37th series of SPP. We also thank Principal & Management of SGBIT
Belgaum for providing necessary infrastructure. We thank Prof Rakhee Kallimani,
KLESCET Belgaum for her encouragement.
BUCK – BOOST CONVERTER FOR EFFICIENT UTILIZATION… 185

REFERENCES

[1] Prajapati Manish, Shrestha Sadhana and Bista Diwakar, “Design and testing of prediction based
battery charging technique for solar home system ”, Rentech Symposium.
[2] jamil Majid, Kirmani Sheeraz, Rizwan Mohammad (2012) “Techno-Economic Feasibility
Analysis of Solar Photovoltaic Power Generation”, Smart Grid and Renewable Energy, 2012, 3,
266-274, doi:10.4236/sgre.2012.34037.
[3] Chen Jingquan, Maksimovic, D. Erickson, R. “Buck-boost PWM converters having two
independently controlled switches”, power electronics specialists conference, vol.2 ,pp. 736 –
741, 2001.
[4] Jafari. M, Hunter. G, Jian Guo Zhu, “A new topology of multi-input multi-output Buck-Boost
DC-DC Converter for micro grid applications”, power and energy conference, 2012, pp.2-5 dec
[5] Kosmatin Peter, Petkovšek Marko, Vončina Danijel,“ High-efficiency DC/DC converter for
low-voltage applications”, Elektrotehniški vestnik 77(2-3): 109-113, 2010,Electrotechnical
Review: Ljubljana, Slovenija.
[6] Bing Hu ,Liuchen Chang , Yaosuo Xue, , ” Research on a novel buck-boost converter for wind
turbine systems “,Sustainable Energy Technologies, 2008. ICSET 2008, IEEE International
conference, Singapore, 24-27 Nov 2008,PP 228-233
[7] Chuan Yao, Xinbo Ruan, ieee, Xuehua Wang, And Chi K. Tse, “isolated buck–boost dc/dc
converters suitable for wide input-voltage range”, ieee transactions on power electronics, vol.
26, no. 9, september 2011.
[8] Diary R. Sulaiman, Hilmi F. Amin,” Design of High Efficiency DC-DC Converter for
Photovoltaic Solar Home Applications”
[9] jamil Majid, Mehdi Zahra, “power factor improvement of cascaded buck boost converter”,
national power electronics conference
[10] Matsui.k, Yamamoto.L, Kishi.T, Hasegawa,M, Mori.H, Ueda.F, “A comparison of various
buck-boost converters and their application to PFC”, IECON [Industrial Electronic Society], Vol
1, pp.30-36, 5-8 Nov.2002
[11] Darbha Lalitha et al “DC-DC Chargers for Efficient Utilization of Renewable Energy”,
International Journal of Engineering Research & Technology (IJERT), Vol. 2 Issue 11,
November – 2013.
[12] Darbha Lalitha, Sneha K, Shradha M “Integrated multi utility hybrid module for isolated
households in remote villages”, International Journal Of Advance Research In Science And
Engineering IJARSE, Vol. No.2, Issue No.10, October 2013.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 187-199

CAPACITY ENHANCEMENT TECHNIQUES OF

DISTRIBUTED UNIFIED POWER QUALITY
CONDITIONER (D-UPQC) - A REVIEW

VEERESHA. K. B.1 AND THEJASWI. A. H.2

1
Asst. Professor. 2Professor
Dept. of E&E Engg., JNNCE, Shimoga-577204, Karnataka, India.

Abstract
The widespread use of Power Electronics based equipment has produced a significant impact on
quality of electric power supply. Conventional power quality mitigation equipment is providing to be
inadequate for an increasing number of applications. One modern and very promising solution that
deals with both load current and supply voltage imperfections is the Unified Power Quality
Conditioner (UPQC). This paper presents comprehensive review on the UPQC capacity enhancement
in high power application using Multi-level/module topologies, such as Multi-converter, Multi-
module converter and Multi-level (power cell) based to improve the electric power quality at
distribution levels.

---------------------------------
Keywords : Power Quality (PQ), Unified power quality conditioner (UPQC), Centralised & Distributed APF
Systems. © http: //[Link]
188 VEERESHA. K. B. AND THEJASWI. A. H.

1. INTRODUCTION
Everyday sophisticated and sensitive loads are wide used for every industry and
organizations. These loads are very much sensitive to power quality issues, power quality
become a major issue, both in customers and utilities point of view. UPQC is one of the
device used to improve power quality as per the study by the researchers [1]. The UPQC has
two converters that share one dc-link capacitor.
The presently developed UPQC operates in much lower dc-link voltage than the operation
voltage of the distribution system. The restriction in dc-link voltage is due to the maximum
sustained voltage of the switching element.
In high power applications, the filtering task cannot be performed for the whole spectrum of
harmonics by using a single converter due to the limitations on switching frequency and
power rating of the semiconductor devices. Therefore, compensating the reactive harmonic
components to improve the power quality of the DG integrated system as well as to avoid the
large capacity centralized APF has meant that parallel operation of multiple low power APF
units is increasing. Like APF, UPQC can also be placed at the PCC or at a high voltage
distribution line as a part of a DG integrated network or in microgrid system to work both in
interconnected or islanded mode.
The UPQC units developed so far for capacity expansion is mainly for medium or high
voltage distribution lines where both the current and voltage are high. Therefore, it may be
easier to use these multi-level or multi-modular type UPQC systems [1-4]. But these
configurations may not be suitable for low voltage distribution lines where the 3-phase
voltage (typically 400Vline) is low, but the required compensating current is comparatively
high (100A or more). In that case, multi-level or module type series APF may not be
required. But for high current compensation, multiple Shunt APF units in a distributed
(parallel) mode and connected with a common dc linked capacitor can be a solution in
developing a Distributed UPQC (D-UPQC) [5-7].

2. SYSTEM CONFIGURATION OF UPQC

The Unified Power Quality Conditioner is a custom power device that is employed in the
distribution system to mitigate the disturbances that affect the performance of sensitive
and/or critical load [8]. It is a type of hybrid APF and is the only versatile device which can
mitigate several power quality problems related with voltage and current simultaneously
CAPACITY ENHANCEMENT TECHNIQUES OF DISTRIBUTED… 189

therefore is multi functioning devices that compensate various voltage disturbances of the
power supply, to correct voltage fluctuations and to prevent harmonic load current from
entering the power system.

Figure 1 : UPQC general block diagram

Fig. 1 shows the system configuration of a single-phase UPQC.
Unified Power Quality Conditioner (UPQC) consists of two IGBT based Voltage source
converters (VSC), one shunt and one series cascaded by a common DC bus. The shunt
converter is connected in parallel to the load. It provides VAR support to the load and supply
harmonic currents. The series converter connected in series to the load provides voltage
compensation [9]. Thus UPQC improves the power quality by preventing load current
harmonics and by correcting the input power factor. The main components of a UPQC are
series and shunt power converters, DC capacitors, low-pass and high-pass passive filters, and
series and shunt transformers:
2.1 Series Converter
It is a voltage-source converter connected in series with the AC line and acts as a voltage
source to mitigate voltage distortions. It is used to eliminate supply voltage flickers or
imbalance from the load terminal voltage and forces the shunt branch to absorb current
harmonics generated by the nonlinear load. Control of the series converter output voltage is
usually performed using sinusoidal pulse-width modulation (SPWM). The gate pulses
190 VEERESHA. K. B. AND THEJASWI. A. H.

required for converter are generated by the comparison of a fundamental voltage reference
signal with a high-frequency triangular waveform.

2.2 Shunt Converter

It is a voltage-source converter connected in shunt with the same AC line and acts as a
current source to cancel current distortions, compensate reactive current of the load, and
improve the power factor. It also performs the DC-link voltage regulation, resulting in a
significant reduction of the DC capacitor rating. The output current of the shunt converter is
adjusted using a dynamic hysteresis band by controlling the status of semiconductor switches
so that output current follows the reference signal and remains in a predetermined hysteresis
band.
2.3 Midpoint-to-ground DC Capacitor Bank
It is divided into two groups, which are connected in series. The neutrals of the secondary
transformers are directly connected to the DC link midpoint. As the connection of both
three-phase transformers is Y/Yo, the zero-sequence voltage appears in the primary winding
of the series-connected transformer in order to compensate for the zero-sequence voltage of
the supply system. No zero sequence current flows in the primary side of both transformers.
It ensures the system current to be balanced even when the voltage disturbance occurs.
2.4 Low-pass Filter
It is used to attenuate high frequency components at the output of the series converter that
are generated by high frequency switching.
2.5 High-pass Filter
It is installed at the output of shunt converter to absorb current switching ripples.
2.6 Equivalent Circuit

Figure 2 : Equivalent circuit for UPQC

CAPACITY ENHANCEMENT TECHNIQUES OF DISTRIBUTED… 191

VS: Voltage at power supply, VSR: Series-APF for voltage compensation,

VL: Load voltage and ISh: Shunt-APF for current and VSR compensation.
Due to the voltage distortion, the system may contain negative phase sequence and harmonic
components.
In general, the source voltage in Figure 2 can be expressed as:
Vs + VSR = VL (1)
To obtain a balance sinusoidal load voltage with fixed amplitude V, the output voltages of
the series-APF should be given by:

VSR  (V – V1p ) sin (t  1p ) – VLN  t     = 2 Vk  t  (2)

k

Where, V1p: Positive sequence voltage amplitude fundamental frequency

θ1p: Initial phase of voltage for positive sequence
VLn: Negative sequence component
The shunt-APF acts as a controlled current source and its output components should include
harmonic, reactive and negative-sequence components in order be to compensate these
quantities in the load current, when the output current of shunt APF Ish is kept to be equal to
the component of the load as given in the following equation:

IL  I1p cos(t  1p )sin1p  ILN    = 2 ILK (3)

K

1p  1p  1p (4)

Where, 1p : Initial phase of current for positive sequence.

As seen from the above equations that the harmonic, reactive and negative sequence current
is not flowing into the power source. Therefore, the terminal source current is harmonic-free
sinusoid and has the same phase angle as the phase voltage at the load terminal.
IS  IL  ISh  I1psin (t  1p ) cos1p (5)

3. CENTRALISED & DISTRIBUTED APF SYSTEMS

3.1 Limitations of centralized system (APF)
 One large unit is required to compensate the reactive and harmonic power arising from
all critical loads rather than the specified loads.
192 VEERESHA. K. B. AND THEJASWI. A. H.

 Impractical for some applications because of its high initial cost,

 Difficulties in installation due to large size and weight
 Single-point-of-failure reduces its reliability.
 High relative cost for achieving redundancy and increasing the capacity with expansion
of the load.
3.2 Distributed/Multi-unit parallel approach
 In a distributed approach, many separate APF units operate in parallel. APF units are
placed flexibly in the system to form a critical network.
 Highly reliable because of redundancy and also easy to achieve this redundancy
 Highly flexible and easy to increase the capacity when the load increases
 User-friendly in terms of maintenance

4. CAPACITY ENHANCEMENT TECHNIQUES

At this place, capacity enhancement is achieved by using Multi-level/module topologies to
reach the higher power levels. These options are as follows [17]:
4.1 Multi-converter UPQC (MC-UPQC)
A multi-level converter is proposed to increase the converter operation voltage, avoiding the
series connection of switching elements. However, the multilevel converter is complex to
form the output voltage and requires an excessive number of back connection diodes or
flying capacitors [1] or cascade converters [3]. A typical multi-level UPQC is shown in Fig
3. It consists of three VSCs (VSC1, VSC2, and VSC3) which are connected back to back
through a common dc-link capacitor. In the proposed configuration, VSC1 is connected in
series with BUS1 and VSC2 is connected in parallel with load L1 at the end of Feeder1.
VSC3 is connected in series with BUS2 at the Feeder2 end [10].
Each of the three VSCs in Fig. 3 is realized by a three-phase converter with a commutation
reactor and high-pass output filter. The commutation reactor and high- pass output filter are
connected to prevent the flow of switching harmonics into the power supply [14].
As shown in Fig. 3, all converters are supplied from a common dc-link capacitor and
connected to the distribution system through a transformer. Secondary (distribution) sides of
the series-connected transformers are directly connected in series with BUS1 and BUS2, and
CAPACITY ENHANCEMENT TECHNIQUES OF DISTRIBUTED… 193

the secondary (distribution) side of the shunt-connected transformer is connected in parallel

with load L1. The aims of the MC-UPQC shown in Fig.3 are:
1) To regulate the load voltage against sag/swell and disturbances in the system to protect
the nonlinear/sensitive load L1; 2) To regulate the load voltage against sag/swell,
interruption, and disturbances in the system to protect the sensitive/ critical load L2; 3) To
compensate for the reactive and harmonic components of nonlinear load current.

Figure 3 : Typical MC-UPQC used in a distribution system

In order to achieve these goals, series VSCs (i.e., VSC1 and VSC3) operate as voltage
controllers while the shunt VSC (i.e., VSC2) operates as a current controller.
4.2 Multi-module converter UPQC (MD-UPQC)
A multibridge converter composed of several H-bridge modules for each phase, was
proposed to increase the converter operation voltage. A multi-module H-bridge UPQC can
be connected to the distribution system without series injection transformers. It has the
flexibility in expanding the operation voltage by increasing the number of H-bridge modules
[4]. Fig. 4 shows a configuration of the proposed UPQC based on several pairs of H-bridge
modules for each phase. Each pair has two H-bridge modules connected in parallel through a
common dc-link capacitor. The H-bridge module in shunt part is connected in series through
a multi winding transformer, while the H-bridge in series part is directly connected in series
and inserted in the distribution line [11].
194 VEERESHA. K. B. AND THEJASWI. A. H.

The proposed UPQC has a bypass function to remove the series converter from service
during the distribution system fault. The bypass function is implemented by the operation of
an insulated- gate bipolar transistor (IGBT) bypass switch and mechanical circuit breaker.
The line overcurrent can be bypassed first by the IGBT switch and then by the mechanical
circuit breaker. However, it is possible to attempt a bypass scheme using the series converter
switches when the maximum fault current is lower than the maximum current rating of the
converter switches [19, 20].

Figure 4 : Typical MD-UPQC used in a distribution system

4.3 Multi-level (power cell) UPQC (ML-UPQC)

These Multi-module techniques [3-6] allow the symmetrical distribution of the load power
among the components of the topology, but the classical design procedure must be modified
or refined to ensure the power cell components should be within its maximum ratings.
The UPQC classical topology [15] can be redefined so as to include a multiple power cell
arranged in order to reach higher power levels, while overcoming the limits of the
semiconductors rating. The electric variable that limits the operation of the series
compensator in the UPQC is the total current according to the transformer turns ratio and for
the shunt compensator is the total voltage, because it has to deal with the total load voltage
[16]. The proposed configuration overcomes these limitations by interconnecting multiple
CAPACITY ENHANCEMENT TECHNIQUES OF DISTRIBUTED… 195

power cells as shown in Fig. 5, where we can see nc cells per phase and thus 3nc cells for
the total system. This connection provides a way for the PCC current and load voltage to be
shared among all the modules of the configuration. This justifies the parallel connection at
the series side of the cells and the series connection at the shunt side.
As shown in Fig. 5, the series side compensator is connected through a series transformer, in
order to compensate the PCC voltage disturbances, because the load voltage vabc
L is the result

of the addition of the PCC voltage v abc abc

pcc and the injected series voltage v s On the other

hand, the shunt side of the compensator is connected at the same load voltage, to sink or
inject current harmonics, and compensate the reactive power, because the PCC current i abc
pcc

is composed of the load current i abc abc

L and the shunt current i p of the UPQC.

Figure 5 : Typical ML-UPQC used in a distribution system

Comparison merits and demerits of Multi-level/module topologies UPQC in capacity

expansion to reach the higher power levels are given table-1 and table-2 [12, 13].
196 VEERESHA. K. B. AND THEJASWI. A. H.

Table 1 : Comparison of Multi-level/module topologies

UPQC-MC UPQC-MD UPQC-ML
(Multi-Converter) (Modular) (Multi-level)

1. In UPQC-MC third 1. UPQC-MD several H- 1. UPQC-ML is based on

converter is added to bridge modules are 3-level neutral point
support dc bus. connected in cascade in clamped topology.
each phase.
2. The third converter is 2. The H-bridge modules for 2. In UPQC-ML three
connected either series shunt inverter is connected level topology require
or parallel with feeder. in series through multi- double semiconductor
winding transformer, while switches.
series inverter is connected
in series without using
series transformer.
3. To improve system 3. UPQC-MD can be useful 3. UPQC-ML can also be
performance, use of to achieve higher power useful to achieve
storage battery or super levels. higher power levels.
capacitor at third
converter.

Table 2: Merits and Demerits of Multi-level / module Topologies

Type of UPQC Merits Demerits
1. Multi- i. High voltage and current i. Voltage unbalance could occur
Converter can be achieved between the different levels
based ii. Can be developed in ii. Requires accessive number of
different ways – diode diode / flying capacity / inverter
clamp / flying capacitor / iii. Central control is required and
cascade inverter based it is complicated
iv. Conduction loss is high
v. Capacity expansion is difficult
vi. Centralized Approach
CAPACITY ENHANCEMENT TECHNIQUES OF DISTRIBUTED… 197

2. Multi- i. No series transformer is i. It requires high number of

modular required, thus reduces the switching devices to enhance
transforme cost the capacity
r-less ii. Capacity expansion is ii. Central control is required
easier than multi-level iii. Conduction loss and switching
converter loss also high
iii. Redundancy is possible iv. Due to symmetrical distribution
of the load power among the H-
bridge inverters, modules may
not work at its maximum rating
3. Multi-level i. Single phase power cell i. Number of H-bridge switching
(power cell) topology helps the unit to device increases, thus increase
work at its maximum the switching loss
rating. ii. Transformer for each shunt part
ii. Capacity expansion is of the power cell could increase
easier the loss as well as make the
iii. Redundancy is possible system bulky
iv. Both Central and iii. Additional single shunt unit
Distributed controls are may not be included if only the
possible iv. required compensating current
v. Due to parallel mode of increases
operation, conduction loss
can be reduced

CONCLUSION
 In distribution system power Quality problem is a major issue. Out of the custom power
devices UPQC is the most effective device for mitigating these issues. In this paper
review on the UPQC capacity enhancement in high power application using Multi-
level/module topologies, such as Multi-converter, Multi-module converter and Multi-
level (power cell) based to improve the electric power quality at distribution levels have
been discussed.
198 VEERESHA. K. B. AND THEJASWI. A. H.

 Compared to a conventional UPQC, the MC-UPQC topology is capable of fully

protecting critical and sensitive loads against distortions, sags/swell, and interruption in
two-feeder systems. This concept can be extended to multibus/multifeeder systems.
 MD-UPQC has flexibility in expanding the operation voltage by increasing the number
of H-bridge modules.
 ML-UPQC classical topology can be redefined so as to include a multiple power cell
arranged in order to reach higher power levels, while overcoming the limits of the
semiconductors rating.
 Comparison, merits and demerits of Multi-level/module topologies have been
discussed.
REFERENCES

[1] Lai J, Peng F Z, Multilevel converters-a new breed of power converters, IEEE Trans Ind Appl,
Vol. 32(3), 1996, pp. 509 – 517
[2] Munoz J A, Espinoza J R, “Design of a Modular UPQC Configuration Integrating a
Components Economical Analysis”, IEEE Trans PD, Vol. 24(4), 2009, pp. 1763-1772.
[3] Peng F Z, McKeever J W and Adams D J, A Power Line Conditioner Using Cascade Multilevel
Inverters for Distribution Systems, IEEE Trans IA, Vol 34(6), 1998, pp. 1293 – 1298
[4] Han B., Bae B., Baek S. and Jang G., “New Configuration of UPQC for Medium-Voltage
Application”, IEEE Trans Power Delivery, Vol. 21(3), 2006, pp. 1438 – 1444
[5] Han B., Baek S., Kim H., and Karady G., “Dynamic characteristic analysis of SSSC based on
multibridge inverter”, IEEE Trans. Power Del. Vol. 17(2), 2002, pp. 623–629
[6] Han B.M. and Mattavelli P., “Operation analysis of novel UPFC based on 3-level half bridge
modules”, IEEE Power Tech Conf., Jun. 2003, vol. 4, pp. 307–312.
[7] Muñoz J, Espinoza J, Morán C, Espinosa E, Sbárbaro D, “Design of a Discrete-Time Linear
Control Strategy for a Multi-Cell UPQC”, IEEE Trans IE, 2011
[8] Sankaran C., “Power Quality”, CRC Press LLC, 2002.
[9] Alexander, Kusko and Marc [Link], “Power Quality in Electrical Systems”, McGraw-Hill,
2007.
[10] Mohammadi Hamid Reza, Varjani Ali Yazdian, and Mokhtar Hossein, “Multiconverter Unified
Power-Quality Conditioning System: MC-UPQC” IEEE Trans. Power delivery, vol. 24, no. 3,
pp. 1679–1686, July 2009.
[11] Han B., Bae B., Baek S., and Jang G., “New Configuration of UPQC for Medium-Voltage
Application” IEEE Trans. Power delivery, vol. 21, no. 3, pp. 1438–1444, July 2006.
[12] Hochgraf C., Lasseter R., Divan D., and Lipo T. A., “Comparison of multilevel inverters for
static var compensation,“ in Conj Rec. IEEE./IAS Ann. Mrg., 1994, pp. 921-928.
[13] Teodorescu R., Beaabjerg F., Pedersen J. K., Cengelci E., Sulistijo S., Woo B., and Enjeti P.,
“Multilevel converters — A survey,” in Proc. European Power Electronics Conf. (EPE’99),
Lausanne, Switzerland, 1999, CD-ROM.
CAPACITY ENHANCEMENT TECHNIQUES OF DISTRIBUTED… 199

[14] Peng F Z, McKeever J W and Adams D J, “A Power Line Conditioner Using Cascade
Multilevel Inverters for Distribution Systems”, IEEE Trans Ind Appl, Vol 34(6), 1998, pp.1293
– 1298
[15] Fujita H. and Akagi H., “The unified power quality conditioner: The integration of series- and
shunt-active filters,” IEEE Trans. Power Electron., vol. 13, no. 1, pp. 315–322, Mar. 1998.
[16] Muñoz J., Reyes J., Espinoza J., Rubilar I., and Morán L., “A novel multi-level three-phase
UPQC topology based on full-bridge single-phase cells,” in Proc. IEEE IECON, Nov. 2007, pp.
1787–1792.
[17] Nasiri A. and Emadi A., “Different topologies for single-phase unified power quality
conditioners,” in Proc. 38th Int. Appl. Soc. Annu. Meet. Ind. Appl. Conf., pp. 976-981, Oct. 12-
16, 2003.
[18] Khadkikar V., Agarwal P., Chandra A., Barry A., and Nguyen T., “A simple new control
technique for unified power quality conditioner (UPQC),” in Proc. 11th Int. Conf. Harmonics
Quality Power, pp. 289-293, Sep. 12-15, 2004.
[19] Peng F. and Lai J., “A multilevel voltage-source inverter with separate DC source for static var
generation,” in Proc. IEEE/IAS Annu. Meeting, Orlando, FL, Oct. 8–12, 1995, pp. 2541–2548.
[20] Basu M., Das S. P., and Dubey G. K., “Comparative evaluation of two models of UPQC for
suitable interface to enhance power quality,” Elect. Power Syst. Res., pp. 821–830, 2007.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 201-212

MODELING AND SIMULATION OF VOLTAGE SOURCE

INVERTER BASED ON SPACE VECTOR MODULATION
TECHNIQUE FOR GRID CONNECTION

HARSHAVARDHAN V. KULKARNI
Department of Electrical & Electronics Engg
Acharya Institute of Technology, Bangalore, India.

Abstract
In this paper modeling and simulation of three phase two level 100 kVA space vector modulation
controlled voltage source inverter and its comparison with sinusoidal pulse width modulation
controlled VSI for L-C filter requirements, L-C and L-C-L filters are designed and simulation results
for isolated grid are presented. The simulation is done using MatLab/Simulink ®[Link]
(R2007a).The need for inverters in distributed generation systems reveals the significance of
achieving low distortion, high quality power. The designed filter should attenuate switching frequency
components and should meet the utility voltage, current and distortion limits.

© http: //[Link]
202 HARSHAVARDHAN V. KULKARNI

I. INTRODUCTION
Electric power systems heavily depend on power electronic apparatus, which enhance the
operation of the system. The ability of high power electronic apparatus in controlling the
flow of real and reactive power through designated electrical paths, improvement of dynamic
and transient performance of the system, and support of voltage profile has led to their
widespread use in modern power networks .Among the most employed high power
electronic apparatus are voltage-sourced converters (VSC’s), VSC converts constant dc
voltage to an ac voltage of desired magnitude, frequency and phase. The basic function of a
VSC in the context of power systems is to act as a controlled voltage source. Based on the
direction of the flow of energy, such a converter can be called either an inverter or a rectifier.
Pulse-width modulation methods are widely used for the synthesis of ac voltages at the
terminals of a voltage-sourced converter (VSC) [1].
Traditionally sinusoidal pulse width modulation (SPWM) has been used. A powerful
alternative for this purpose is space-vector modulation (SVM), in which the converter is
placed in a finite number of states in order to best approximate the reference voltage. This
method offers better utilization of the dc bus voltage and provides several degrees of
freedom for enhancement of the harmonic spectrum. Space Vector Modulation (SVM)
Technique has become the most popular and important PWM techniques for three phase
Voltage Source Inverters for the control of drive applications [3-7].

2. METHODOLOGY
The need for inverters in distributed generation systems reveals the significance of achieving
low distortion, high quality power export through inverters. The designed filter should
attenuate switching frequency components and should meet the utility voltage, current and
distortion limits. The output LC and LCL filters are used to reduce the switching frequency
components. The output LC and LCL filters are designed and simulated.

Figure 1 : System configuration of three phase inverter with LC or LCL filter

MODELING AND SIMULATION OF VOLTAGE SOURCE… 203

The Low pass LC filter is used to filter out undesired switching frequency components from
the converter output voltage, in order to meet the requirements set up by Standards for grid
connection as well as loads. The Filter components are sized aiming at getting an ideally
sinusoidal output voltage. As the pulse number is high for the VSI converters considered
here, the effect of low order harmonics on the Filter sizing would be minimal, and hence care
should be taken in handling the dominant harmonics of order and its multiples. fs .Thus by
choosing an odd pulse number for three phase inverters results in an odd symmetry, and
hence all the even harmonics will disappear from the out put phase voltages. And also if the
pulse number is a multiple of 3 in a three phase SPWM, the dominant harmonics at odd
multiples of ‘p’ in the line to line voltages would be suppressed. The Low pass filter
components sizing here is for the purpose of handling harmonics from the SPWM switching
of the VSI, and aimed at getting an ideally ripple free voltage before connecting to the grid
or to loads as specified by standards. According to IEEE 1547 standard which is a standard
for interconnecting distributed resources with electric power systems, and IEEE 519
recommended practices for utilities, for voltages below 69 kV, the Total Voltage Distortion,
and THD must be less than 5%. The assumptions made in design with regard to the AC
system are as follows
 Rated AC voltage: 440 V, 3-phase, 4-wire
 Rated power of converter: 100 kVA
 DC link voltage is assumed to be constant.
For any harmonic order n, per phase equivalent circuit of the LC Filter can be drawn as
shown in Fig.2.

Figure 2 : Equivalent Circuit of LC Output Filter for a harmonic order n

XL 34.333
| |≥
XC n2
n2
XC ≤ × XL
34.333
204 HARSHAVARDHAN V. KULKARNI

For the pulse number of 81 in this design, an attenuation of 3% of the harmonic at the
switching frequency, according to the above assumption
1 812
(X c )p. u= = 34.333 × 0.15 p.u = 28.664 p.u
𝜔𝑐

(X c ) = Zb × (X c )p. u = 1.6 × 28.664 p. u = 45.862 Ω

1 1
|C| = | |= = 69.405 µF
ωb × X C 2π × 50 × 45.862
C = 69.405 µF
Cut off frequency of low pass LC filter will be
1
fc = 2π√LC = 2.19 p. u = 2.19 50 = 109.5 Hz

The designed values of L and C are L= 0.764 mH, C = 69.405 . With these LC filter
values with thefollowing input parameters

Table 1: Parameters used for SVPWM based VSI

Input DC voltage, Vdc 816 Volts
Filter Inductance , Lf 0.99 mH
Filter Capacitance ,Cf 100 µF
Three phase transformer Ratings
kVA capacity 100 kVA
Primary Voltage 600 Volts
Secondary Voltage 440 Volts
Winding configuration Δ / Үg
Frequency 50 Hz
Three phase Grid parameters
Phase to phase voltage 440 volts
Frequency 50 Hz
Space Vector PWM Block
Carrier frequency 6000 Hz
Switching frequency 6000 Hz
MODELING AND SIMULATION OF VOLTAGE SOURCE… 205

Modulation index 0.8

Frequency of output voltage 50 Hz
Three phase resistive load, Yg configuration 105 kW

Table 2 : Parameters used for SPWM based VSI

Input DC voltage, Vdc 816 Volts

Filter Inductance , Lf 0.70 mH

Filter Capacitance ,Cf 70 µF

Three phase transformer Ratings

kVA capacity 100 kVA

Primary Voltage 650 Volts

Secondary Voltage 440 Volts

Winding configuration Δ / Үg

Frequency 50 Hz

Three phase Grid parameters

Phase to phase voltage 440 volts

Frequency 50 Hz

Sinusoidal PWM Block

Carrier frequency 6000 Hz

Switching frequency 6000 Hz

Modulation index 0.8

Frequency of output voltage 50 Hz

Three phase resistive load, Yg configuration 175 kW

206 HARSHAVARDHAN V. KULKARNI

Figure 3 : Simulink model of 100 kVA SVM/SPWM controlled VSI

Figure 4 : SPWM Harmonic order versus line voltage Vab before filtering

Figure 5 : SPWM Harmonic order versus line voltage Vab after filtering
MODELING AND SIMULATION OF VOLTAGE SOURCE… 207

Table 3 : SPWM Line voltage THD comparison

Before Filtering After Filtering
Line Voltages
THD (%) THD (%)
Vab 91.54 1.235
Vbc 91.54 1.235
Vca 91.54 1.235

Figure 6 : Line voltage before filtering

Figure 7 : Filtered line after filtering

Figure 3 shows the simulink model of SVPWM/SVM controlled VSI connected to
[Link] 6 and 7 shows the line voltage before and after filtering Table 3 shows the line
 208 HARSHAVARDHAN V. KULKARNI

voltage THD comparison. Figure 4 and 5 shows harmonic spectrum of line voltage before
and after filtering
Table 4 : Phase voltages before filtering harmonic spectrum
Va Vb Vc
THD =139 % THD =140 % THD =143%

Table 5 : Phase voltages after filtering harmonic spectrum

Va Vb Vc
THD =2.714 % THD =2.8892 % THD =3.0133 %

Table 4 and 5 shows the THD before and after filtering in SPWM technique.
Table 6 : Filter values comparison
LC filter SPWM SVPWM
L 0.9 mH 0.7 mH
C 100 µF 70 µF

Table 6 shows the filter comparison of LC filter requirements for SVPWM and SPWM
control techniques
Table 7 : Phase voltages before filtering harmonic spectrum

Va Vb Vc
THD=103.862 % THD=100.0785 % THD=106.1696 %

Table 8 : Phase voltages after filtering harmonic spectrum

Va Vb Vc
THD=4.0152% THD=4.3319% THD=4.2159%

Table 8 and 9 shows the THD of phase voltages before and after filtering in SVPWM
[Link] THD has reduced below 5% after filtering using LC filter. In the design of
LCL filter the resonance frequency is considered as a design parameter .The requirement for
resonance frequency is that it should be included in the range 10ffund< fres<fsw/2 [20]. Where
ffund is the fundamental frequency of the utility network [Link] the converter side
inductance L1 is determined from equation
MODELING AND SIMULATION OF VOLTAGE SOURCE… 209

(1)

Where is the angular switching frequency?

and are the phase voltage and current on the converter side of the filter
The resonance frequency of the LCL filter is defined as:

(2)

Where r =
The attenuation of the line current amplitude at the switching frequency is defined as

(3)

Where

The resonance frequency in equation 2 can be presented as

and then eqn 2 is

(4)

Where d = is the attenuation of the switching frequency line current amplitude across the

filter .
Finally, r is obtained and it is

r (5)

In order to achieve the capacitor value C, equation (2) is reformulated. The equation for the
capacitor C is

(6)

A 5% THD limit for the line current is specified, that is rule of thumb value for acceptable
current waveform. The parameter values are given in per unit (p.u).The per unit value is
210 HARSHAVARDHAN V. KULKARNI

based on the nominal apparent power of the converter and the line-to-line voltage

.The filter is designed for and .The switching frequency of the

converter is 1.8 kHz.
The base values are determined as

(7)

The value of base impedance = = 1.6 Ω

Base values of the inductance, = (8)

Base value of the capacitance, (9)

and
The allowed switching frequency current amplitude at the converter side inductor
is = 0.1 p.u and at the grid side inductor .

The resonance frequency is selected to be 0.5 .

The amplitude of the voltage components at the switching frequency are assumed to be as
0.2s p.u, 0.056 p.u = 0.056 × 10.2mH = 0.5712 mH and d= 0.5 and
The ratio between the filter inductors is r = -1/[Link] reason for negative value of r is that the
resonance frequency and switching frequency current amplitude have not got a common
solution. The problem is solved by increasing the attenuation or decreasing .Decreasing
switching frequency is not reasonable and the attenuation is increased. That is ‘d’ is
decreased to 0.3, resulting in r =0.11. = 0.006 × 10.2 mH = 0.0612 mH
C = 0.56 p.u = 0.56 × 995 µF = 557.2 µF are obtained with inductor ratio r = 0.11. The
reason for high value of capacitor is that when using a low switching frequency the
resonance frequency requirement is fulfilled by choosing a large capacitor. To obtain smaller
capacitance values a smaller is to be allowed which causes to increase. Hence the

LCL values of the designed filter are , 0.0612 mH and C =557.2

µF.
MODELING AND SIMULATION OF VOLTAGE SOURCE… 211

CONCLUSIONS
The simulation study reveals that in SVM controlled VSI the fundamental peak of line and
phase voltage in SVM based VSI is higher compared to SPWM controlled VSI. It is also
observed that utilization of dc bus voltage is higher in SVM controlled VSI than SPWM
controlled VSI. It is also observed that SVM controlled two-level VSI generates less
harmonic distortion in output phase voltage, line voltage and currents compared with SPWM
technique. Output LC filter requirement increases in SPWM controlled VSI compared to
SVM controlled VSI for the same switching frequency and modulation index Voltage THD
in SVM is less compared to SPWM technique based VSI for resistive star connected load. It
is also observed that in SVM controlled VSI, LC filter requirement decreased by 30%
compared to SPWM controlled VSI for a resistive load of 105 kW. The output LCL filter
designed for SVM controlled VSI for switching frequency of 1800 Hz and 0.8 modulation
index the maximum load that inverter can handle is 230kW with voltage THD less than 5%
and phase and line voltage tolerance limits. The designed LCL filter component values are
lower compared to LC filter values and hence can reduce the cost of output filter with better
attenuation. It is also observed from the harmonic spectrum of output voltage that output LC
and LCL filter is able to reduce the amplitude of harmonics other than fundamental with
increase in fundamental component.

REFERENCES

[1] Willi Van Der Broeck, Hans Christoph Skudelny and Georg Viktor Stanke “Analysis and
realization of a pulse width modulator based on voltage space vectors” IEEE Trans. Ind. Appl.
Vol. 24, No. 1, Jan / Feb 1988.
[2] Jose R. Espinoza, Geza Joos and Hua Jin, “Modeling and implementation of space Vector PWM
techniques in active filter applications” IEEE trans. Ind. Appl. Vol.12, No. 6, Jan / Feb 1999.
[3] Joachim Holtz, “Pulse Width Modulation-A Survey” IEEE trans. on Industrial Electronics, Vol.
39, No. 5, December 1992
[4] Application notes Texas Instruments, “Space vector control of three leg PWM inverters”.
[5] Rathnakumar, Lakshmana Perumal.J and Srinivasan T. “A new software implementation of
space vector PWM” in IEEE Proc. Southeast Conf., pp.131-136, April 2005
[6] Ye Yang, Kazerani M. and Quintana, V.H, “Modeling, control and implementation of three
phase PWM converters” in IEEE Transactions on Power Electronics Vol. 18, No. 3, pp. 857-
864, May 2003
212 HARSHAVARDHAN V. KULKARNI

[7] Atif Iqbal, Adoum Lamine, Imtiaz Ashraf and Mohibullah, “Space Vector PWM for Three-
Phase Voltage Source Inverter” in IEEE Proc. of UPEC ‘06, vol. 3, Pp.1096-1100, 6-8
Sept.2006
[8] Boys J.T. and Handley P.G. “Harmonic analysis of space vector modulated PWM Waveforms”
IEE Proc, Vol. 137, Pt. B, No. 4, July 1990.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 213-225

BIDIRECTIONAL DC-DC CONVERTER FED DRIVE

FOR ELECTRIC VEHICLE SYSTEM

PAVITRA KUMARI 1 AND SHERYL GRACE COLACO2

1
[Link], Dept. of E&EE, St. Joseph Engineering college, Mangalore, India .
2
Associate professor, Dept. of E&EE, St. Joseph Engineering College, Mangalore, India.

Abstract
This study is about the electric vehicle powered by bidirectional dc-dc converter. Batteries are the
primary energy-storage devices in Electric Vehicles (EVs). Now days Battery fed electric drives are
commonly being used for electric vehicles applications, due to various advantages, such as: nearly
zero emission, guaranteed load leveling, good transient operation and energy recovery during braking
operation. To fulfill these requirements converters with bidirectional power flow capabilities are
required to connect the battery to the dc link of the motor drive system. Battery fed Electric Vehicles
(BFEVs) is required to function in three different modes namely: acceleration mode, normal (steady-
state) mode and braking (regenerative) mode. During acceleration and normal modes the power flow
is from battery to motor where as during braking or regenerative mode the kinetic energy of the motor
is converted into electrical energy and fed back to battery. In the present work closed loop operation
of bi-directional dc-dc converter feeding a dc motor and its energy recovery due to regenerative
braking has been demonstrated.

---------------------------------
Keywords : Electric Vehicle(EVs), Battery fed Electric Vehicles (BFEVs),Hybrid Electric Vehicle
(HEVs), Pulse Width Modulated (PWM),Proportional-Integral(PI), Energy Storage System(ESS),
Nickel-Metal hydride (NiMH)
© http: //[Link]
214 PAVITRA KUMARI AND SHERYL GRACE COLACO

1. INTRODUCTION
As technology advances daily, rapidly changing technologies are causing environmental
damage. Living quality, life, and human health are repeatedly threatened by crises such as
housing circumstances influence, severe environmental damage of the Earth, rupture of the
Ozone Layer and the Greenhouse effect. Environmental protection and environmental
pollution improvement are a growing concern in most industrialized countries. Recently bi-
directional dc-dc converters are widely researched and developed for various applications
such as battery charger-dischargers, electric vehicles and UPS systems. In case of the battery
fed electric vehicles (BFEVs), electric energy flows between motor and battery side. For
achieving zero emission, the vehicle can be powered only by batteries or other electrical
energy sources. Batteries have widely been adopted in ground vehicles due to their
characteristics in terms of high energy density, compact size, and reliability. This can be
applied in Hybrid Electric Vehicle (HEVs) with a battery as an energy storage element to
provide desired management of the power flows. However the high initial cost of BFEVs as
well as its short driving range has limited its [Link] use of a Bi-directional dc-dc converter
fed dc motor drive devoted to electric vehicles (EVs) application allows a suitable control of
both motoring and regenerative braking operations, and it can contribute to a significant
increase the drive system overall efficiency. This paper deals with the use of a Bi-directional
dc-dc converter for a battery fed electric vehicle drive system. A closed loop speed control
technique of the proposed battery fed electric vehicle is designed and implemented using PI
controller. The overall drive system reduces the system complexity, cost and size of a purely
electric based vehicular system.

2. PROPOSED CONFIGURATION OF ELECTRIC VEHICLE SYSTEM

2.1 Block diagram of Proposed System
The proposed block diagram of electric vehicle integrated with energy storage system(ESS)
which drives motor drives is shown in fig .1
2.2 Circuit Structure and Operating Principles of Two-quadrant DC Motor Drive
A two-quadrant dc motor drive is comprised of two single-quadrant drivers. Fig.2 shows the
circuit structure of separately excited dc motor connected with two-quadrant dc motor drive,
and equivalent circuit.
BIDIRECTIONAL DC-DC CONVERTER FED DRIVE... 215

Figure 1 : Block diagram of a electric vehicle

Figure 2 : Equivalent circuit of a separately excited dc motor Connected with two quadrant
drive
Semiconductor switch Q1 and diode D1 form a buck converter. However, semiconductor
switch Q2 and diode D2 form a boost converter. A separately excited dc motor is driven by
the buck converter when the motor accelerates. Moreover, regenerative braking is generated
by the boost converter when the motor decelerates and delivers energy back to the battery.
Semiconductor switches Q1 and Q2 are alternately conducted by two PWM control signals.
The Q1 is conducted during TON period, and Q2 is conducted during TOFF period. To
avoid a short circuit in the power source, the circuit should prevent Q1 and Q2 from
operating simultaneously. To achieve the above purpose, a time delay circuit is connected
between control signal of Q1 and control signal of Q2. Namely, a small time delay in a
twinkling of control signal of VG1 alternating with control signal of VG2 is needed to
216 PAVITRA KUMARI AND SHERYL GRACE COLACO

ensure two control signals cannot occur simultaneously. Furthermore, only one of four
elements Q1, Q2, D1 and D2 can operate at one time.
The detailed working modes of circuit configuration are as follows:
Mode 1: 0≤t≤TON, Q1 conducts, VT=V, iA≥0 (Fig.3)

Figure 3 : Circuit with Q1 conducted

Motor armature is connected to power source via Q1 such that armature voltage VT equals
source voltage V when Q1 conducts. Because the source voltage V is greater than counter
electro-motive force EA, positive armature current iA flows through Q1. Therefore, time rate
of armature current iA is positive when the inductor begins to store energy.
Mode II : TON≤t≤T, Q1 cut off, D1 conducts, VT=0, iA≥0 (Fig.4)

Figure 4 : Circuit with D1 conducted

BIDIRECTIONAL DC-DC CONVERTER FED DRIVE... 217

When Q1 is cut off and D1 is conducted, in the same interval Q2 has a control signal.
However, because D1 is conducted such that it provides a reverse bias voltage on Q2, the
semiconductor switch Q2 cannot be conducted even though a control signal triggers Q2.
Motor armature is short-circuited such that armature voltage equals zero. As the inductor
begins to release energy, flow direction of armature current iA is unchanged, so the armature
current iA flowing through D1 is still positive, but time rate of armature current iA is
negative.
Mode III: TON ≤ t ≤ T D1 cut off, Q2 conducts, VT=0, iA≤0 (Fig.5)

Figure 5 : Circuit with Q2 conducted

The inductor continuously releases power energy until iA declines to zero such that the
inductor voltage is lower than the counter electro-motive force EA. The flow direction of
armature current iA then changes, D1 become cut off and Q2 become conducted, armature
current iA is negative, and time rate of armature current iA is negative.
Mode IV: 0≤t≤TON, Q2 cut off, D2 conducts, VT=V, iA≤0 (Fig.6)
When Q2 is cut off, in the same interval, Q1 has a control signal. However, counter electro-
motive force EA exceeds power source voltage V therefore, armature current iA
continuously flows in reverse, which causes D2 is conducted such that a reverse bias voltage
is applied to Q1. Semiconductor switch Q1 cannot be conducted even though a control signal
triggering Q1. The armature voltage VT then equals power source voltage V, and armature
current iA is negative.
218 PAVITRA KUMARI AND SHERYL GRACE COLACO

Figure 6 : Circuit with D2 conducted

Counter electro-motive force EA is lower than the source voltage V until the armature
current iA equals zero, then Q1 is conducted and come back to Mode I.

Figure 7 : Waveforms of two-quadrant drive

BIDIRECTIONAL DC-DC CONVERTER FED DRIVE... 219

3. CONTROL STRATEGY
To control the speed of the dc drive; one possible control option is to control the output
voltage of the bidirectional converter. To control the output voltage of the bidirectional
converter for driving the vehicle at desired speed and to provide fast response without
oscillations to rapid speed changes a PI controller is used and it shows satisfactory result. In
this control technique the motor speed ωm is sensed and compared with a reference speed
ωref. The error signal is processed through the PI controller. The signal thus obtained is
compared with a triangular signal equal to switching frequency to generate pulse width
modulated (PWM) control signals.

Figure 8 : Closed loop operation of the drive

The block diagram of feedback speed control system for DC motor drive is shown in Figure
8; the control objective is to make the motor speed follow the reference input speed change
by designing an appropriate controller. The proportional-integral(PI) controller is used to
reduce or eliminate the steady state error between the measured motor speed (ωmotor) and
the reference speed (ωref) to be tracked. The transfer function of PI controller is given by
𝐾𝑖
Gc(s) = Kp + 𝑆
(1)

Where, Kp and Ki are the proportional and integral gains. The value of Kp and Ki are
obtained by using Manual tuning method. Manual tuning is achieved by arranging the
parameters according to the system response. Until the desired system response is obtained
Ki, Kp and Kd are changed by observing system behavior. Example (for no system
oscillation): First lower the derivative and integral value to 0 and raise the proportional value
100. Then increase the integral value to 100 and slowly lower the integral value and observe
the system’s response. Since the system will be maintained around set point, change set point
220 PAVITRA KUMARI AND SHERYL GRACE COLACO

and verify if system corrects in an acceptable amount of time. If not acceptable or for a quick
response, continue lowering the integral value. If the system begins to oscillate again,
record the integral value and raise value to 100. After raising the integral value to 100, return
to the proportional value and raise this value until oscillation ceases. Finally, lower the
proportional value back to 100.0 and then lower the integral value slowly to a value that is
10% to 20% higher than the recorded value when oscillation started (recorded value times
1.1 or 1.2).
A proportional controller (Kp) will have the effect of reducing the rise time and will reduce,
but never eliminate, the steady-state error. An integral control (Ki) will have the effect of
eliminating the steady-state error, but it may make the transient response worse. Effects of
each of controllers Kp and Ki on a closed-loop system are summarized in the table shown in
Table.1
Table.1: Effects of Kp & Ki
Controller respond Rise Time Overshoot Settling Time Steady state error
Kp Decrease Increase Small change Decrease
Ki Decrease Increase Increase Eliminate

4. BATTERY REQUIREMENT FOR AUTOMOTIVE APPLICATION

Mainly Nickel-Metal hydride (NiMH) and Lithium-ion batteries are used in vehicular
application due to their characteristics in terms of high energy density, compact size and
reliability. The battery is being recharged by the regenerative capabilities of the electric
motors which are providing resistance during braking helping to slow down the vehicle. The
lithium-ion battery has been proven to have excellent performance in portable electronics
and medical devices .The lithium-ion battery has high energy density, has good high
temperature performance, and is recyclable. The promising aspects of the Li-ion batteries
include low memory effect, high specific power of 300 W/kg, high specific energy of 100
Wh/kg, and long battery life of 1000 cycles. These excellent characteristics give the lithium-
ion battery a high possibility of replacing NiMH as next-generation batteries for vehicles.

5. SIMULATION RESULTS
Performance of the dc motor drive with the above battery model and bidirectional converter
is simulated under different speed command. The simulations are carried out using
BIDIRECTIONAL DC-DC CONVERTER FED DRIVE... 221

MATLAB/SIMULINK. The inductor parasitic resistance and MOSFET turn-on resistance

are not considered in this case. For the test condition of the proposed drive topology the
following values of the different components of the converter are considered. A separately
excited DC motor model is used as load to the bidirectional dc-dc converter. The motor rated
at 5 hp, 240 V, and 1750 rpm. Principal parameters of the bidirectional converter are
switching frequency=1kHz, State of Charge= 95% , Battery voltage =200V ,Battery
capacity = 15 Ah.
Simulink Model of Drive system
A total of two cases of the drive system are studied:
Case I: steady state operation, the reference motor speed is 120rad/sec with a constant torque
demand of 10Nm.

Figure 8 : Simulink model of drive system

222 PAVITRA KUMARI AND SHERYL GRACE COLACO

Figure 9 : Waveforms of motor speed,motor armature Current,field current,motor torque

Figure 10 : Waveforms of Battery state of charge, Battery current, Battery voltage

BIDIRECTIONAL DC-DC CONVERTER FED DRIVE... 223

Case II. Transient State Results

During transient, simulations are performed when the motor speed is changed from
60rad/sec to 120rad/sec.

Figure 11 : Waveforms of motor speed, motor armature current, field current, motor torque
Simulations are also performed for the regenerative braking operation when the speed is
changed from 120 rad/sec to 0rad/sec.

Figure 12 : Waveforms of motor speed, motor armature current, field current, motor torque
224 PAVITRA KUMARI AND SHERYL GRACE COLACO

Figure 13 : Waveforms of motor power

Figure 14 : Waveforms of Battery state of charge

CONCLUSION
In this work we demonstrate the performance of a battery operated electric vehicle system and it
shows satisfactory performance at different driving condition. The proposed control technique with PI
controller find suitable for this electric drive. The performance of the BFEV is verified under forward
motoring mode, regenerative mode and when there is step change is speed command. The overall cost
and volume of the battery operated electric vehicle is less with the least number of components used
in the system
BIDIRECTIONAL DC-DC CONVERTER FED DRIVE... 225

REFERENCES

[1] Power Management of Fuel Cell and Battery fed DC Motor Drive for Electric Vehicle
Application, Premananda Pan1, R. K. Singh2, R.K.Tripathi3, Member, IEEE.
[2] Bidirectional DC/DC converter fed drive for electric vehicle system, International journal of
engineering, science and technology, vol.3, no.3, 2011
[3] Design and Implementation of a Regenerative Braking System for Electric Bicycles with a DSP
Controller, Chih-Chiang Hua, Shih-Jyun Kao, Yi-Hsiung Fang Department of Electrical
Engineering National Yunlin University of Science & Technology Douliou , Yunlin,
Taiwan,ROC
[4] A Full Hybrid Electric Scooter T.W. Ching, Senior Member, IEEE and K.U. Chan
[5] Design and Realization of a Bi-directional DC/DC Power Storage Converter’s Control
SystemYang Juan, Li Weiguo, Yan Tao, Hui Dong, Song Li, Zhang Chenyun, IEEE transactions
2012
[6] Energy Storage for Electric Vehicles, Juan Dixon, Senior Member IEEE.
[7] A Battery With Ultracapacitor HybridEnergy Storage System In Electric Vehicles Neenu.M, PG
Scholar, PSNACET, Dindigul, [Link] - International Conference On Advances In
Engineering, Science And Management, 2012
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 227-242

A BIDIRECTIONAL DC/DC CONVERTER FOR PARALLEL

MPPT IN PV STAND-ALONE APPLICATIONS

SHYNI R NAMBIAR1 AND AJITHANJAYA KUMAR M K2

1
PG Scholar, Dept. of E&EE, St. Joseph Engineering College, Mangalore, India.
2
Assistant professor, Dept. of E&EE, St. Joseph Engineering College, Mangalore, India.

Abstract
The rising cost of electricity and environmental concerns have driven research in the renewable
sources of energy especially solar energy. In this paper a bidirectional dc-dc converter is proposed for
parallel MPPT to increase the overall efficiency of the PV system compared to the series connected
dc-dc converter. For tracking the maximum power from PV module classical Perturb and Observe
(P&O) algorithm is used. Both MPPT and constant voltage control is used for maintaining the
constant voltage at the load side. Working is verified by modelling the PV panel with proposed DC-
DC converter in a closed loop with the help of Matlab/Simulink software.

----------------------------------
Keywords: PV module, bidirectional DC/DC converter, Maximum power point tracker
(MPPT),Perturb and observe (P&O) © http: //[Link]
228 SHYNI R NAMBIAR AND AJITHANJAYA KUMAR M K

1. INTRODUCTION
Among several renewable energy resources, energy harvesting from the photovoltaic (PV)
effect is the most essential and sustainable way because of abundance and easy accessibility
of solar radiant energy around the earth. In spite of the intermittency of sunlight, solar
energy is widely available during daylight and it is free to use. Recently, photovoltaic system
is recognized to be in the forefront in renewable electric power generation because it can
generate direct current electricity without heavy environmental impacts and contamination.
In order to extract maximum power from the PV module usually DC-DC converter is
connected in series between source and load. In this case total efficiency of PV system
depends upon this series converter. Because all power generated is processed by this series
converter only. To increase the overall efficiency of the system in this paper a Bidirectional
DC-DC converter is proposed which is connected in parallel configuration. So only a part of
power generated is processed by the converter which will reduces the negative impact of the
converter.
Perturb and observe (P&O) algorithm is used to track the maximum power from PV module
which is one of the same algorithm used in a series configuration also. In addition to this to
maintain the constant voltage at the load side when source is a battery than constant voltage
control is used.

2. PARALLEL MPPT CONFIGURATION FOR PV SYSTEM

2.1 Block diagram
The block diagram of parallel MPPT for PV system with proposed Bidirectional converter is
shown in Fig 1.

Figure 1 : Block diagram of PV module connected in parallel MPPT configuration.

A BIDIRECTIONAL DC/DC CONVERTER FOR... 229

The parallel connection of the MPPT is introduced in [1].Advantage of this configuration

compared to the series one is only a part of power generated is processed in a dc-dc
converter which increases the efficiency compared to series configuration. And at the same
time same MPPT algorithms are used both in series and parallel. For this parallel
configuration a Bidirectional DC-DC converter is proposed which having multiple functions:
battery charging in buck mode and battery discharging in boost, when load current equal to
PV module MPP current value (Impp), where the dc–dc converter does not process power.
Depending on the dc load voltage level, the dc loads can be connected in parallel with the
PV module (“DC Load-2”) and/or in parallel with the battery bank (“DC Load-1”).However,
the voltage level of the PV module is higher than the battery voltage. If the system must also
supply the ac loads, a dc–ac converter can be connected in parallel with the PV module. In
this case, with the help of output transformer it will generate the nominal ac voltage, or by
increasing the no of PV modules in a series connection. So in this configuration both
efficiency and reliability of the system increases significantly.

3. COMPONENTS OF PARALLEL CONFIGURATION

3.1 PV Module
Solar cells consist of a p-n junction fabricated in a thin wafer or layer of semiconductor. In
the dark, the I-V output characteristic of a solar cell has an exponential characteristic similar
to that of a diode. When exposed to light, photons with energy greater than the band gap
energy of the semiconductor are absorbed and create an electron-hole pair. These carriers are
swept apart under the influence of the internal electric fields of the p-n junction and create a
current proportional to the incident radiation.
3.1.1 Modelling the Solar Cell
Thus the simplest equivalent circuit of a solar cell is a current source in parallel with a diode.
The output of the current source is directly proportional to the light falling on the cell. The
diode determines the I-V characteristics of the cell. This model included irradiance and
temperature dependence of the photo-current IL and the saturation current of the diodeI0. A
series resistance Rs was included, but not a shunt resistance Rsh. A single shunt diode was
used with the diode quality factor set to achieve the best curve match. This model is a
simplified version of the two diode model presented in [2]. The circuit diagram for the solar
cell model is shown in Fig. 2.
230 SHYNI R NAMBIAR AND AJITHANJAYA KUMAR M K

Figure 2 : The circuit diagram of the solar cell

The equations which describe the I-V characteristics is
I=IL-IO[eq(V-IRS)/AKT --- 1]
Where I and V are solar cell output current and voltage respectively IO is dark saturation
current, q charge of electron, A is the diode quality factor, K Boltzmann constant, T absolute
temperature. For simulation purpose PV module of Kyocera Solar KC40T is utilized.
Table.1 gives the PV panel ratings measured under standard test condition (STC).i.e at
1000W/m2 and 25oC in [3].
Table 1 : Specifications of Kyocera solar KC40T
Maximum power (Pmax) 43W
Maximum power voltage (Vmpp) 17.4V
Maximum power current (Impp) 2.48A
Open circuit voltage (Voc) 21.7V
Short circuit current (Isc) 2.65A
Temperature coefficient of Voc -8.21*10-2V/oC
Temperature coefficient of Isc 1.06*10-3A/oC

The maximum power of the PV module at particular irradiation and temperature is only at
one value current (Impp) and voltage (Vmpp), which defines the maximum power point
(MPP) [4]. Fig.3 Shows I-V characteristics of Kyocera solar [Link] power is.
obtained at 17.4V and 2.48A at 1000W/m2 and 25oC.
A BIDIRECTIONAL DC/DC CONVERTER FOR... 231

.
Figure 3 : I-V characteristics.
3.2 Proposed Bidirectional DC-DC converter
3.2.1 Circuit operation
It consists of a capacitor(C), an inductor (L), a battery bank to storage energy, and a power
semiconductor leg (S1-D1 and S2-D2) as shown in [Link] verify the operation
characteristics of proposed circuit only dc load-2 is considered. The proposed bidirectional
converter operates as a buck converter to charge the battery and as a boost converter when
load energy demand is higher than the energy generated or when power from PV module is
zero.

Figure 4 : Bidirectional DC-DC converter

3.2.2 Buck mode of operation
For correct operation battery voltage must be lower than the PV module voltage. When the
energy generated is excess to supply the load, the exceeding energy is used to charge the
battery. When the power switch S1 is turned on, the inductor L stores energy, and the energy
flows from the PV module to the battery bank. When the power switch S1 is turned off, the
232 SHYNI R NAMBIAR AND AJITHANJAYA KUMAR M K

diode D2 conducts, and the energy stored in the inductor L is transferred to the battery. The
switches’ command signals are complementary, and therefore, the switch S2 is turned on
during the conduction of the diode D2.
3.2.3 Boost mode of operation
When energy generated at the MPP is not enough to supply the load the circuit operates as a
boost converter, transferring energy from the battery to the load. In this case, the switch S2 is
turned on, the inductor L stores energy from the battery. When the switch S2 is turned off,
the energy stored in the inductor is transferred to the load.
3.2.4 Operation modes
A bidirectional DC-DC converter for parallel MPPT having five operating modes. For
simplification PV module is represented by a current source (If).
a) Mode 1-Operation without Load (Ic = 0) as shown in Fig.5.

Fig.5: Operation mode 1.

When the load is not connected, the battery is charged with the PV module current (If = Ib,
and Ic =0). The MPPT algorithm ensures that the converter current (Ib) is equal to the PV
module MPP current (Impp) and sum of the battery and the capacitor voltages(Vb+Vc) equal
to the PV module MPP voltage(Vmpp).It is the buck mode of operation to charge the
battery.
b) Mode 2-Operation with load current that is lower than the MPP current(Ic < Impp) as
shown in Fig. 6.
A BIDIRECTIONAL DC/DC CONVERTER FOR... 233

Figure 6 : Operation mode 2.

If the load current value (Ic) is lower than the PV module MPP current (Impp), part of the
energy generated by the PV module is used to supply the load, and another part is used to
charge the battery. The control algorithm maintains the sum of the load current and the
converter current equal to the MPP current (If = Impp = Ic + Ib). This mode also operates as
a buck converter to charge the battery.
c) Mode 3-Operation with load current that is equal to the MPP current (Ic=Impp) as shown
in Fig. 7.

Figure 7 : Operation mode 3.

When the load current value (Ic) is equal to the PV module MPP current (Impp), the power
processed by the dc–dc converter is zero. In this case, the efficiency of the power system can
be considered close to 100% because the maximum power of the PV module is transferred to
the load without any power processing by the dc–dc converter.
234 SHYNI R NAMBIAR AND AJITHANJAYA KUMAR M K

d) Mode 4-Operation with a load current that is higher than the MPP current (Ic > Impp) as
shown in Fig. 8. If the load current value (Ic) is higher than the PV module MPP current
(Impp), all energy generated by the PV module is supplied to the load. The additional power
that is necessary to complement the load power must be supplied by the dc–dc converter.
The control algorithm maintains the PV module current at the MPP (If = Impp = Ic − Ib).
The battery is discharged during this operation mode, and the dc–dc converter operates as a
boost converter.

Figure 8 : Operation mode 4.

e) Mode 5-Operation without solar irradiation (If = 0) as shown in Fig. 9.

Figure 9 : Operation mode 5.

The dc–dc converter supplies the load during operation without solar irradiation or under
shading. In this case, the MPPT algorithm is disabled, and a digital voltage control loop
A BIDIRECTIONAL DC/DC CONVERTER FOR... 235

regulates the load voltage. The battery is discharged during this operation mode, and the dc–
dc converter operates as a boost converter, considering the battery as the input source. The
control system also verifies the battery discharge state and must turn off the dc–dc converter
when the battery energy level reaches its lowest limit.
3.3 Control Algorithms
The control strategy of the proposed system is composed of different algorithms, for modes
1, 2, and 4 MPPT control and the output voltage regulation in mode 5.
3.3.1 MPPT Algorithm
PV voltage and current are monitored, and the PV power can be controlled as in the series
connected system. The transition between the different operation modes must occur without
discontinuity of the load in the parallel-connected MPPT system. The convergence time of
the MPPT algorithm is, normally, not important for the series connection but it is important
in parallel connection in order to minimize the transient in the output voltage with the load
variation. Perturb and observe method was utilized [5]. In Fig.10. sample of the PV module
voltage Vf(n) and current If(n) are used to calculate the instantaneous power Pf(n). The
power obtained in the last iteration Pf(n) is compared with the power calculated in the
previous iteration Pf(n−1). In addition, the value of the duty cycle at the last iteration D(n) is
compared with the previous duty cycle D(n−1) in order to ascertain whether the converter
duty cycle must be increased or decreased in order to reach the MPP.
3.3.2 Voltage Control Loop
The voltage control loop is also activated in operation mode 5 when the energy from the PV
module is very low or null. In this case, the MPPT algorithm is disabled, the dc–dc power
circuit operates as a boost converter, and the battery bank supplies the load with a constant
voltage. In this situation, the reference voltage defined in the control algorithm is compared
with the output voltage, and the error signal is applied to the digital voltage compensator.
The control action defines the converter duty cycle in order to regulate the output voltage.
The voltage reference considered is equal to 15 V. As occurs with the MPPT algorithm
during the voltage control loop operation, the battery state of charge is verified, and the
system is turned off when the battery reaches its lowest.
236 SHYNI R NAMBIAR AND AJITHANJAYA KUMAR M K

Figure 10 : Control algorithm for parallel MPPT

3.3.3 Design Parameters

The MPP voltage (Vmpp) of the PV module utilized in this paper is between approximately
12.5 and 17.5 V . The battery nominal voltage is equal to 6 V. The continuous conduction
mode operation is considered for determination of components shown in a circuit in Fig. 4.
The components and semiconductors of the parallel MPPT system must be designed,
considering the operation condition under the highest current level, which occurs in the boost
operation mode without solar irradiation (mode 5). Switching frequency = 30 kHz.
Maximum power of the PV module = 45 W, and this value was considered in the design
[Link] Inductor
The average inductor current in the boost operation is defined by (1), considering an
efficiency equal to η = 90%
A BIDIRECTIONAL DC/DC CONVERTER FOR... 237

IL = P/(Vb * η) =45/(6 * 0.9) = 8.333 A (1)

A current ripple that is equal to 20% was considered in the inductance determination
ΔI = 0.2 * IL = 0.2 * 8.333 = 1.667 A.
The operation point defined is a duty cycle that is equal to D = 0.67, which is in order to
obtain a load voltage that is equal to 18 V, with a battery voltage equal to 6 V. The
inductance is calculated by

An inductance value that is equal to 90 μH is utilized for simulation.

[Link] Capacitor
A capacitor voltage ripple that is equal to 5% was adopted in the design of the filter
capacitor
ΔVC = Vo * 0.05 = 18 * 0.05 = 0.9 V.
The capacitance is calculated by
C = ( Icap * D)/(f * ΔVC).
The capacitor current is defined by
Icap = Ic +ΔI/2
Icap = (45/18)+ (1.667/2) = 3.333 A.
Thus
C= (3.333*0.667)/(30*1000*0.9) = 82.337 μF.
A capacitor of100 μF/50 V is utilized for simulation.

4. SIMULATION RESULTS
Simulation is carried for different irradiation from 1000W/m2 to 100W/m2 at constant
temperature 25oC to verify different modes of operation. Perturb and observe method is used
for the MPPT algorithm and for mode 5 operation under the closed loop constant output
voltage control is used.
1) Simulation of parallel configuration at irradiation 1000W/M2, and temperature 250c of
PV module maximum power track by the circuit is 42.5 W at duty cycle 0.45 and average
output voltage [Link] in Fig 11a,b, Waveforms of different parameters.
238 SHYNI R NAMBIAR AND AJITHANJAYA KUMAR M K

Figure 11.a : Simulation ckt of parallel configuration at 1000w/m2.

Figure 11.b : PV Power and Dutycycle.

A BIDIRECTIONAL DC/DC CONVERTER FOR... 239

Figure 11.c : O/P current and O/P voltage

Figure 11.d : PV voltage,current,power,inductor current

In this inductor current and battery power is positive so it is operating in buck mode i.e.
charging the battery(mode 2).
240 SHYNI R NAMBIAR AND AJITHANJAYA KUMAR M K

Figure 11.e : Battery power,voltage,current.

1) At different irradiation 1000w/m2,800w/m2,600w/m2 and 100w/m2.

Figure 12.a : PV Power and Dutycycle

At 1000W/m2 max power tracked is 42.5.800w/m2 max power tracked is 38W.600w/m2

max power tracked is 25 W.100w/m2 PV power is less than 10 W so it running in mode 5
(i.e. constant output voltage control).
A BIDIRECTIONAL DC/DC CONVERTER FOR... 241

Figure 12.b : O/P current and O/P voltage

Figure 12.c : PV voltage,current,power,inductor current

242 SHYNI R NAMBIAR AND AJITHANJAYA KUMAR M K

Figure 12.d : Battery power, voltage, current

Inductor current and battery power is changing from positive to negative which shows that as
the irradiation decreases operation of the bidirctional DC-DC converter changes from buck
mode to boost mode and battery charging to discharging.

CONCULSION
In this study the parallel MPPT configuration is verified by Matlab/Simulink. Perturb and
observe algorithm has been used to track maximum power in modes 1,2and 4 and constant
output voltage control is used in mode 5. The parallel configuration increases the overall
efficiency compared to series. There is study in progress to design closed loop simulation for
mode 3 and upgrade the results in simulation and also for hardware implementation.

REFERENCES
[1] Enslin J. H. R. and Snyman D. B., “Combined low-cost, high-efficient inverter, peak power
tracker and regulator for PV applications,” IEEE Trans. Power Electron., vol. 6, no. 1, pp. 73–
82, Jan. 1991.
[2] Gow J. A., Manning C. D. “Development of a photovoltaic array model for use in power
electronics simulation studies,” IEE Proceedings on Electric Power Applications, vol. 146, no. 2,
pp. 193-200, March 1999.
[3] Kyocera data sheets, [Link].
[4] Solodovnik E. V., Liu S., and R. A. Dougal, “Power controller design for maximum power
tracking in solar installations,” IEEE Trans. Power Electron., vol. 19, no. 5, pp. 1295–1304, Sep.
2004.
[5] Femia N., Petron G., Spagnuolo G., and Vitelli M., “Optimization of perturb and observe
maximum power point tracking method,” IEEE Trans. Power Electron., vol. 20, no. 4, pp. 963–
973, Jul. 2005.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 243-252

MODERN INTELLIGENT CONTROL SYSTEM

USING KALMAN FILTER

NISHA ANIE ALEXANDER1, PRADEEP KUMAR2,

VISHVANATH A. N.3 AND SANTHOSH MUTHE4
1
MTech Student, Microelectronics, NMAMIT, Nitte.
2
Assistant Professor, E&E Department, NMAMIT, Nitte.
3
Scientist ‘G’, GTRE, Bangalore, India.
4
Scientist ‘C’, GTRE, Bangalore, India.

Abstract
A jet engine is a reaction engine that discharges a fast moving jet which generates thrust by jet
propulsion in accordance with Newton's laws of motion. Design and development of a gas turbine
engine for military aircraft presents challenging and unique opportunities for control system evolution
and its development. Engine control design is crucial, as the propulsion system must reliably provide
optimum thrust at the given conditions in the shortest amount of time. Modern gas turbine engine
control systems are closed loop control systems that consist of all four types of control components:
controller, actuator, sensor and harness. Full authority digital Electronic Control system (FADECS)
is common in all modern jet engines. FADECS manipulates main fuel, variable geometries, exhaust
nozzle area, clearance control actuators in a coordinated manner to achieve multiple unmeasured
objectives like thrust and speed regulation and limit on temperatures, pressures, stall margin. In
modern aircrafts there is an increasing demand for reliability and safety. A fault in any component of
control system may lead to system instability or may result in unacceptable system performance
degradation. In the area of model based fault-tolerant control fault detection and isolation uses state or
output estimators. As part of this project an attempt is made for developing of engine parameter
models and fault detection and diagnostic algorithms using state space and Kalman filter techniques.

----------------------------------
Keywords : Kalman filter, linear optimal control, pole placement, thrust.
© http: //[Link]
244 NISHA ANIE ALEXANDER, PRADEEP KUMAR, VISHVANATH A. N. AND SANTHOSH MUTHE

1. INTRODUCTION
Turbofan is an airbreathing jet engine uses in aircraft propulsion. In turbofan four main
phases are carried out such as compression phase where pressure and temperature of fuel in
compressor increases, in heating phase temperature and volume increases at constant
pressure at combustor, expansion phase is adiabatic where mechanical work is extracted
from the air to power the compressor in turbine, further in expansion phase in nozzle where
speed is increases as it returns to inlet pressure. Hence entire action of turbofan or turbojet is
carried out using brayton cycle principle. Thrust is obtained by the difference in speed
between the outlet and inlet of turbojet. Thrust is mechanical force which moves an aircraft
through the air. It is an unmeasured value, Thrust is generated through the reaction of
accelerating a mass of gas obtained by Newton's third law of motion. F=m(V f-Vn). Where Vf
and Vn are inlet and outlet speed of turbojet.

2. INTELLIGENT CONTROL SYSTEM

Main aim of this paper is to estimate effective thrust produced by system. Control system is
an arrangement of different physical elements connected in such a manner so as to regulate,
direct or command itself or other system. Consider the state variable form of the system
X(1)=AX+BU; Y=CX+DU; Where A and B are system state and control input matrix, C and
D are output and feed forward matrices. System used is multi input and multi output system.
Generally thrust is an unmeasured value, it can be estimated by speed of the engine. In this
project considering high pressure and low pressure spool speed (NH and NL respectively
(rps)). From Newton’s second law of motion F=ma; where force F is thrust which is directly
proportional to accelerating mass of gas. Acceleration of gas is depends on velocity and
pressure, constant pressure at inlet and outlet of gas turbine leads to change in velocity of
the gas by applying brayton principal. The inlet of the system is fuel flow (Wf(kg/s)) and
nozzle area (A8(cm2)). In this project I have to obtain desired value of spool speed (NL and
NH) by maintaining A8 and Wf at specific range. Both NL and NH are independent
variables and NH is always greater than NL. Thrust can be estimated by using only NL value
if any error in NL value controller should automatically shift to NH value and reduce the
error in NL value.
Figure 1 represents the model based control and diagnostics concept. The actual output of
the system and desired output of the system must be equal with some acceptable control
MODERN INTELLIGENT CONTROL SYSTEM… 245

error. Usually engine model will produce desired correct output, but real engine will produce
some percent change in output. The difference between desired and actual output of the
system should be zero, if any error is detected then try to reduce this error. The error can be
reduced by making change in control laws.

FIG1: Model based control and diagnostics concept.

The main aim if model based control is to have the model reflect the actual condition of the
engine. Using fleet average model method predicts the engine models and detects the errors
and limits the capability of model to predict the unmeasured variables. Tracking filter is used
to estimate the model parameters, error can be estimated by using fault diagnostic algorithm.
This error can be reduced by using state and parameter estimation technique of adaptive
model parameter by KALMAN Filter. [1]

3. POLE PLACEMENT APPROACH

Entire engine model is nonlinear in nature which can be linearized by using piecewise
linearization algorithm. Each linear system is designed for different PLA values and all are
open loop system. First check the stability, controllability and observablity of each plant. If
system is stable in nature then obtain the frequency response so the system, if it is satisfied
then convert the system into closed loop system to estimate desired output value. It can be
done by pole placement approach where characteristic of plant will change by placing the
poles of the closed loop system to desired pole location. Sometimes system becomes
unstable while converting to closed loop system. In this case to maintain the system to stable
246 NISHA ANIE ALEXANDER, PRADEEP KUMAR, VISHVANATH A. N. AND SANTHOSH MUTHE

state we can adjust the proportional gain. The system can be reach desired value of output by
using PI controller.

FIG 2 : Closed loop system with gain

Pole placement can be carried out by using full state feedback method which provides the
sufficient control parameters to move all closed loop poles independently . some parameters
may be unmeasured in nature due to measurement noise, these parameteres can be estimated
by observing the output of the system and then reconstruct the system using compensator
shown in Figure 3. Observer can estimate the variables using reduced order observer.[4]

FIG 3 : Observer of the system

4. KALMAN FILTER
Filter is a deterministic system to which random signals are input and output is a random
signal with desired statistical properties. Filter is designed in such a way that it should
achieve desired set of performance objectives such as cut off frequencies, desired attenuation
of signals above or below the cut off frequency etc., with the minimum possible distortion
and it is also observed that a greater attenuation of noise and greater distortion of signal. A
MODERN INTELLIGENT CONTROL SYSTEM… 247

stochastic system such as a noisy system can be modeled by passing white noise through an
approximate linear system.
Consider a system X(1)=AX+BU+FV; Y=CX+DU+Z; V and Z are white noises called
process noise and measurement noise. The output of the stochastic system must be non-
stationary white noises. We cannot rely on full state feedback in the case of stochastic
system because we cannot predict the state vector. Hence observer is required based on
desired output value. Kalman filter is an optimal observer, which minimizes a statistical
measure of the estimation error (e=x-x0). Best estimation of state vector implies error should
be zero and reduce the conditional covariance matrix R . Kalman filter observer matrix is
X(0)=AX0+BU+L(Y-CX0-DU), output matrix is Y=CX+DU+GV+Z. Kalman filter results
naturally when the plant is time invariant and noise signals are stationary white noises.
Kalman filter is used because the system and system parameters are nonlinear in nature. In
MATLAB Kalman filter can be solved by using lqe/lqe2.m command.[4]

5. EXPERIMENTS AND RESULTS

Each linear system in the form of state variable is analyzed. If the system is stable,
controllable, observable in nature then obtained then obtain frequency response (peak time,
overshoot, undershoot, rise time, settling time etc…) of the system.[6] When passing input
A8 and Wf to the open loop system, we can obatain the output NL and NH shown in graph1,
where we can observe that NH value is always higher than NL value.
248 NISHA ANIE ALEXANDER, PRADEEP KUMAR, VISHVANATH A. N. AND SANTHOSH MUTHE

Make the open loop system to closed loop by using pole placement to obtain desired output
value. Where poles of the closed loop system will change to desired output value. This can
be carried out by using place.m file. Here we can’t use acker.m because system used is multi
input and multi output system. In place.m we need to have the pre knowledge of desired pole
location. When this proportional gain is directly send to engine model we can obtain the
desired output shown in graph2.

The overabundance of design parameter for multi input and multi output system and
additional condition requirement in traditional pole placement method (place.m, acker.m
etc...) and determining observer gain matrix are becomes a drawback. Hence linear optimal
control provides an alternative design strategy which directly formulates the performance
objectives of control system. For a linear plant the optimal control problem posed as finding
optimal regulator gain matrix K. in this process the value of gain matrix K can be obtained
directly by using MATLAB command lqr/lqry.m. The output of the system when
proportional gain K is considered as controller is shown in graph3.
MODERN INTELLIGENT CONTROL SYSTEM… 249

Consider the desired output of the plant is step signal with amplitude one. To reach the
desired output we can use PI controller instead of only proportion controller. Here we need
to change the gain matrices if system becomes unstable. To maintain NL to desired output
range first make all the control parameters to zero other than control parameter of NL with
respect to fuel flow then change control parameter of NL with respect to nozzle area. The
output of system when considering only NL by maintaining NH as independent variable as
shown in graph4.
250 NISHA ANIE ALEXANDER, PRADEEP KUMAR, VISHVANATH A. N. AND SANTHOSH MUTHE

Same procedure is carried out for obtain desired value of NH by adjusting gain matrices of
NH with respect to Wf and A8. And maintaining NL as independent variable as shown in
graph5.

Output of the system when both NL and NH are maintain in desired value by adjusting gain
matrices as shown in graph6.
MODERN INTELLIGENT CONTROL SYSTEM… 251

Consider the case when thrust value is estimated only by using NL value. If any
measurement error occurred with respect to NL value then controller automatically shifts to
NH value. Consider at the begging we there is no error in the system parameters so NL value
is considering by keeping NH as independent variable, suddenly error occurred at time 25
sec then systems is shifting to NH value by keeping NL as independent variable as shown in
graph7. This condition is occurred due to the presence of measurement error. While shifting
NL to NH the value we should consider both input A8 and Wf should be in acceptable range.

Due to the presence of process noise and measurement noise some system parameters are
unmeasured variable so we need to use observer. Kalman filter is an optimal observer use to
reduce estimation error. In matlab Kalman gains can be obtained by using lqe/lqe.m file.
Here Kalman filter is used because system and system parameters are nonlinear in nature.

CONCLUSION
Reduce error using Kalman filter. Compare output of Kalman filter with SISO system.
Improved fuel efficiency, increased durability and life, decreased life cycle costs, and
improved operations.
252 NISHA ANIE ALEXANDER, PRADEEP KUMAR, VISHVANATH A. N. AND SANTHOSH MUTHE

ACKNOWLEDGMENT
Author is like thank scientist A.N. Viswanath and Santhosh Muthe for providing their
thoughts and lab facilities. Also like to thank Asst. Prof. Pradeep Kumar and Prof. Vasudev
Shettigar.

REFERENCES

[1] TR-AVT-128 ([Link]), Chapter 3 – “INTELLIGENT CONTROL AND HEALTH

MONITORING” by Garg Sanjay (NASA Glenn Research Center), Kumar Aditya and Mathews
H. Kirk (GE Global Research), Rosenfeld Taylor (GE Aviation), Rybarik Pavol (Vibro-Meter
SA), and Viassolo Daniel E. (GE Global Research).
[2] Goebel K., Eklund N., Brunell B., “Rapid Detection of Faults for Safety Critical Aircraft
Operation”, Paper #1360, 2004 IEEE Aerospace Conference, Big Sky, Montana, March 6-13,
2004.
[3] Maybeck P. and Hanlon P., “Interrelationship of Single-Filter and Multiple-Model Adaptive
Algorithms”, IEEE Transactions on Aerospace and Electronic Systems, Vol. 34, No. 3, July
1998.
[4] “Modern control design with matlab and simulink” by Ashish Tewari.
[5] “Control system” by [Link] and [Link].
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 253-268

DSP BASED OPEN LOOP SPEED CONTROL OF THREE PHASE

INDUCTION MOTOR USING SPACE VECTOR MODULATED
VOLTAGE SOURCE INVERTER

AKSHAYA KATKAR, VEERESH BYAHATTI AND A. B. RAJU

Department of Electrical and Electronics Engineering,
B.V.B. College of Engineering and Technology,
Hubli 580031, Karnataka, India

Abstract
This paper describes developing mathematical modeling and simulation of Space Vector Modulation
[SVM] of 3 phase 2 level Voltage Source Inverter[VSI] using Matlab/ Simulink environment, and
verifying the same experimentally using TI’s 320LF2407A DSP. Simulation results obtained using
SVM are compared with that of Sinusoidal Pulse Width Modulation[SPWM] results to show that for
the same switching frequency, load and modulation index SVM results in higher output voltage and
reduced harmonic. Open loop speed control of 3 phase induction motor using space vector modulated
voltage source inverter with constant v/f ratio is implemented and experimentally verified using TI’s
320LF2407A DSP and the experimentally obtained results are presented.

---------------------------------
Keywords: Space Vector Modulation, Sine Pulse Width Modulation, Digital Signal Processor,
Voltage source inverter and speed control. © http: //[Link]
254 AKSHAYA KATKAR, VEERESH BYAHATTI AND A. B. RAJU

1. INTRODUCTION
Inverter are used in AC motor drives and uninterruptible AC power supplies where the main
objective is to produce a sinusoidal AC output Voltage whose magnitude and frequency can
be controlled so that of speed of induction motor can be controlled. The input to switch
mode inverter is assumed to be a constant DC voltage source which can be supplied from
rectifiers. The Fig. 1 shows the basic configuration of voltage source inverter. 3 phase
voltage source inverter [1] can be classified into square wave inverter and PWM inverter.
PWM inverter can be further classified into SPWM inverter and SVM inverter. PWM
inverter will be discussed in this paper. In PWM inverter switches are pulse width modulated
in order to shape the AC output voltage to be close to a sine wave as possible.

Figure 1 : Basic configuration of voltage source inverter

2. SINUSOIDAL PULSE WIDTH MODULATION

In Sinusoidal Pulse Width Modulation the width of each pulse is varied in proportion to the
amplitude of a sine wave.
In this technique the gating signals are generated by comparing a sinusoidal reference signal
with a high frequency triangular wave also called as carrier wave [2]. It is as shown in Fig. 2.
The gating signals are generated as follows
 when Vcontrol1 > Vtri S1 is on and S2 is off
 when Vcontrol1 < Vtri S1 is off and S2 is on
DSP BASED OPEN LOOP SPEED CONTROL OF THREE PHASE… 255

3. SPACE VECTOR MODULATION

Voltage space vector is a compact way to represent all the three-phase voltages desired at
any instant by a single variable [1]. This switching-cycle-averaged space vector is
synthesized using space-vector PWM (SVPWM) by spending appropriate time at each
vector. SVPWM fully utilizes the available dc-bus voltage, and results in the AC output,
which can be approximately 15% higher than that possible by using the Sine-PWM
approach.

Figure 2 : SPWM gating pulse

4. SPACE VECTOR MODULATION THEORY

The main idea behind SVPWM is to divide the 2D-plane into six equal areas each of them is
called sector. As shown in Fig. 3 the plane contains 6 active vector and two null vectors.
When active vectors are applied to the power module Fig. 1 the output voltage of the power
module, will be greater than zero i.e. one of the switches S1, S2, S3 will not be off. The
other two vectors V0 and V7 are called inactive or null vectors, because all switches will be
off or on. These two vectors allocate in the center of the circle of Fig. 4. The desired output
assigned Vref will rotate with angular velocity equal to desired output AC voltage frequency.
Rotating Vref vector will scan all sectors with the time. For every sample time we can
determine the sector containing Vref and calculate the time period for each vector of the
determined sector.
256 AKSHAYA KATKAR, VEERESH BYAHATTI AND A. B. RAJU

Figure 3 : SVPWM 2D plot

5. SIMULINK MODEL OF SVM

Equations used to develop simulink model of SVM. Consider the desired three-phase output
voltages to be synthesized, as shown in Fig. 5. The waveforms can be divided into six
sectors, but initially we will consider only sector 1, shown shaded in Fig. 5. In sector 1
shown in Fig. 4, the voltage for phase-a is maximum, for phase-c is minimum, and for phase-
b is in the middle therefore the line voltage is as shown in Eq. 1.

Figure 4 : 3 phase voltage to be synthesized

It is possible to achieve VLL=Vd only if da=1 and dc=0 Therefore at any other instant one of
the ways a lower line to line voltage can be obtained by decreasing the phase a duty ratio da
below 1 by amount ∆d and by increasing phase c above0 by amount ∆d
da = 1 - ∆d (2)
DSP BASED OPEN LOOP SPEED CONTROL OF THREE PHASE… 257

dc = 1 + ∆d (3)
Eq. (1), (2) and (3) show that the line-line voltage will be reduced below Vd by an amount
equal to (2∆d)Vd. Adding Eq. 2 and Eq. 3
da + d c = 1 (4)
Now considering the pole output voltages which contain the common-mode voltages:
VaN = Vcom + Van = da*∆Vd (5)
VcN = Vcom + Vcn = dc*∆Vd (6)
Adding Eq. (5) and (6)
Vcom = (Vd/2) - (Van + Vbn)/2 (7)
Eq. 7 is valid only for sector 1. It can be generalized for all the sector, referring to Eq. 7 for
sector 1, Van is the maximum of the three voltages similarly Vcn is the minimum of the three
phase voltages. Therefore, the general expression that is valid in all sectors is,
Vcom(t)=Vd/2+(0.5)(max(Van,Vbn,Vcn)*min(Van,Vbn,Vcn) ) (8)
Vcom(t) = Vd=2 + Vk(t) (9)
The Eq. 9 shows that common mode voltage in SVPWM consists of Vd/2, as in SPWM plus
an extra voltage term Vk(t).
Then pole voltage are calculated as follows
Vao = (Vdc/2)*(2*S1-1) (10)
i.e if S1=1 then Vao=Vdc/2 else Vao=-Vdc/2 in similar manner the pole voltage of other two
phase are calculated.
Calculation of Phase voltage
Vno = (Vao + Vbo + Vco)/3 (11)
Van = Vao - Vno (12)
Vbn = Vbo - Vno (13)
Vcn = Vco - Vno (14)
Calculation of line voltage
Vab = Van - Vbn (15)
Vbc = Vbn - Vcn (16)
Vca = Vcn - Van (17)
Calculation of Direct axis and Quadrature axis component
Vd = Van (18)
Vq = (1/√3)*(Vbn- Vcn) (19)
258 AKSHAYA KATKAR, VEERESH BYAHATTI AND A. B. RAJU

6. SIMULINK MODEL OF SVM

Fig. 5 shows simulink model of Space Vector modulated inverter. The model consists of
subsystem that generates the Space Vector modulating signal using Eq. 8. This modulating
signal is compared with high frequency triangular wave to generate the Gating signals.
Further these gating pulses are used to calculate line and Phase voltage using Equations 10 to
17. Function block is used to realize the Equations and repeating sequence is used to
generate high frequency triangular wave.

7. SIMULINK AND EXPERIMENT RESULTS

Fig. 6 shows line voltage using SVM technique with Vdc = 100 volts at frequency of 50 Hz
and carrier frequency fs=1950 Hz. Fig. 8 shows the Harmonic spectrum of line voltage using
SVM technique where in the harmonic are shifted towards higher frequency. From Fig. 7 we
can see that the trajectory followed by reference vector is circular in nature hence the current
drawn by motor or load will be more sinusoidal in nature with reduced harmonic. Fig. 11
shows SVM pulse generated using PIC18F4520 [4] and Fig. 12 shows SVM pulse generated
for two different Phase. These pulses are generated by comparing lookup table values of
modulating signal and triangle wave. If modulating value is greater than triangular value
then a particular bit of port will be set high or else set to zero.

Figure 5 : Simulink model of SVM

DSP BASED OPEN LOOP SPEED CONTROL OF THREE PHASE… 259

Figure 6 : line voltage waveform

Figure 7 : Plot of Vd v/s Vq

8. COMPARISON OF HARMONICS WITH DIFFERENT MODULATION

TECHNIQUE
From table 1 we can see that square wave modulation gives highest fundamental voltage but
harmonics appear at low frequency due to which the current drawn by load or motor is non
sinusoidal. With SPWM choosing mf as an odd integer results in an odd symmetry as well as
a half-wave symmetry. Therefore, only odd harmonics are present and the even harmonics
260 AKSHAYA KATKAR, VEERESH BYAHATTI AND A. B. RAJU

disappear from the waveform. Moreover, those for the cosine series are zero the coefficients
of the sine series in the Fourier analysis are only finite; The harmonics in the inverter output
voltage waveform appear as side-bands, centered around the switching frequency and its
multiples, that is, around mf , 2mf , 3mf and so on. This general pattern holds true for all
values of ma in the range 0-1. With SVM the value of fundamental voltage is 187.2 which is
15% more then SPWM and also the value of harmonic voltage is reduced. Harmonics are
expressed as percentage of fundamental output voltage.
Table 1 : Comparison Of Harmonics

Harmonic Square wave SPWM SVM

Fundamental 218.24 169.91 187.82
5th 20.67% -- --
7th 12.91% -- --
11th 19.8% -- --
13th 5.26% -- --
17th -- -- --
mf±2(mf=39) -- 29.03% 16.57%
2mf±1 -- 18.10% 8.84%
3mf±2 -- 13.84% 1.7%

Figure 8 : Harmonic Spectrum of SVM

DSP BASED OPEN LOOP SPEED CONTROL OF THREE PHASE… 261

Figure 9 : Harmonic Spectrum of SPWM

9. FEATURES OF 320LF2407A DSP

The Texas Instruments TMS320LF2407 DSP Controller is a programmable digital controller
with a C2xx DSP central processing unit (CPU) as the core processor 40 million instructions
per second (MIPS) performance.
The LF2407 peripheral set includes:
Two Event Managers (A and B)
 General Purpose (GP) timers
 PWM generators for digital motor control
 Analog-to-digital converter
 Controller Area Network (CAN) interface
 Serial Peripheral Interface (SPI) synchronous serial port
 Serial Communications Interface (SCI) asynchronous serial port
 General-Purpose bi-directional digital I/O (GPIO) pins
 Watchdog Timer (time-out DSP reset device for system integrity) [5] [6].
262 AKSHAYA KATKAR, VEERESH BYAHATTI AND A. B. RAJU

10. EXPERIMENTAL RESULTS

The SVPWM algorithm is implemented on TMS320F2407A DSP based digital controller
and tested on induction motor drive on open loop powered from three phase inverter. Basic
block diagram of the system is shown in Fig. 10. The program is built and complied in the
system the generated hex file is dumped on to DSP, SVM pulse are generated on PWM ports
of DSP. Gating signal are generated with appropriate dead time in order to avoid shoot
through or cross conduction current through leg of inverter. The Fig. 11 shows picture of
complete hardware set-up along with load and motor setup. 3rd harmonic injected reference
signal observed on DAC port of DSP is shown in Fig. 12 and Fig. 13 shows SVPWM gating
signals observed on PWM channel of DSP. Waveforms for line voltage and no load current
of induction motor for ma=0.8 and freq=50 Hz are shown in Fig. 14 and Fig. 15. From
Fig.14 we can see that the current drawn by the motor is sinusoidal. Observations for
different frequency on no load are given in table II and we can see that as frequency is
varied, V/f ratio is maintained constant. DC generator coupled with induction motor was
loaded in steps with frequency held constant to test induction motor with load and test results
are shown in Fig. 16 and Fig. 17. The current drawn by the motor during loaded condition is
still sinusoid. The observation for freq=50 Hz with ma=0.8 are given in table III. Even with
load the ratio V/f is maintained constant but there is drop in speed as the load is varied.

11. DRIVE DETAILS

The details of induction motor and DC generator are given in table

Figure 10 : Basic block diagram of Experimental Set-up

DSP BASED OPEN LOOP SPEED CONTROL OF THREE PHASE… 263

Figure 11 : Experimental Set-up

Figure 12 : Modulating signal for ma=0.8 and f=50 hz

264 AKSHAYA KATKAR, VEERESH BYAHATTI AND A. B. RAJU

Figure 13 : SVM pulse

Table 2 : Observation For Ma=0.8

Freaquency (hz) speed(rpm) voltage(volts) V/f ratio
10 290 31 3.1
20 550 64 3.2
30 860 96 3.2
40 1160 127 3.1
50 1420 164 3.2
DSP BASED OPEN LOOP SPEED CONTROL OF THREE PHASE… 265

Figure 14 : No load current waveform

Figure 15 : No load voltage waveform

266 AKSHAYA KATKAR, VEERESH BYAHATTI AND A. B. RAJU

Figure 16 : Current waveform under load condition for Idc=0.5A

Figure 17 : Line voltage waveform under load condition for Idc=0.5A

DSP BASED OPEN LOOP SPEED CONTROL OF THREE PHASE… 267

Table 3 : Observation For Ma=0.8 F=50 HZ

freq(hz) speed(rpm) vtg(volts) V/f ratio Idc Vdc
50 1434 160 3.2 0.5 121
50 1416 160 3.2 1 114.5
50 1380 160 3.2 1.4 107.6
50 1374 160 3.2 1.7 101

Table 4 : Drive Details

Induction motor
Voltage 415/230
RPM 1380
KW 0.55
Hz 50
A 1.4/2.7
Efficiency 70
DC motor
KW .75
RPM 1500
Voltage 180
A 5.5
Field DC 220
Current .53

XII. CONCLUSIONS: From the harmonic spectrum of SPWM Fig. 9 and SVM Fig. 8 we
can see that for same switching frequency and load the harmonic content in SVM is less as
compared to SPWM and also results in higher value of fundamental component and the
magnitude of harmonic is reduced in case of SVM. Speed control of 3 phase induction motor
fed with space vector modulated voltage source inverter with constant V/f is experimentally
verified using DSP 320LF2407A on on load as well with load. From experimental results we
can see that as the frequency is varied the ratio V/f is maintained constant and this true for
motor on load. The current drawn by load is also sinusoid.
268 AKSHAYA KATKAR, VEERESH BYAHATTI AND A. B. RAJU

REFERENCES

[1] Mohan, Undeland, Riobbins, Power electronic converters, applications and design WILEY
STUDENT EDITION,
[2] K. Vinoth Kumar, Prawin Angel Michael, Joseph P. John and Dr. S. Suresh Kumar, ”Simulation
and Comparison of SPWM and SVPWM Control for three phase Inverter” IEEE Transaction on
Industry Applications, Vol 24. No, I. January /February 2001.
[3] V T Ranganathan ”Space vector pulse width modulation - A status review” Sadhana, Vol. 22,
Part 6, December 1997.
[4] Alexandru Savulescu, ”The analysis and the simulation of the svm generator used for the control
of the electric drives with asynchronous motors”, 6th international conference on
Electromechanical and Power system October 4-6 2007.
[5] TMS320F/C24x DSP Controllers Reference Guide CPU and Instruction Set Literature Number:
SPRU160C June 1999 Texas Instruments
[6] TMS320LF/LC240xA DSP Controller Reference Guide System and Peripheral Literature
Number: SPRU357 C Revised May 2006 Texas Instruments
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 269-281

STUDY ON CORONA LOSS

SOUMYA ASANGIMATH
Lecturer, Department of Electrical & Electronics Engineering
S G Balekundri Institute of Technology, Belgaum-590010, India.

Abstract
Energy crisis is one of the toughest issues nowadays. Though how much ever the technology is
growing with new inventions and methods of generating energy, but still it is not possible to bridge
the gap between the generation & demand. In order to meet the increased demands regarding energy
we need to concentrate on losses that occur during transmission. This paper presents the mathematical
calculations of corona loss for higher voltages and the comparison between the I2R loss, corona loss
during rainy season & corona loss during normal fair weather conditions. This paper also highlights
on the methods to reduce the corona loss.

----------------------------------
Keywords: Corona power loss, resistive loss, Critical disruptive voltage, Irregularity factor, Bundled
conductors, RTV coating, Hydrophobicity © http: //[Link]
270 SOUMYA ASANGIMATH

1. INTRODUCTION
Improved standard of living of the people has led them to use more electrical energy now a
days. As a result of this, per capita energy consumption of our country is also increasing day
by day. Since generating stations are very far away from the consumer end so there is a need
to transmit and distribute the generated energy. For transmitting the electrical energy it is
usual & reliable to prefer High voltage transmission, because of the reason that it has
numerous advantages such as, it requires less conductor material, improves line regulation,
increases the power handling capacity & decreases the I2R losses etc...But at the same time
corona loss increases with increase in transmission voltage.
This corona loss occurs in the unusual situation in the overhead transmission line when the
intensity of the electric field exceeds the dielectric strength of air, then around the conductor
is electric drilling, which causes increasing losses and increasing the apparent conductivity.
This phenomenon is called the corona. Therefore, corona, defined as a self-sustained electric
discharge in which the field intensified ionization is localized only over a portion of the
distance between the electrodes. When the voltage higher than the critical voltage is applied
between to parallel polished wires, the glow is quite even. After operation for a short time,
reddish beads or tufts form along the wire, while around the surface of the wire there is a
bluish white glow. The a.c. corona viewed through a stroboscope has the same appearance as
direct current corona. As corona phenomenon is initiated a hissing noise is heard and ozone
gas is formed which can be detected by its characteristic colour. Mathematical calculations
prove that the corona loss is greater than the I2R loss for higher voltages.
To calculate corona loss there are many methods available such as Peek’s empirical formula,
Peterson’s formula, Runge kutta method, Finite element method, Haar wavelet methods
etc..This paper makes use of Peek’s empirical formula. Since corona losses play a major and
dominant role for a high voltages and when these losses get added to the other losses, the
total loss is more as a result of this, the whole efficiency of the system reduces. So this paper
makes an attempt to discuss few of the methods to reduce the corona loss.
Corona loss is characterised by the following voltages i) Critical disruptive voltage ii) Visual
critical disruptive voltage iii) Corona power loss at ac voltage
1.1 Critical disruptive voltage
If one-phase transmission line shown in Fig.1.
STUDY ON CORONA LOSS 271

Let ‘r’ be the radius of each conductor and d the distance between the conductors such that d
>> r. In this single-phase transmission line, let q be the charge per unit length on one of the
conductors and hence – q on the other. If the operating voltage is V, the potential of
conductor A with respect to neutral plane N will be V/2 and that of conductor B will be -
V/2.
Neutral plane
r
+q x P -q

B
A d

Figure 1 : 1-ph transmission line

The electric field intensity at point ‘P’ situated at distance ‘x’ meters from the centre of first
conductor A,is
𝑞
𝐸=
2𝛱ℇ0 𝑥
The electric field intensity at point ‘P’ from the entre of conductor B having charge density –
q coulomb/m is
−𝑞
𝐸=
2𝛱ℇ0 (𝑑 − 𝑥)
Since both the fields are in the same direction, the total electricity field intensity at point’P’
is
𝑞 𝑑
𝐸𝑥 = 2𝛱ℇ (1)
0 𝑥(𝑑−𝑥)

If V’ is the voltage of one conductor with respect to the neutral plane, we can write
𝑟 𝑞 𝑑−𝑟
V’= -∫𝑑/2 𝐸𝑥 𝑑𝑥 = 𝑙𝑛 volt (2)
2𝛱ℇ0 𝑟
𝑞
Substituiting the value of 2𝛱ℇ in equation (1) from equation (2) we get
0

𝑉′ 𝑑
𝐸𝑥 =
𝑙𝑛[(𝑑 − 𝑟)/𝑟] 𝑥(𝑑 − 𝑥)
From fig 1 we know that d>>r, hence
V′ d
Ex = ln[(d/r)] x(d−x) (3)
272 SOUMYA ASANGIMATH

From above equation it is clear that the field intensity (gradient) increases as x decreases.
The gradient (gmax) is maximum at the surface of conductor (x=r) and this will be
𝑉′ 𝑑 𝑉′
𝐸𝑚𝑎𝑥 = 𝑔𝑚𝑎𝑥 = 𝑙𝑛[(𝑑/𝑟)] 𝑟(𝑑−𝑟) ≈ 𝑟 𝑙𝑛(𝑑/𝑟) (4)

The value of potential gradient at which ionization of air takes place under normal
temperature(250C) and pressure of (76 cm of Hg) and without any impurities is equal to 30
Kv(peak)/cm or 21.1 kv(rms)/cm and is denoted by ‘go ’ or ‘ 𝑔0 ’. Thus the voltage
required to ionize the air
𝑑
𝑉 ′ = 𝑟 𝑔0 ln 𝑟

The density correction factor ‘δ’ at ‘t0 C’ and at pressure of ‘b cm of Hg’ is given by the
following equation
3.92𝑏
δ=273+𝑡 (5)
𝑑
Thus 𝑉 ′ = 𝑟𝑔0 𝛿 𝑙𝑛 𝑟
(6)
Critical disruptive voltage is the voltage at which complete disruption of dielectric occurs. It
corresponds to the gradient at the surface of conductors equal to the breakdown strength of
air. This voltage can be calculated form equation (6).
But irregularity factor ‘m’ is taken into account which further results in reduction of
breakdown voltage. So now the critical disruptive voltage (phase to neutral) will be
𝑑
𝑉𝑐 (Critical disruptive
𝑖𝑛 𝐾𝑣
voltage
) = rgomδ ln (7)
𝑟

Irregularity factor ‘m’ is different for different surface of conductors & the values are given
in the below table
Table 1 : Irregularity factor for different configuration of conductor
Sl. no Practical situation Irregularity factor (m)
1. Smooth polished conductor 1.0
2. Rough & weathered wire 0.98-0.93
3. Seven stranded cable 0.87-0.83
4. Concentric layer cables 0.85-0.80
1.2 Visual critical disruptive voltage
At critical disruptive voltage corona occurs, but it is not visible because the charged ions in
air must receive some finite energy to cause further ionization by collision. When the voltage
further increases at some point, the corona becomes visible. This voltage is called visual
STUDY ON CORONA LOSS 273

critical disruptive voltage (Vv). The gradient for critical visual corona is denoted by’ gv ‘
Peek gave the following empirical formula for ‘ gv ‘ which is the function of radius of
conductor
0.3
𝑔𝑣= 𝑔0 𝛿 (1 + √(𝑟𝑑)) Kv/cm (8)

The visual critical disruptive voltage is given by

𝑑 0.3 𝑑
𝑉𝑣= r𝑔𝑣 mv ln 𝑟 =21.1𝑚𝑣 𝑟𝛿 (1 + √(𝑟𝑑)) ln 𝑟 Kv (rms) (9)

iii) Corona power loss at ac voltage

According to Peek corona power loss can be determined from
𝑓+25 𝑟
𝑃𝑐 = 241 ∗ 10−5 ( ) (√ ) (𝑉𝑝 − 𝑉𝑐 )2 (10)
𝛿 𝑑

Where 𝑃𝑐 = Corona power loss in Kw/ph/km ,

f=Frequency in Hz,
𝑉𝑝=Phase voltage in Kv =VL /√3
r= radius of conductor in cm,
d= spacing between the conductors in cm
δ= density correction factor
𝑉𝑐 =Critical disruptive voltage in Kv

2. DESCRIPTION OF THE PROBLEM (MATHEMATICAL FORMULATION,

SIMPLIFYING ASSUMPTIONS)
2.1 Calculation of corona loss
So as to calculate corona loss, a problem statement of 750 kv transmission line is taken in to
account. Where conductors are of 2 cm in diameter and are spaced equilaterally at a distance
of 10m, with conductors having smooth surface with irregularity fator of m=0.96,
barometric pressure of 73 cm of Hg & temperature of 20o C.
On calculating the values of ‘δ’ , ‘𝑉𝑝’ and ‘𝑉𝑐 ’ using equations (5) &(7) we get the following
values\
δ=0.976, 𝑉𝑝=433.012 Kv, 𝑉𝑐 = 136.56Kv
Once phase line voltage (𝑉𝑝 ) and critical disruptive voltage (𝑉𝑐 ) are known, and if 𝑽𝒑 > 𝑽𝒄
then corona loss exists. However the above mentioned condition is used only to check the
274 SOUMYA ASANGIMATH

existence of corona loss for low [Link] the problem statement is for high voltage, it
is sure that corona loss will exist and on calculation it is found that
Corona loss (Pc)=514.28 Kw/ph/km and
Total corona loss/km=1542.84 Kw/km
The calculated corona loss exists in fair weather condition. But during the rainy season
corona loss increases and is found by the following Peek’s empirical formula
𝑓+25 𝑟
PC= 241 ∗ 10−5 ( 𝛿
) (√𝑑) (𝑉𝑝 − 0.8𝑉𝑐 )2 (11)

Using this formula corona loss during rainy season is calculated and is found to be
Corona loss (PC)=613.00 Kw/ph/km &
Total corona loss/km=1839 Kw/km
2.2 Calculation of resistive loss
Once corona losses are calculated, resistive losses are to be found so as to compare both the
values. Three phase resistive losses are found by the formula 3I2R, where the value of
current ‘I’ & resistance (R) are required. The current value ‘I’ can be calculated using power
equation
P=√3 VI cos Ø (12)
2.3 Assumptions
1) Power factor is assumed to be Unity.
2) The range of Characteristic impedence ‘ZC ’ value is found to be 250-400Ω.But here the
in between value of 300Ω is chosen throughout the calculation i.e for all different voltage
[Link] current value ‘I’ can be found provided we know power handling capacity of 3Ø
,750 Kv transmission [Link] power handling capacity is again found by the formula
[𝑉𝑜𝑙𝑡𝑎𝑔𝑒 𝑟𝑎𝑡𝑖𝑛𝑔 𝑖𝑛 𝐾𝑉]2
P(MW)=𝐶ℎ𝑎𝑟𝑎𝑐𝑡𝑒𝑟𝑖𝑠𝑡𝑖𝑐 𝑖𝑚𝑝𝑒𝑑𝑒𝑛𝑐𝑒 (𝑍𝑐) (13)

Thus on calculating the power handling capacity for 3Ø, 750 Kv using the above formula we
get P=2000 [Link] Equation (12) current value is calculated & is found to be I=1539.6
Amperes and on referring the standard specification table for 750 Kv transmission,
resistance value which is one of the line parameters is found to be R=0.0136 Ω/km.
Using the above values, Resistive loss= 3I2R is calculated and is found to be P(Resistive
loss)=96.711 Kw/[Link] comparing both Corona loss & Resistive loss it is observed that
P(Corona loss)>P(Resistive loss)
STUDY ON CORONA LOSS 275

It is obvious that corona loss is greater than resistive loss because the analysis is done on 750
Kv transmission line, which is a high voltage transmission. Using the same approach corona
loss & resistive losses are found for different voltage levels such as 220 Kv, 400 Kv & 1200
Kv. Spacing between the conductor (d) and resistance (R)varies for different voltage levels
and is given in the table below
Table 2 : Clearance between lines of various voltage levels
Voltage level in (KV) Spacing between the conductor Resistance ‘R’ in
‘d’ in (meters) Ω/Km
220 4 0.068
400 6.1 0.031
750 10 0.0136
1200 12.3 0.0107

On calculation we get different values for power handling capacity ratings and current
ratings for different transmission voltage levels and are listed in the table below

Table 3. Power handling capacity of various voltage level

Voltage level in (KV) Power handling capacity in (Mw) Current ‘I’in (Amperes)
220 150 393.648
400 500 721.68
750 2000 1539.6
1200 4000 1924.5

4. RESULTS
Table 4: Results
Voltage Power Corona Corona loss Resistive % of corona loss % of
rating rating loss in fair during rainy loss(3 Fair Foul & (3 I2R)
in Kv in Mw weather in season I2R) in weather rainy loss
(KW/Km) (Kw/Km) (Kw/km) condition season
220 150 2.01 28.77 31.61 0.00134 0.0191 0.021
400 500 243.408 376.5 48.4 0.048 0.0753 0.0096
750 2000 1524.84 1839 96.711 0.076 0.09195 0.0048
276 SOUMYA ASANGIMATH

Voltage Power Corona Corona loss Resistive % of corona loss % of

rating rating loss in fair during rainy loss(3 Fair Foul & (3 I2R)
in Kv in Mw weather in season I2R) in weather rainy loss
(KW/Km) (Kw/Km) (Kw/km) condition season
1200 4000 4827.42 5331.81 118.88 0.12 0.1333 0.0029

Using the approach discussed in the case study, corona loss during fair weather condition,
foul weather condition &I2R loss are calculated for the different voltage levels and are
tabulated in the above table. These values are used in plotting the graph which is shown
below to indicate that the corona loss is higher during rainy & foul weather conditions as
compared to the normal weather condition

6000

5000

4000 Pc normal in Kw/km

Pc rain in Kw/km
3000
I^2 R loss in Kw/km
Y axis-Loss in Kw/Km
2000
X axis- Voltage in KV

1000

0
220 Kv 400 Kv 750 Kv 1200 Kv

Figure 1 : Comparison of corona losses &I2R loss

Percentage of corona loss & I2R losses are also calculated for all the voltage levels and it is
observed that corona loss increases for higher voltages and it is evident that percentage of
corona loss is higher during foul weather conditions. But I2R loss decreases with increase in
voltage level.
STUDY ON CORONA LOSS 277

0.14

0.12

% of corona loss in fair

0.1 weather
% of corona loss during
0.08 foul weather
% of I^2 R loss
0.06
X axis- Voltage in Kv
0.04
Y axis - loss in %

0.02

0
220 Kv 400 Kv 750 Kv 1200 Kv

Figure 2 : Corona losses & I2R loss in Percentage

In order to design corona free lines it is necessary to take into consideration following
factors:
1) Electrical
2) Atmospheric
3) Conductor
1) Electrical Factors:
a) Frequency and waveform of the supply: Corona loss is a function of frequency. For
that reason the higher the frequency of the supply voltage the higher is corona loss.
This means that corona loss at 60 Hz is greater than at 50 Hz. As a result direct
current (DC) corona loss is less than the alternate current (AC).
b) Line Voltage: Line voltage factor is significant for voltages higher than disruptive
voltage. Corona and line voltage are directly proportional.
c) Conductor electrical field:Conductor electrical field depends on the voltage and
conductor configuration i.e., vertical, horizontal, delta etc. In horizontal
configuration the middle conductor has a larger electrical field than the outsides
278 SOUMYA ASANGIMATH

ones. This means that the critical disruptive voltage is lower for the middle
conductor and therefore corona loss is larger.
2) Atmospheric Factors: Air density, humidity, wind, temperature and pressure have an
effect on the corona loss. In addition rain, snow, hail and dust can reduce the critical
disruptive voltage and hence increase the corona loss. Rain has more effect on the corona
loss than any other weather conditions. The most influential are temperature and
[Link] condition such as air density is directly proportional to the air strength
breakdown.
3) Conductor Factors: Several different conductor factors affectthe corona loss:
a) Radius or size of the conductor: The larger the size of the conductor (radius) the
larger the power lower loss. For a certain voltages the larger the conductor size, the
larger the critical disruptive voltage and therefore the smaller the power loss.

Ploss  (Vln  Vc )2
Vln = line-to-neutral (phase) operating voltage in kV
Vc = disruptive (inception) critical voltage kV (rms)
b) Spacing between conductors: The larger the spacing between conductors the smaller

r
the power loss. This can be observed from power loss approximation Ploss 
D
r = conductor radius
D = distance (spacing) between conductors
c) Number of conductors / Phases: In case of a single conductor per phase for higher
voltages there is a significant corona loss. In order to reduce corona loss two or more
conductors are bundled together. By bundling conductors the self-geometric mean
distance (GMD) and the critical disruptive voltage are greater than in case of a single
conductor per phase which leads to reducing corona loss.
d) Profile or shape of the conductor: Conductors can have different shapes or profiles.
The profile of the conductor (cylindrical, oval, flat, etc.,) affects the corona loss.
Cylindrical shape has better field uniformity than any other shape and hence less
corona loss.
STUDY ON CORONA LOSS 279

e) Surface conditions of the conductors: The disruptive voltage is higher for smooth
cylindrical conductors. Conductors with uneven surface have more deposit (dust,
dirt, grease, etc.,) which lowers the disruptive voltage and increases corona.
f) Clearance from ground: Electrical field is affected by the height of the conductor
from the ground. Corona loss is greater for smaller clearances.
g) Heating of the conductor by load current: Load currentcauses heating of the
conductor which accelerates the drying of the conductor surface after rain. This
helps to minimize the time of the wet conductor and indirectly reduces the corona
loss

5. METHODS TO MINIMIZE CORONA LOSS

a) Conductors of large diameter of conductors: If conductor radius is larger, surface field
intensity is less and hence corona losses are lower. For the same current carrying
capacity, an ACSR conductor has larger radius, therefore the transmission lines with
ACSR conductors have lower corona loss. Also, for bundled conductors lines effective
radius is larger and hence corona loss is less.
b) Bundled conductors: corona losses in the power transmission lines with one conductor
is solved with by using two or more than two conductors per phase or as they are
known as bundling of conductors. By bundling the conductors the self geometric mean
distance (GMD) of the conductors is increased thereby the critical disruptive voltage is
increased and hence corona loss is reduced.
c) RTV coating: RTV (Room Temperature Vulcanizing silicone) is a rubber which is
used as the anti-corona coating, which is spaying onto the surface of the wearing
conductor. RTV coating is a widely used coating for its good heat and cold resistant
and electric insulation property. The good hydrophobicity and migration of
hydrophobicity make it have the performance of pollution flashover characteristics,
which is the best plan in solving the problem of pollution flashover of the outer
insulation of the power equipment. How to put the coating on the surface of the
conductor is another problem, because the shape of the coating will greatly influence
the effect of anti-corona. There are various methods used for coating such as
quantitative brush method, pouring method, spray dyeing method, etc. the quantitative
brush method and pouring method could not make the surface of the conductor
280 SOUMYA ASANGIMATH

uniform, so spray dyeing method is widely adopted. Thus the tests prove that the
corona loss of the conductor with RTV coating is less than that with no coating
because of increase in coating thickness.
d) Increasing the conductor spacing: The increased spacing increases the voltage level at
which corona starts. Thus the corona effects are reduced. But practically spacing
cannot be increased beyond certain limits as the cost of the supporting structure
increases with the increased spacing.

6. ADVANTAGES OF CORONA
a) Due to corona, the air surrounding the conductor is ionized and becomes conducting.
This increases the virtual diameter of the conductor.
b) Corona reduces the effects produced by the surges and conductor is saved form
possibility of lightning. It acts as a safety device.

7. DISADVANTAGES OF CORONA
a) The corona power loss is the biggest disadvantage which reduces the transmission
efficiency.
b) The effective capacitance of the conductor is increased which increases the flow of
charging current.
c) Due to electromagnetic and electrostatic induction field corona interferes with the
communication lines which usually run along the same route as the power lines
d) The third harmonic components produced due to corona makes the current non
sinusoidal. This increases the corona loss.
e) The ozone gas formed due to corona chemically reacts with the conductor and can
cause corrosion.

CONCLUSION
Corona loss is higher than I2R loss for extra high voltage (EHV) and ultra high voltages
(UHV) and this corona loss is highest during rainy and foul weather conditions. This loss
cannot be eliminated but can be reduced to some extent by adopting the methods which are
mentioned above.
STUDY ON CORONA LOSS 281

REFERENCES

[1] Fang-Cheng Lü, Shao-Hua You etc, “AC Conductors’Corona-Loss Calculation and Analysis in
Corona Cage”, IEEE transactions on power delivery, VOL. 27, NO. 2, APRIL 2012
[2] Greg Clawson Mira Lopez, “Design and fault analysis of a 345kv 220 mile overhead
transmission line”, A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree
of Master of science to California State University, Sacramento, SPRING 2012
[3] Singh S.N., “Electric Power Generation, Transmission Distribution”,PHI learning Private
Limited,2nd edition, June 2010
[4] Uppal S.L. and Rao S. “Electrical power systems”,Khanna publishers, fifteenth edition,2009
[5] Begamudre Rakosh Das “Extra high voltage AC transmission engineering”, New age
interntional publishers, 4th edition,2011
[6] Zakariya M. Al-Hamouz, Member, IEEE “Corona Power Loss, Electric Field, and Current
Density Profiles in Bundle Horizontal and Vertical Bipolar Conductors”,IEEE transactions on
industry applications, vol. 38, no. 5, september/october 2002
[7] Zakariya Mahmoud Al-Hamouz, Member, IEEE “Corona Power Loss on Bundled Conductors:
Experimental and Computational Results”, IEEE transactions on industry applications, vol. 35,
no. 6, november/december 1999
[8] Dion Savio Antao, “A study of Direct Current Corona Discharges in Gases and Liquids for Thin
Film Deposition”, A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree
of Master of science to Drexel University, September 2009
[9] Gentile Thomas J, etc,“Merits of 765 Kv and insights intoits Application”,Quanta technology,
Electric transmission technology(ETA) America, September 15, 2009
[10] Visacro S. *, Silveira F.H., “Evaluation of current distribution along the lightning discharge
channel by a hybrid electromagnetic model”, Journal of Electrostatics 60 (2004) 111–120
[11] R. Allen bernstorf, principal engineer, insulators david ryan,polymer chemist,“Silicon
compounds for high voltage insulators:HPS testing of silicon compounds part 2of a 2 part series
“, by Hubbell Power Systems, Inc.
[12] Tirumala Rakshit, “Corona discharges in asymmetric electric fields and its impact on ionic wind
generation”, A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy to university of Notre Dame,march 2013
[13] krasniqi Isuf, komoni Vjollca, “Corona Losses Dependence from the Conductor Diameter”,
Recent Researches in Power Systems & Systems Science .
[14] Snigdha Sharma, etc, “Corona effects on EHV AC transmission lines” International Journal of
Scientific Research Engineering & Technology, volume 1,August 2012
[15] Ersoy A., Kuntman A., “An experimental study on corona losses in conductors”,Online Journal
on Electrical Engineering,volume 3
[16] Xu Ziqiang, Li Ren, “Research on the Anti-corona Coating of the Power Transmission Line
Conductor”, Energy and Power Engineering, April 2013, 5, 148-150
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 283-294

DESIGN OF SERVO MOTOR CONTROLLER

FOR BIPED ROBOT

ROOPA C1, N.MUNIYAPPA2 AND NITHIN AWASTHI3

1
Student, The Oxford College of Engineering, VTU Karnataka, India,
2
Assistant Professor, The Oxford College of Engineering, VTU Karnataka, India,
3
Manager, Technical, APSIS Solutions, Bangalore, India

Abstract
With the rapid advancement and development in the field of Electrical and Electronics Engineering, the
motor controlling in the field of Robotics has become most challenging task, so in this paper number of
servo motors being used to control and design of the Biped Robot. Hence Design of servo motor
controller for Biped Robot is proposed in this paper. This servo motor controller is used to control the
Biped Robot motion, which has servo motors HXT900 to accomplish the movement of Biped Robot
legs. DSPIC30F2010 being used to design the controller for servo motors. Biped robot walks like a
human by balancing the Centre of Mass.

----------------------------------
Keywords : Biped Robot; DSPIC30F2010; HXT900-9g Micro Servomotor
© http: //[Link]
284 ROOPA C, [Link] AND NITHIN AWASTHI

1. INTRODUCTION
With advances in science and technology, the interest to study the human walking has
developed the demand for building the Biped robots. The development of Biped walking robot
involves research in heterogeneous areas. This Paper describes the first attempt in building the
Bipedal walking robot. DC Servo Motors become an important device in a wide range of
industrial applications that require high dynamics on position control such as numerically
controlled machinery, robotics, automation and other mechanism where the starting and
stopping functions are quickly and accurately. These applications require a high-speed control
accuracy and good dynamic respond. In robotic applications, servo motors are used to move
the robotic arm to a relevant position by means of controllers in the automated manufacturing
lines of industries. The rotor construction of servo motors made of special material with less
weight to decrease the inertia of armature but capable to produce the necessary magnetic flux.
Low rotor inertia increases the capability of immediately starting and stopping during the
on-off conditions. The high cost of servo motor becomes a major issue. Therefore, the
small-scale manufacturers or users cannot afford to use this type of DC Motor. Hence
HXT900-9g Micro Servomotors are used to make Biped Robot. One more aspect of
implementing Biped Robot with servomotor is reducing the time delay between the
commands. By using data sheet of HXT900-9g micro servomotor frequency, duty cycle being
calculated. According to duty cycle PWM signals are generated and are given to the
servomotors to achieve the movement of the Biped Robot. These PWM signals are controlled
by DSPIC30F2010.

2. THE PROPOSED METHOD

In this paper the Biped Robot is constructed by several servomotors, HXT900, which are
controlled by servo motor controller. The block diagram for implementing Biped Robot is
shown in figure1 as below
DESIGN OF SERVO MOTOR CONTROLLER FOR BIPED ROBOT 285

Figure 1 : Block diagram for Biped Robot

2.1 Generation of PWM signal
Pulse Width Modulation (PWM) or Pulse Duration Modulation (PDM) signals are a two state
signal (square or digital) with a fixed frequency and whose duty cycle can [Link] simplicity
of such coding is widely used in reduced model. Receiver transmits servo-motor position
information through a PWM signal whose frequency is 50Hz (20ms period) and whose duty
cycle takes values between 1ms and 2ms. 1.5ms being the servo motor central [Link]
transmission is one of the easiest ways to have two microcontrollers exchanging variables.
Peripheral to generate and read PWM are very common on most microcontrollers. On the
DSPIC, two peripherals can generate PWM signals and three peripherals can read such
signals:
 PWM Motor and
 [[Block/Output_Compare|Output Compare (OC)] or Output Compare Hardware
(OC_HW) based on the same peripheral implemented with a slightly different way)
Three peripheral can read PWM signals:
 Input Capture (the most precise)
 External Interrupt
 Change Notification
A PWM signal requires only one wire (two if we take into account the ground wire). However,
the communication through one wire is one way only (not bi-directional). As many wires as
variable being sent are required. Despite the two state values of the signal, this is an analog
coding: the value of the variable is analog to the duty-cycle). Therefore, this analogue coding
could add noise to the reconstructed value on the receiver side. This noise is due to:
 Quantization of the duty cycle at the emitter side
286 ROOPA C, [Link] AND NITHIN AWASTHI

 Transition from one state to the other slow down due to wire capacity
 Quantization at the receiver side
The quantization at the emitter or receiver side depends on both the period of the PWM signal
and the speed of the microcontroller. How many ticks can the microcontroller count during
this period? The respective provide these values through variables present in the workspace
like PWMmax, OCxmax, ICxmax etc… These variables can be used as constant in the model.
There exist a trade of between the resolution and the bandwidth. The bandwidth of a PWM
transmission is typically not very high. (Transmissions are done at 50Hz in reduced model, but
can be increase to 500Hz) We note however that information transmitted through a PWM
signal is robust to electrical noise because it is based on a temporal coding. Received value
may be received with a delay.
The generated PWM signal for following specification is shown in figure2
Pulse cycle 20ms,
Pulse width 450-2450 µs
Frequency 50Hz

Figure 2 : PWM Output for 50Hz Frequency

The following formulas were used to calculate frequency for generating PWM signals:
PTPER-PWM Time based Period

PTPER= –1

Duty cycle =
DESIGN OF SERVO MOTOR CONTROLLER FOR BIPED ROBOT 287

2.2 Principle of operation of servomotor

Servos are controlled by a pulse of variable width. The sent signal of this input pulse is
characterized by a minimum pulse, maximum, and a repetition rate as seen in Figure 3. Given
the rotation constraints of the servo, neutral is defined to be the position where the servo has
exactly the same amount of potential rotation in a clockwise direction as it is in a counter
clockwise direction. The angle is determined by the duration of applied pulse to the signal
wire which is called PWM or Pulse Coded Modulation. The servo should detect a pulse every
20 ms. The length of the pulse will determine how far the motor turns. For example, a 1.5 ms
pulse will make the motor turn to a 90 degree position (neutral position). The position pulse
must be repeated to instruct the servo to stay in position.

Figure 3 : input pulse to servo-motor

When a pulse is sent to a servo that is less than 1.5 ms, the servo rotates to a position and holds
its output shaft some number of degrees counter-clockwise from the neutral point. When the
pulse is wider than 1.5 ms, the opposite operation is occurred. The minimal width and the
maximum width of pulse that will command the servo to turn to a valid position are functions
of each servo. Generally the minimum pulse will be about 1ms wide (some servo is 0.5 ms)
and the maximum pulse will be 2 ms wide (some servo is 2.5 ms). The servo motor operates in
the range of 5 % to 10 % of duty cycles. Figure4 shows relationship between pulse and
direction of servo motor.

Figure 4 : Relationship between pulse and direction of servo-motor

288 ROOPA C, [Link] AND NITHIN AWASTHI

The PWM is a commonly used technique for controlling power into electrical device. The
average value of voltage (and current) fed to the load is controlled by turning the switch
between supply and load, ON and OFF at a fast pace. The longer the switch is ON compared to
the OFF periods, the higher the power supplied to the load is. AC drives required the
sinusoidal signal and modulation generator. The triangular signal is the carrier or switching
frequency of the inverter. The modulation generator produces a sine wave signal that
determines the width of the pulses, and therefore the RMS voltage output of the inverter. For
DC drives, the PWM signal is generated by comparing a triangular wave signal with a DC
signal as shown in Figure5. The DC signal can be ranged between the minimum and maximum
voltages of the triangle wave. The PWM signal is measured using the percentage of duty cycle
where the pulse duration over the pulse period.

Figure 5 : Generation PWM for DC Drives

The mathematical model of DC servo motor can be simplified by mean of the circuit as shown
in Figure 6. The electrical part represented by armature and the mechanical part by T and J. As
the field excitation is constant, the armature controller only depends on armature voltage. The
mechanical equations describing this system can be written with the assumptions that the loss
is included in load torque and neglecting viscous friction constant as given below:

Figure 6 : DC servomotor equivalent circuit

Va = Rai + Ladi/dt + Eb
Jdω/dt = Te – T
DESIGN OF SERVO MOTOR CONTROLLER FOR BIPED ROBOT 289

With
Eb = Kω; Te = Ki
Where;
Va = Armature voltage, Ra = Armature resistance
I = Armature current, La = armature inductance
K = Torque and back electromagnetic constant (Nm.A-1)
ω = Rotor angular speed, Te = Electromagnetic Torque
T = Total load torque, J = Rotor inertia

3. SIMULATION OF SERVOMOTOR PARAMETERS SIMULATION CIRCUIT

Figure 7 : simulation circuit for servomotor

The simulation circuit consists of four major blocks
1. Pulse generator
2. Driver circuit
3. DC servomotor
4. Angular position sensor
According to motor specification PWM signals are generated with the frequency of 50Hz and
duty cycle of 2% to 13%. Driver circuit consists of H-Bridge configuration which provides
motor to rotate in forward and reverse direction. Then next block will be consisting of
servomotor, here motor itself is controlled in closed loop control system. To sense the motor
shaft position, angular position sensor being used. Angular position sensor will give three
outputs they are: step angle, velocity and acceleration. The step angle will be in radians to
convert it into degree, radians to degree converter block being used. The out for the simulation
is shown below:
290 ROOPA C, [Link] AND NITHIN AWASTHI

(i) Step angle

Figure 8 : Step Angle

(ii) Velocity

Figure 9 : Velocity
(iii) Acceleration:

Figure 10 : Acceleration

4. CONTROL OF BIPED ROBOT

Generally any robot has a combination of motors and sensors, which are controlled by
microcontrollers. There are wide varieties of motors, sensors and microcontrollers available.
In this project low cost DSPIC30F2010 and actuators are used. There are three D.O.F, each
D.O.F has one servomotor and it is controlled by DSPIC30F2010. The robot controller board
has been specifically designed for this project and it measures 90X70mm. The controller
DESIGN OF SERVO MOTOR CONTROLLER FOR BIPED ROBOT 291

board has the capability to control upto eight actuators and it has a provision for providing
sensory inputs to the controller. The robot has the capability to work in closed loop with the
help of sensory inputs. The robot is controlled and actuated using a pre-defined sequences and
it implements an open loop control and thus does not use sensors.
Algorithm:
All the three motors are controlled and actuated simultaneously while maintaining the
previous positional values. Initially, the first motor will be serviced with on-time pulse period
and during the off-time pulse period of the motor, second motor will be serviced with on-time
pulse period. This type of actuation is continued till all the three motors are serviced.
Positional values loaded in the Look-up table and are retrieved and pulses are sent to the
motors accordingly. It is shown in the figure below with various ON and OFF time periods.
No special algorithms are used for balancing the bipedal robot. Currently, the walking gait
was developed by studying possible walking movements using the prototype and by
simulating various walking gaits using the simulink. In the future we hope to add sensor-based
active balancing.

Figure11 : PWM pulses for Servomotors

Arrangement of Servomotors for Biped Robot:

Figure12 : Arrangement of servomotors

292 ROOPA C, [Link] AND NITHIN AWASTHI

To make Biped Robot, three servo motors being used. Two servomotors are used for knee of
the legs of Biped Robot. S1 for right leg and S2 for left leg knees respectively. S3 is used as
mass at the center of the Robot.

5. SCHEMATIC DIAGRAM
5.1 High Performance Modified RISC CPU
Modified Harvard architecture, C compiler optimized instruction set architecture, 84 base
instructions with flexible addressing modes, 24-bit wide instructions, 16-bit wide data path,
12 Kbytes on-chip Flash program space, 512 bytes on-chip data RAM, 1 Kbyte non-volatile
data EEPROM, 16 x 16-bit working register array, Up to 30 MIPs operation:
- DC to 40 MHz external clock input
- 4 MHz-10 MHz oscillators input with PLL active (4x, 8x, 16 xs)
Up to 42 interrupt sources, three external interrupt sources
- 8 user selectable priority levels
- 54 interrupt vectors
- 8 processor exceptions and software traps

Figure13 : Schematic diagram of Biped Robot

DESIGN OF SERVO MOTOR CONTROLLER FOR BIPED ROBOT 293

5.2 DSP Engine Features

Modulo and Bit-Reversed modes, Two, 40-bit wide accumulators with optional saturation
logic, 17-bit x 17-bit single cycle hardware fractional/integer multiplier, Single cycle
Multiply-Accumulate (MAC) operation,40-stage Barrel Shifter, Dual data fetch.
5.3 Peripheral Features
High current sink/source I/O pins: 25 mA/25 mA, Three 16-bit timers/counters; optionally
pair up 16-bit timers into 32-bit timer modules, Four 16-bit Capture input functions, Two
16-bit Compare/PWM output functions- Dual Compare mode available, 3-wire SPITM
modules (supports 4 Frame modes), I2CTM module supports Multi-Master/Slave mode and
7-bit/10-bit addressing, Addressable UART modules with FIFO buffers.
5.4 CMOS Technology
Low power, high speed Flash technology, Wide operating voltage range (2.5V to 5.5V),
Industrial and Extended temperature ranges, Low power consumption.
5.5 Special Microcontroller Features
Enhanced Flash program memory:- 10,000 erase/write cycle (min.) for industrial temperature
range, 100K (typical) Data EEPROM memory:- 100,000 erase/write cycle (min.) for
industrial temperature range, 1M (typical), Self-reprogrammable under software control,
Power-on Reset (POR), Power-up Timer (PWRT) and Oscillator Start-up Timer (OST),
Flexible Watchdog Timer (WDT) with on-chip low power RC oscillator for reliable
operation, Fail-Safe clock monitor operation, Detects clock failure and switches to on-chip
low power RC oscillator, Programmable code protection, In-Circuit Serial Programming™
(ICSP™), Selectable Power Management modes-Sleep, Idle and Alternate Clock modes.
5.6 Motor Control PWM Module Features
Six PWM output channels- Complementary or Independent Output modes- Edge and Center
Aligned modes, 4 duty cycle generators, Dedicated time base with 4 modes, Programmable
output polarity, Dead-time control for Complementary mode, Manual output control, Trigger
for synchronized A/D conversions.

CONCLUSION
A Biped Robot being designed and which is controlled by servo motor controller using DSPIC30F2010
is proposed in this paper. An extensive Literature Survey conducted for the project gave profound insight
294 ROOPA C, [Link] AND NITHIN AWASTHI

on the requirements for building the Biped robot. Based on the Literature survey, the inputs for designing
the robot have been decided. Hardware will be implanting in the future work.

ACKNOWLEDGMENT
Our special thanks to [Link] for giving up an idea about Robotics and also all Professors of The
oxford college of Engineering, Bangalore, VTU India for giving good support.

REFERENCES

[1] Shih-An Li, Chung-Wei Weng, Yi-Hong Chen, Chia-Hung Lo, Min-Hao Yang, Yi-Chun Lin,
Ming-Hua Hsieh, “Servomotor controller design for robotic manipulator” 2012 IEEE International
Symposium on Intelligent Signal Processing and Communication Systems (ISPACS 2012)
November 4-7, 2012

[2] Moshi A.A., Cynthia S.S., Islam E., Rahman R., and Azad A., "Performance analysis of robotic
arm manipulators control system under multitasking environment," 2011 IEEE 18th International
Conference on Industrial Engineering and Engineering Management pp.613-617, Sept. 2011

[3] Seki K., Yokoi H. & Iwasaki M., Experimental evaluations of friction behavior in
micro-displacement region positioning for servo motor with air bearings, Proceeding of IEEE
International Conference on Advanced Intelligent Mechatronics, 2012.

[4] Xiaosheng L., Yuqiang W., Nantian H. & Yue H., The Networked Virtual Test System for Servo
Motor and Drive Based on LabVIEW, Proceeding IEEE 7th international conference on Power
Electronics and Motion Control, 1-5, 2012.

[5] Andre Senior and Sabri Tosunoglu – “Design of a Biped Robot”, Florida Conference on Recent
Advances in Robotics, FCRAR, May25-26, 2006.

[6] Lee C.K., Design and implementation of the FPGA-based rotor flux oriented control law for an
Induction motor drive, Master Thesis, Department of Electrical Engineering, National Ilan
University, Taiwan, 2009

[7] Chen S.Y., Application of SOPC embedded system on high precision motion controller, Master
Thesis, Department of Electrical Engineering, National Cheng Kung University, Taiwan, 2006.

[8] [Link] – USED FOR THE UNDERSTANDING ABOUT DSPIC30F2010.

[9] [Link] – used for the learning about HXT900-9g micro servomotor
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 295-307

THE IMPLEMENTATION OF FIELD ORIENTED

CONTROL FOR PMSM DRIVE BASED ON
TMS320F28035 DSP CONTROLLER

ROHITH KUMAR C1 AND V. CHINAPPAN2

1
. PG Student, Computer Application in Industrial Drives, Dept of EEE,
The Oxford College of Engineering, Bangalore-560068,
Affiliated to Visvesvaraya Technological University, Karnataka, India.
.2 Professor and HOD, Dept of EEE , The Oxford College of Engineering, Bangalore-560068,
Affiliated to Visvesvaraya Technological University, Karnataka, India.

Abstract
This paper presents the implementation of the Permanent magnet synchronous motor (PMSM)
controller by using Field Oriented Control (FOC) method. The digital signal processor (DSP) was
used as a controller to interface between the FOC and the PMSM. In this project, a floating 32 bit
DSP controller TMS320F28035 is used to realize the drive system. TMDSHVMTRPFCKIT board by
Texas Instrument is used to run the motor. The results show that the speed of PMSM was successfully
follows the reference speed.

-----------------------------------
Keywords : Digital Signal Processor, Field Oriented Control, Permanent Magnet Synchronous Motor
© http: //[Link]
296 ROHITH KUMAR C AND V. CHINAPPAN

1. INTRODUCTION
Permanent magnet synchronous motor (PMSM) become popular in industry because of their
advantageous such as light weight, compactness and cost performances [1]-[3]. This type of
motor is most applicable where the high speed performance is needed. With elimination of a
commutator, PMSM is become more reliable than a DC motor and become more efficient
than an AC induction motor because the production of the rotor flux is from a permanent
magnet. In order to achieve high performance control characteristics, vector control is used
to control the PMSM [4],[5].
The FOC is one of the vector based method that aims to control the torque and rotor flux of
the PMSM effectively. The FOC is carry out to control the space vector of magnetic flux,
current and voltage of the machines in order to achieve the precise speed target.
The stator currents is set as a control variable and this three phase static reference frame of
the stator current is transformed and is performed in the d-q coordinate reference frame of
the motor. The magnetic flux produced from the rotor of PMSM is locked to the vector of
the rotor flux and will rotate at the stator frequency. The voltage supplied to the motor is
transformed from d-q coordinate reference frame of the rotor to the three phase static
reference frame of the stator before it can be fed to Space Vector Pulse Width Modulation
(SVPWM) to get PWM output. The implementation of vector based FOC which is build
upon stator-flux oriented method gives a better steady state operation.
In order to yield a high performance system, the selection of the microprocessor is important.
Many controllers available in market have high capability to achieve high performance
application in electrical motor drive. The use of DSP controllers now becomes beneficial
because it can incorporate with a multiple advanced power electronics peripherals and
simplify the design process, it also have the capability to incorporate with various extra
features in the drive [6], [7], [8].
This paper describes the implementation of the PMSM controller by using FOC method and
application of the DSP to interface between the FOC and PMSM. All the architecture of the
FOC as shown in Figure 1 was implemented in a TMS320F28035 DSP controller
THE IMPLEMENTATION OF FIELD ORIENTED CONTROL… 297

2. FIELD ORIENTED CONTROL

Mathematical model of PMSM used is according to the d-q Synchronous reference frame.
The stator voltages and magnetic flux equation in the d-q synchronous reference frame are
given as follows:
dids
vds = R s ids + Lds dt
− ωr Ψqs (1)
diqs
vqs = R s iqs + Lqs dt
− ωr Ψds (2)

Ψds = Lds ids + Ψm (3)

Ψqs = Lqs iqs (4)
vds and vqs are the stator’s d-q axis voltage respectively, ids and iqs are the stator’s d-q axis
currents, 𝑅𝑠 is the stator resistance, 𝐿𝑑𝑠 and 𝐿𝑞𝑠 are the d-q axis stator inductances, Ψ𝑑𝑠 and
Ψ𝑞𝑠 are the d-q axis stator magnetic flux, 𝜔𝑟 is the electrical rotor speed and Ψ𝑚 is the
rotor’s permanent magnetic flux.
The current model can be represented by combining equation (1) and (3) for d-axis current
and equations (2) and (4) for q-axis current and can be described by equations (5) and (6) as
follows.
𝑑𝑖𝑑𝑠 𝑅 𝜔𝑟 𝐿𝑞𝑠 1
𝑑𝑡
= − 𝐿 𝑠 𝑖𝑑𝑠 + 𝑖𝑞𝑠 + 𝐿 𝑣𝑑𝑠 (5)
𝑑𝑠 𝐿𝑑𝑠 𝑑𝑠

𝑑𝑖𝑞𝑠 𝑅 𝜔𝑟 𝐿𝑑𝑠 1 𝜔𝑟 𝜓𝑚
= − 𝐿 𝑠 𝑖𝑞𝑠 − 𝐿𝑞𝑠 𝑞𝑠
𝑖 + 𝐿 𝑣𝑞𝑠 − (6)
𝑑𝑡 𝑞𝑠 𝑞𝑠 𝐿𝑞𝑠

The developed torque motor is given by;

3 𝑃
𝜏𝑒 = 2 ( 2 ) (Ψ𝑑𝑠 𝑖𝑞𝑠 − Ψ𝑞𝑠 𝑖𝑑𝑠 ) (7)

The mechanical torque motor is given by;

𝑑𝜔𝑚
𝜏𝑒 = 𝜏𝐿 + 𝐵𝜔𝑚 + 𝐽 𝑑𝑡
(8)

The rotor mechanical speed equation is obtained by rearranging equation (8);

𝜏𝑒 −𝜏𝐿 +𝐵𝜔𝑚
𝜔𝑚 = ∫ 𝐽
(9)

And;
2
𝜔𝑚 = 𝜔𝑟 𝑃 (10)

𝜏𝑒 is developed torque, 𝜏𝐿 is load torque, P is number of pole, 𝜔𝑚 is rotor mechanical speed

and 𝜔𝑟 is rotor electrical speed. Based on the FOC scheme diagram shown in figure 1, the
closed loop control is applied to the PMSM. Due to the use of permanent magnet type of the
298 ROHITH KUMAR C AND V. CHINAPPAN

rotor, the flux linkage remains constant. For this project, the torque of PMSM is controlled
by the q-axis and the d-axis of the stator current is set to [Link] implementation of
SVPWM has been adapted to the inverter circuit as illustrated in Figure 1.

Figure 1 : Block diagram of the drive system

Park transformation is applied to change the phase current to the d-q components to obtain a
complete decoupling of torque ( 𝑖𝑞 ) and flux ( 𝑖𝑑 ). By taking equation (3), the d-axis flux
linkage remain unchanged when 𝑖𝑑 is set to zero such that
𝛹𝑑𝑠 = 𝛹𝑚𝑠 (11)
The electromagnetic torque is given by following equation.
3
𝜏𝑒 = 2 𝑃[𝑖𝑞𝑠 𝛹𝑚𝑠 ] (12)

The torque control of PMSM is done when stator currents interact with rotor flux linkage.
The transformation of the voltage and current is done by the following transformation:
 Clarke transformation: stationary a-b-c frame to stationary 𝛼 – 𝛽 frame.
2 −1 −1 𝑓
𝑓𝛼 𝑎
[𝑓 ] = [ 3 3
1
3
−1 ] [ 𝑓𝑏] (13)
𝛽 0
√3 √3 𝑓𝑐

 Inverse Clarke transformation: stationary 𝛼- 𝛽 frame to stationary a-b-c frame

THE IMPLEMENTATION OF FIELD ORIENTED CONTROL… 299

1 0
𝑓𝑎 −1 √3 𝑓𝛼
[𝑓𝑏 ] = 2 2 [ ]
𝑓𝛽 (14)
𝑓𝑐 −1 −√3
[2 2 ]

 Park transformation: stationary 𝛼- 𝛽 frame to synchronously rotating d-q frame.

𝑓𝑑𝑠 cos 𝜃 sin 𝜃 𝑓𝛼
[𝑓 ] = [ ][ ] (15)
𝑞𝑠 − sin 𝜃 cos 𝜃 𝑓𝛽
 Inverse park transformation: synchronously rotating d-q frame to stationary 𝛼- 𝛽
frame.
𝑓𝛼 cos 𝜃 −sin 𝜃 𝑓𝑑𝑠
[𝑓 ] = [ ][ ] (16)
𝛽 sin 𝜃 cos 𝜃 𝑓𝑞𝑠

3. DSP CONTROLLER
The TMS320F28035 Digital Signal Processor (DSP) is used in this project. The selected
DSP controller, TMS320F28035 has High-Performance Static CMOS Technology operating
at 60MHz (16.67-ns Cycle Time), low-power supply 1.8V for DSP Core and 3.3V for the
I/O buffers [9]. The DSP controllers offer 60 MIPS of 32-bit DSP performance possible to
position the edge of a PWM signal with 150 ps precision or 16 bits accuracy in a 100 KHz
control loop [10]. Figure 2 shows a block diagram of TMS320F28035 controller. The
controllers combine a number of peripherals, such as 64 KB of flash memory, Boot ROM
(8K x16), 45-GPIO pins,16-channel (2 x 8 Channel Input Multiplexer) 12 bit analog to
digital converter (ADC), 3 32-bit CPU timers, 14 independent PWM channels, 1 quadrature
encoder pulse (QEP), and 1 CAP input for position sensing [9]. A 32-bit wide data path
enough to give awful system performance while mixed 16-bit/32-bit instruction achieves
code density. Key communication interfaces include multiple serial port peripheral such as
SCI, CAN, I2C, UART, SPI ports and Watch Dog timer module.
For this project the reference speed is set at pre- specified values in DSP controller. The
High Voltage Digital Motor control (DMC) and Power Factor Correction (PFC) kit
(TMDSHVMTRPFCKIT), provides a great way to control the high voltage motors digitally.
In this project 3 phase PMS motor is connected with a incremental encoder, the output of the
encoder is fed to QEP module at TMS320F28035 to read the actual speed and resulting
angle. The voltage and current sensors sense the voltage across the motor and current
through the motor respectively.
300 ROHITH KUMAR C AND V. CHINAPPAN

Figure 2 : Block diagram of DSP(TMS320F28035) [9]

This voltage and current values are in analog form, using ADC in TMS320F28035 DSP the
is converted to digital. On the basis of the obtained values from encoder, voltage sensor and
current sensor the DSP will configure and generate 6 PWM signals. This PWM signals
drives the driver circuit block to operate the 3 phase Voltage Source Inverter (VSI) as shown
in Figure 3. VSI 3-phase output is given to motor.
The voltage and current obtained by sensors is monitored continuously. This analog signal
will be converted to digital signal to be processed by the DSP. The FOC operation of Park,
Inverse Park, Clark, Inverse Clark, SVPWM and PI controller are executed in this DSP.
THE IMPLEMENTATION OF FIELD ORIENTED CONTROL… 301

Figure 3 : TMDSHVMTRPFCKIT with F28035 DSP Controller[10]

302 ROHITH KUMAR C AND V. CHINAPPAN

4. HARDWARE EXPERIMENTAL SETUP

Figure 4 shows the experimental setup for this project. It consists of TMS320F28035 DSP
controller, PMSM and TMDSHVMTRPFCKIT. The 3-phase, 8 poles and a incremental
encoder mounted along with shaft of the PMSM used is manufactured by ESTUN (EMJ-
04APB22). The main controller which is DSP development board is interfaced with three
phase voltage source inverter.

Figure 4 Hardware experimental setup

This TMDSHVMTRPFCKIT board has the following specification,

 Kit contents
 TMS320F28035 DSP Control card
 High Voltage DCM board
 400V to 15V and 5V power supply
 USB-B to A cable
 Onboard isolated JTAG emulation
 Heat sink attached with DC fan
THE IMPLEMENTATION OF FIELD ORIENTED CONTROL… 303

 3-phase inverter stage

 350V DC max input voltage
 1.5KW max load
 QEP and CAP inputs available for speed and position measurement
 Power factor correction stage
 750W max power rating
 400V DC max output voltage
 200KHz switching frequency for power stage
 85-132VAC/ 170-250VAC rectified input
 Up to 100KHz PFC control loop frequency
 AC rectifier stage
 750W max power rating
 85-132VAC/ 170-250VAC input.
In this project, the response of the system is evaluated by using difference pre defined
speeds. The parameters of the motor is shown in Table 1.0
Table 1 : Motor parameters

Parameters Values
Voltage (V) 220
Output Power (kw) 0.4
Rated speed (rpm) 3000

Rated torque (Nm) 1.27

No of Poles 8
Stator resistance (ohm) 0.79
Stator inductance (mH) 1.17
Flux ([Link]/rad) 0.017666

5. EXPERIMENTAL RESULTS
Figure 5 until figure 8 shows the obtained response of the motor. The system is tested under
no-load and load conditon at Deadband = 0.83usec, dlog.trig_value=100, Vdcbus=300V,
[Link] =3 at room temperature.
304 ROHITH KUMAR C AND V. CHINAPPAN

Figure 5 : Output of 𝑻𝒂 , 𝑻𝒃 , 𝑻𝒄 and 𝑻𝒃 -𝑻𝒄 waveform

Figure 5 shows the output wave form of 𝑇𝑎 , 𝑇𝑏 and 𝑇𝑐 are 1200 apart from each other .

Figure 6 : The waveform to measure theta and phase A & B current waveform
THE IMPLEMENTATION OF FIELD ORIENTED CONTROL… 305

Figure 7 : Measured theta, svgen dutycycle and phase

A&B current under no-load and 0.3 pu speed

Figure 8 : Measured theta, svgen dutycycle and phase

A&B current under 0.33pu load and 0.3 pu speed
306 ROHITH KUMAR C AND V. CHINAPPAN

As shown in figure 7 at no load condition noice is pressent in current waveform at low

speed. After enploying 0.33pu load on motor at same low speed the noise is reduced as
shown in figure 8. At the low speed range , the performance of speed response relies heavily
on the good rotor position angle provided by QEP encoder.
Figure 9 shows the speed response when the motor is observed through GUI. At starting, the
motor is under forward mode of operation which was the speed set at 750 rpm. Then, after
some time, the motor speed is changed to 450 rpm and estimated speed is recorded as 456
rpm. It clearly shows in Figure 9 that the actual speed follows the reference speed.

Figure 9 : Speed response

THE IMPLEMENTATION OF FIELD ORIENTED CONTROL… 307

CONCLUSION
From the result obtain, it shows that the speeds follow the reference speed under various
mode of operation. Therefore, FOC method can be used to control the PMSM and DSP
controller can be implemented to give the high performance drives system.

ACKNOLOWLEDGEMENT
Financial support and lab facilities from The Oxford College Of Engineering, Bangalore for
carrying out this work is gratefully acknowledged.

REFERENCES

[1] Bose B.K., Power electronics and variable frequency drives –technology and application, IEEE
Press, 1997.
[2] Rahman M.A, Hoque M.A, “Online Adaptive Artificial Neural Network Based Vector Control
of Permanent Magnet Synchronous Motors”, IEEE Trans on Energy Conversion, Vol. 13, no 4,
December 1998.
[3] Pillay P, Krishnan R, “modeling, Simulation and Analysis of Permanent Magnet Synchronous
Motor Drives, Part 1: The Permanent magnet Synchronous Motor Drive”, IEEE Trans, Ind. App,
Vol. 25, March, April 1989.
[4] Narayanan G., Zhao Di, Krishnamurthy Harish K, Ayyanar Rajapandian, Ranganathan V.T,
“Space Vector Based Hybrid PWM Techniques for Reduced Current Ripple”, IEEE Transaction
on India Electronics, Vol. 55, No 4, April 2008.
[5] Radwan T.S., Rahman M.A., Osheiba A.M. and Lashine A.e., “Dynamic Analysis of A High
Performance Permanent Magnet Motor Drive,” Electrical and Computer Engineering, 1996,
Canadian Conference on vol. 2.
[6] Trzynadlowski A.M, “Field Orientation Principle in Control of Induction Motors,” Kluwer
Academic Publishers, 1994.
[7] Shireen W, Arefeen M S and Figoli D, “Multiple Motor Control Using a Single DSP
Controller,” IEEE Applied Power Electronic Conference (IEEE-ApEC) 1999.
[8] Mohammed Arefeen, Godbole Kedar and Konghirun Mongkol “Dual field oerinted permanent
magnet synchronous motor drive using a sinlge DSP controller.”
[9] TMS320F28035 Digital Signal Processors Datasheet, Texas Instruments.
[10] The High Voltage Digital Motor control (DMC) and Power Factor Correction (PFC) kit
(TMDSHVMTRPFCKIT) Technical References.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 309-323

OPTIMIZATION OF HYBRID RENEWABLE ENERGY

SYSTEM USING GENETIC ALGORITHM

VEERESH BYAHATTI, AKSHAYA KATKAR AND A. B. RAJU

Department of Electrical & Electronics Engineering,
B. V. B College of Engineering & Technology,
Hubli- 580031, Karnataka - India

Abstract
In this paper we have used the Genetic Algorithm for optimization of Hybrid Renewable Energy
System (HRES). Our objective in optimization is to minimize the total cost consisting of capital and
maintenance costs and at the same time achieve the required value of Loss of Power Supply
Probability (LPSP). This project has the objective of creating an optimization program for feasibility
assessment and optimal sizing of HRES. This involves the development of a program which would
analyze information input about the climate of the area and the load demand. Climatic data for
temperature, wind speed and solar irradiance will be needed for the accurate execution of the program.
With the information specified above, this program would then be used to calculate the size and type
of the HRES that is best suited, in view of the energy demands calculated for that area. Another aspect
of this model would be energy storage: the method of modeling energy storage, sizing the optimal
amount of energy storage and its operation and control in HRES. Because of the inherent
intermittency of renewable energy techniques which are reliant on wind speed and irradiance levels, it
is important to store energy when there is an excess of energy and to be able to supply energy when
there is insufficient being generated.

-----------------------------------
Keywords : Genetic Algorithm; Optimal configuration; HRES; Standalone Systems; Cost
Minimization. © http: //[Link]
310 VEERESH BYAHATTI, AKSHAYA KATKAR AND A. B. RAJU

1. INTRODUCTION
Proper design of HRES is a challenging task, as the coordination among renewable energy
resources, generators, energy storages and loads is very complicated. Generally the main
objectives of the optimization design are power reliability and cost.
In this paper an optimal sizing method using the genetic algorithm (GA) is proposed. The
number of wind turbines, size of photo voltaic panels and the capacity of batteries can be
optimized when sizing a HRES, which may be defined as a constrained optimization
problem.
The main objective is that minimize the cost of the system keeping the Loss of Power Supply
Probability (LPSP) within the specified limit. By using the climatic data and load demand
the LPSP can be calculated and simultaneously cost of the stand alone energy system should
be minimized.

2. LOSS OF POWER SUPPLY PROBABILITY

2.1 System Configuration
The configuration of HRES is shown in Fig.1.

Figure 1 : Block diagram of HRES

2.2 Models of System Components
[1] Present the following method of calculating the power of the PV panels at the specified
temperature and irradiance:
𝑃𝑚𝑝 = 𝑉𝑜𝑐𝑖 𝐼𝑠𝑐𝑖 𝐹............................................... (1)
OPTIMIZATION OF HYBRID RENEWABLE ENERGY… 311

Where, Vosci is the open circuit voltage at the specified temperature and irradiance, and is
calculated in equation 3, Isci is the short circuit current at the specified temperature and
irradiance, and is calculated in equation 2, and FF is the fill factor of the panel.
𝐺
𝐼𝑠𝑐𝑖 = [𝐼𝑠𝑐 + 𝐾𝐼 (𝑇𝑐 − 𝑇𝑛 )] 𝐺 ……………………………………………. (2)
𝑛

Where Isc is the short circuit current at STC, KI is the temperature coefficient for short circuit
current, Tc is the calculated temperature, Tn is the STC temperature, G is the irradiance and
Gn is the irradiance at STC.
𝑉𝑜𝑐𝑖 = 𝑉𝑜𝑐 − 𝐾𝑉 𝑇𝑐 ………………………………………………….……….(3)
Where Voc is the open circuit voltage at STC, KV is the temperature coefficient for open
circuit voltage and Tc is the calculated temperature, as calculated in equation 4.
𝑁𝐶𝑂𝑇−20
𝑇𝑐 = 𝑇 + 800
𝐺………………………………..(4)
Where T is the PV panel operating temperature, NOCT is the Nominal Cell Operating
Temperature as specified by the manufacturer and G is the irradiance.

Figure 2 : Wind Speed - Power Profile of Kestrel e400i

As wind speed measurements are not always taken at the same height as the proposed height
of the wind turbine, the wind speed must be altered to ensure that the wind speed at the
312 VEERESH BYAHATTI, AKSHAYA KATKAR AND A. B. RAJU

proposed height of the wind turbine is used rather than the wind speed at the height
measured. Equation 5 is used to calculate the wind speed at the appropriate height:
ℎ 𝛼
𝑣 = 𝑣0 (ℎ ) …………………………..……….…..(5)
0

Where v is the wind speed at the height of the wind turbine, h and v0 is the reference wind
speed at the height of measurement, h0. 𝛼 is the wind speed power law coefficient, which
1 1 1
varies from 4 to 7 depending on the type of terrain at the site. For open land 7 is used as the
standard factor. But in this paper the height is kept constant as 10m.
To find out the power generated by wind turbine can be done by creating a look up table
using the wind speed/power profile and details supplied by the manufacturer as shown in
Fig.2 the data generally supplied by the manufacturer is usually substantial enough to do this
and interpolate values without losing much accuracy in the system.
This is the approach that I’ve taken considering Matlab’s vast ability to interpolate and store
data efficiently. The following code was used to implement the initialization of the look up
table.
WSda = transpose([2:0.5:12]);
WPda=[23;45;78;123;185;263;361;481;624;794;991;1219;1480;1775;2108;2479;
2891;3050;3120;3150;3150];
WSdai = 2:0.01:12;
WPdai = interp1(WSda,WPda,WSdai);
Where WSda and WPda represent the wind speed and wind power data supplied by the
datasheet and WSdai and WPdai represent the interpolated data with values at every 0.01
interval.
The wind power can then be looked up by finding the value closest to the input of wind
speed from the user and allocating the matching wind power value.
Lead acid batteries are main energy storage devices in standalone power systems. The
battery charge efficiency is set equal to the round-trip efficiency, and the discharge
efficiency is set equal to 1. The maximum battery life can be obtained if the depth of
discharge (DOD) is set equal to 30%-50%, for example DOD=50%. The batteries are
normally installed inside a building where the temperature is not expected to change
drastically, so the temperature effects are not considered in this paper. The energy stored in
batteries at any hour t, EB,t is subjected to the following constraint:
OPTIMIZATION OF HYBRID RENEWABLE ENERGY… 313

𝐸𝐵𝑚𝑖𝑛 ≤ 𝐸𝐵,𝑡 ≤ 𝐸𝐵𝑚𝑎𝑥 ………………………………………….………….(6)

where,
EBmin is battery minimum allowable energy level
EBmax is battery maximum allowable energy level
Let EBmax as battery´s nominal capacity with the value of the battery state of charge (SOC) as
1, then
𝐸𝐵𝑚𝑖𝑛 = (1 − 𝐷𝑂𝐷)𝐸𝐵𝑚𝑎𝑥 ………………………………………….………..(7)
The MPPT, the battery controller, the inverter and distribution lines are assumed to have
constant efficiencies. Assume the efficiencies of the MPPT, the battery controller and
distribution lines as 1 and that of the inverter as 0.9.
A. Loss of Power Supply Probability
The Loss of Power Supply Probability (LPSP) [2], which is defined in terms of the SOC, is
the power reliability index of a system. LPSP can be defined as the long-term average
fraction of the load that is not supplied by the standalone power system.
The SOC of the batteries at any time t1 depends on the SOC in the previous moment t0 and
the sequence of generated power and load demand levels in the time interval t1 - t0. The SOC
is used as a decision variable for the control of over charge and over discharge. This is
important to prevent batteries from shortening their life or even their destruction.
The simulation period is one year and the time step is one hour. The load, solar energy and
wind energy are assumed to be constant during a time step. The energy generated by the PV
array and the wind turbine for hour t, EG,t, can be expressed as:
𝐸𝐺,𝑡 = 𝑁𝑝𝑣 𝐸𝑝𝑣,𝑡 + 𝑁𝑤𝑡 𝐸𝑤𝑡,𝑡 …………..…………………………………………………...(8)
where,
Epv,t energy generated by the PV array in that hour t
Npv number of PV panels in the PV array.
Ewt,t energy generated by the wind turbine in that hour t
Nwt number of wind turbines.
Batteries form a large part of the capital investment of RESs and therefore it is important to
manage the system to ensure that the maximum lifespan of the battery is achieved by
controlling its charging and discharging cycles. In addition the system must be sized
correctly to ensure that the load is always met by the supply and to minimize the total cost of
the system.
314 VEERESH BYAHATTI, AKSHAYA KATKAR AND A. B. RAJU

Modeling and simulation of batteries within a RES design has a twofold purpose: firstly, to
study their performance in a stand-alone system as an energy storage device, and secondly to
enable the optimization of the size of batteries in order to minimize the overall operating
costs. Modeling batteries can be a challenging task as there are many parameters to take into
account, such as the State of Charge (SoC) of the battery, various voltages, internal
resistances, battery capacitance, charge resistance, discharge resistance and other values
which do not remain constant. These parameters are difficult to measure and predict within a
system where the charging and discharging cycles are often unpredictable and the
temperature of the system is not constant, as is the case with most RESs.
Lead acid batteries are currently the most used types of batteries in power applications. They
are used for varying energy density requirements, but are most effective for lower power
densities. While they have a good energy density, their power density is limited and
therefore the amount of energy that can be supplied to the system and the time taken to
charge the battery is significant. Lead acid batteries have a relatively short lifespan and
therefore need to be replaced periodically. This is therefore the limiting factor in isolated
power generation.
Due to the inherent unpredictable nature of renewable energy sources and the significant
dependence of the power generated on climatic conditions, it is very important to simulate
all components of the system together before implementation to ensure that the power supply
needed for the load is met at all times. Therefore the extent to which the battery can store
and supply energy is very important within a standalone system and must be simulated to
ensure that the system will be able to meet the load.
If the generated energy from the PV array exceeds that of the load demand, the batteries will
be charged with the roundtrip efficiency:
𝐸𝐿,𝑡
𝐸𝐵,𝑡 = 𝐸𝐵,(𝑡−1) (1 − 𝜎) + [𝐸𝐺,𝑡 − ] ƞ𝑏𝑎𝑡 ………………………………………(9)
ƞ𝑖𝑛𝑣

where,
EBmin is battery minimum allowable energy level
EBmax is battery maximum allowable energy level
DOD is Depth of Discharge
EB,t is energy stored in the batteries in hour t
EB,(t-1) energy stored in the batteries in previous hour
OPTIMIZATION OF HYBRID RENEWABLE ENERGY… 315

EL,t load demand in hour t

ƞbat is round-trip efficiency of the batteries
ƞinv is inverter efficiency
σ is self-discharge rate per hour of the batteries
When the load demand is greater than the available energy generated, the batteries will be
discharged by the amount that is needed to cover the deficit. This can be expressed as:
𝐸
𝐸𝐵,𝑡 = 𝐸𝐵,(𝑡−1) (1 − 𝜎) − [𝐸𝐺,𝑡 − ƞ 𝐿,𝑡 ] ƞ𝑏𝑎𝑡 …………………………………….……..…(10)
𝑖𝑛𝑣

When the energy generated is greater than the load demand, the batteries will be charged.
This can be expressed as:
𝐸
𝐸𝐵,𝑡 = 𝐸𝐵,(𝑡−1) (1 − 𝜎) + [𝐸𝐺,𝑡 − ƞ 𝐿,𝑡 ] ƞ𝑏𝑎𝑡 ………………………………………(11)
𝑖𝑛𝑣

When the available energy generated and stored in batteries is insufficient to satisfy the load
demand for hour t, that deficit called Loss of Power Supply (LPS) for hour t can be
expressed as:
∑𝑇
𝑡=1 𝐿𝑃𝑆𝑡
𝐿𝑃𝑆𝑃 = ∑𝑇
…………………………………………(12)
𝑡=1 𝐸𝐿𝑡

The LPSP for a considered period T is the ratio of all LPSt values for that period to the sum
of the load demand, as defined by [2]:
𝐿𝑃𝑆𝑡 = 𝐸𝐿𝑡 − (𝐸𝑔𝑡 + 𝐸𝐵𝑡−1 − 𝐸𝑚𝑖𝑛 )ƞ𝑖𝑛𝑣 …..………………………..……..(13)

3. GENETIC ALGORITHM
Genetic algorithm is a type of minimum seeking optimization which mimics the behavior of
chromosome genetics and survival of the fittest in biology.
As shown in Fig.3 this is done by creating a number, or population, of ‘chromosomes ’,
which represent possible solutions to the problem to be optimized. These ‘chromosomes’ are
then measured against some criteria, and the specified percentage of those that match the
criteria the best are kept, while those that match the criteria least are discarded. This is in
keeping with the biological principles of natural selection. The kept chromosomes are then
sorted and some chosen to ‘crossover ’with other chromosomes. This can be done in a
number of ways, but the outcome is that the two chromosomes are combined to form new
chromosomes. The population are then crossed over in this way to make up the population
size to replace those that were discarded. Another important biological aspect of genetics
316 VEERESH BYAHATTI, AKSHAYA KATKAR AND A. B. RAJU

which is used in genetic algorithms is mutation. In genetic algorithms, a percentage of the

population will undergo mutation which can also happen in a variety of ways which will be
discussed, but essentially where a random section of a chromosome is changed or mutated to
give a different chromosome. This introduces some change especially when the parent
chromosomes are all very similar. The process of crossover and mutation is then reiterated
until either the specified number of population generations has been reached or until some
other stopping criterion is reached.

Figure 3 : Block diagram of Genetic algorithm

OPTIMIZATION OF HYBRID RENEWABLE ENERGY… 317

Genetic algorithms have the advantage of being able to solve discrete or continuous
problems, because of the fact that there is a population of solutions, of being able to look at a
wide variation of solutions at the same time and therefore is more able to find a global
optimum solution rather than get stuck in a local optimum, of not being restricted in the
number of variables it uses and of being able to handle very tricky objective functions with
ease, amongst other advantages. Whilst in some simpler cases, it may not be the best or
quickest method of finding a solution, when it comes to complex problems with many
constraints as many variables, such as the problem that is presented in this project, it is a
very effective way of tackling a problem. And the flowchart of Genetic Algorithm as shown
below in Fig. 4.
In this paper, integer encoding and the elitist strategy are employed. The form of the
individual of the GAs population is [Npv Nwt Nbat]

4. OPTIMAL SIZING USING THE GENETIC ALGORITHM

4.1 Problem Description and Sizing Procedures
The reliability and cost are two objectives of sizing standalone hybrid PV power systems.
The cost index is the capital cost [1] in this paper. The size of PV panels and the capacity of
batteries have a considerable influence upon the power reliability and capital cost and can be
optimized. The optimization of standalone hybrid PV power systems is a mixed multiple
criteria integer programming problem [3].
𝑐𝑜𝑠𝑡(𝑁𝑝𝑣 , 𝑁𝑤𝑡 , ℎ, 𝑁𝑏 ) = 𝑁𝑝𝑣 (𝐶𝑐𝑝𝑣 + 20𝐶𝑚𝑝𝑣 ) + 𝑁𝑤𝑡 (𝐶𝑐𝑤𝑡 + 20 𝐶𝑐𝑚𝑡 + ℎ 𝐶𝑐𝑤𝑡𝑡 +
20 ℎ 𝐶𝑚𝑤𝑡𝑡 ) + 𝑁𝑏 (𝐶𝑐𝑏 + 𝑦𝑏 + 𝐶𝑐𝑏 + (20 − 𝑦𝑏 − 1)Cmb )…………....(14)
where,
h = 10 Height of wind system
Ccpv is capital cost each PV panel
Cmpv is maintenance cost each PV panel per year
Ccwt is capital cost each wind turbine
Cmwt is maintenance cost each wind turbine per year
Cwtt is capital cost per metre of each wind turbine
Cmwtt = 0.01(Ccwtt) is maintenance cost per metre of each wind turbine per year Ccb is capital
cost each battery
318 VEERESH BYAHATTI, AKSHAYA KATKAR AND A. B. RAJU

Cmb = 0.01(Ccb) is maintenance cost each battery yb The number of replacements that need to
be made to the battery over the 20 years.

Figure 4 : Analogy between a numerical GA and biological genetics.

4.2 The Genetic Algorithm
The GA is a stochastic global search method that mimics the metaphor of natural biological
evolution and does not require derivative information or other auxiliary knowledge. It is
OPTIMIZATION OF HYBRID RENEWABLE ENERGY… 319

important to note that GA provides a number of potential solutions to a given problem and
the choice of final solution is left to the user [4].
In this paper, integer encoding and the elitist strategy are employed. The form of the
individual of the GAs population is [Npv Nwt Nbat]. The LPSP of every individual is calculated
by simulations for 8760 hours.

5. APPLICATION EXAMPLE
As the climatic data consists of solar irradiance and temperature. The typical data set is as
shown in table IV [5]. The capacity of a single battery used was 200Ah. That battery has a
round-trip efficiency of 0.7 and DOD=10%. From the table IV the maximum load demand is
9000W 5,for reliable operation the maximum LPSP = 0:01.
The cost of each material is shown in table I and the maximum output power of each device
is shown in table II.

Figure 5 : Load during a day

Table 1 : Cost Of Each Material And Maintenance Cost

Material Cost
Capital cost of each PV panel 30,000
320 VEERESH BYAHATTI, AKSHAYA KATKAR AND A. B. RAJU

Material Cost
Maintenance cost of each PV panel per year 300
Capital cost of each wind turbine 4,09,850
Maintenance cost of each wind turbine per year 4,098
Capital cost of per metre of each wind turbine 16,900
Maintenance cost of per metre of each wind turbine per year 169
Capital cost of each battery 11,265
Maintenance cost of each battery per year 112

Table 2 : Maximum Power Rating Of Each Device

Device Power
Wind Generator 3kW
PV Panel 200W
Battery 200Ah

The different parameters which are set in algorithm are shown in table III.
Table 3 : Ga Parameters
Parameter Value
Maximum Number of iterations 500
Initial Population 100
Mutation rate 0.15
% selection 50%

6. RESULTS
The results of the GA are shown in table V for both rated load and doubled rated load.
The convergence of cost function as shown in below figures. The cost of HRES has reduced
with increase in the iteration and has reached optimal value and remains constant at this
value till the end of iterations. The Fig. 6 corresponds the rated load and the Fig.7
corresponds the doubled rated load.
OPTIMIZATION OF HYBRID RENEWABLE ENERGY… 321

Table 4 : Climatic Data And Load Data

Temperature in 0C Solar Irradiance in W=m2 Wind speed in m/s Load in Watts
22.2 015 3.2 5400

23.0 157 4.1 4950

25.0 369 5.5 5800

27.1 557 6.9 6400

28.9 690 8.2 6500

30.0 758 7.9 6600

30.8 752 8.4 6400

30.7 691 10.4 6350

30.6 568 11.5 6500

30.0 382 11.5 6500

29.1 188 9.7 6300

27.9 033 7.4 6600

27.0 001 4.8 7800

26.4 000 2.7 9000

26.0 000 1.7 8200

25.6 000 1.5 6500

25.2 000 1.1 5100

24.7 000 1.1 4500

24.3 000 1.4 4000

23.8 000 1.7 3900

23.4 000 1.9 3800

23.1 000 2.2 3850

22.8 000 2.4 4500

22.5 000 2.6 5950

322 VEERESH BYAHATTI, AKSHAYA KATKAR AND A. B. RAJU

Figure 6 : Convergence of GA for rated load

Figure 7 : Convergence of GA for doubled rated load

OPTIMIZATION OF HYBRID RENEWABLE ENERGY… 323

Table 5 : Results Of Parameters

Parameters With rated load With doubled rated load
Number of PV Panels 27 85
Number of Wind Turbines 2 1
Number of Batteries 3 4
Cost of the system (In Indian Rupees) 24.04 Lakhs 40.38 Lakhs
LPSP 0.00874 0.003454

CONCLUSIONS
A methodology of optimizing the HRES using the genetic algorithm is proposed in this
paper. Studies have proved that the genetic algorithm converges very well and the
methodology proposed is feasible for optimization of HRES.

REFERENCES

[1] Koutroulis E., Kolokotsa D., Potirakis A. and Kalaitzakis K., “Methodology for optimal sizing
of stand-alone photovoltaic/windgenerator systems using genetic algorithms”, Solar Energy,
vol.80, pp. 1072-1088, 9, 2006.
[2] Borowy Bogdan S., Salameh Ziyad M., “Methodology forOptimally Sizing the Combination of
a Battery Bank and PV Array in a Wind/PV Hybrid System”, IEEE Trans. Energy Conversion,
vol. 11, no. 2, pp. 367- 375, 1996.
[3] Baoding Liu, Ruiqing Zhao, Monte Carlo Programming and Fussy Programming, Beijing:
Tsinghua University Press, 2001.
[4] Chipperfield Andrew, Fleming Peter, Pohlheim Hartmut, Fonseca Carlos, Genetic Algorithm
Toolbox Users Guild, Department of Automatic Control and Systems Engineering, University of
Sheffield, 1995.
[5] [Link]
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 325-336

ANALYSIS OF DIGITALLY CONTROLLED BUCK

CONVERTER DRIVING RGB LED BASED
LIGHTING SYSTEM

PREETI BALIGA B.1 AND PINTO PIUS A. J.2

1
[Link] Student, Dept. of E&EE,
St. Joseph Engineering College, Mangalore, India.
2
HOD, Dept. of E&EE,
St. Joseph Engineering College, Mangalore, India.

Abstract
Solid State Lighting (SSL) technology is growing and will become more robust than it was in the last
decade. A key challenge to energy sustainability is cutting energy consumption without negatively
impacting our lifestyle. LED lighting clearly offers an eco-friendly alternative to energy-wasting
conventional lighting.
In this paper, analysis of buck converter for RGB LED based lighting system is presented. The
forward characteristics of Red, Green and Blue LED colours indicate the region of control to produce
acceptable level of light intensities. As the intensity of light produced by an LED is proportional to its
forward current, a suitable region is identified to obtain distinctive colours and to explore various
energy saving options. This project works towards designing a simple, portable, efficient and cost-
effective colourful lighting system. The system is simulated in MATLAB Simulink. The ongoing study
includes implementation of the proposed converter system using a Digital Signal Processor (DSP).

-----------------------------------
Keywords: Buck Converter, RGB LED, Current Mode Control (CMC), Pulse Width Modulation
(PWM) © http: //[Link]
326 PREETI BALIGA B.1 AND PINTO PIUS A. J.

1. INTRODUCTION
The lighting technology is undergoing a radical transformation and has provided exciting and
ground-breaking possibilities; all thanks to LED. Although the technology has been around
for quite some time, meaningful innovations have made LED lighting more relevant than ever
before.
Today, highly energy-efficient white LEDs are revolutionizing the domain of LED. Not only
are they aesthetically pleasing and easily controllable, they also have a long lifespan [2] and
reduced maintenance. It also has a powerful effect in office lighting.
The coloured LEDs have launched a new dimension in architectural lighting, stage and
studio lighting and can also be used in traffic light systems. Because coloured LEDs are
highly efficient [3] and controllable they are used for beautification and creating dynamic
ambiences. Therefore the RGB LED (Multi-colour LED) based lighting schemes are useful
in commercial lighting applications. LED lights use 1/10th of the electricity used by
incandescent lights, greatly reducing electricity costs [6] and CO2 emissions. The
lumen/Watt of the LED light source being much higher than any other source [4], the energy
saving concept is getting popularized in all sectors of lighting. RGB LED lighting fixtures
offer endless colour changing possibilities with high brightness to enhance our environment.
In essence, LEDs on the whole serve as replacements for older technologies such as
fluorescent, incandescent, neon and halogen lights [9]. They're also about twice as bright.
These LED lights have no harmful ultraviolet rays.
The block diagram of the control system along with possible disturbances is shown in fig.1.
The LED array consists of a total of 14 LEDs with two in series and seven such in parallel
[10]. According to the datasheet of the LED the forward current is considered to be 350 mA.
Current Mode Control [1] will be adopted to provide control over the entire region of the
diode graph which includes both linear and nonlinear regions which will serve the purpose of
dimming. The duty cycle of the switches of the converters and hence the intensities of the
filaments to produce distinctive colours, will be varied using Pulse Width Modulation
(PWM) [7] which is a technique to dim an LED from 0-100% of its nominal brightness. A PI
controller is used for error compensation.
ANALYSIS OF DIGITALLY CONTROLLED BUCK… 327

Fig.1: Block diagram of proposed LED array control

The rest of this paper is arranged as follows: Section II presents the architecture of the
converter both in open loop and closed loop form. Three converters are needed for each of
the three filaments of one RGB LED. In this paper the case of red colour with a forward
voltage of 2.6 V is demonstrated. In Section III, the two cases of disturbances; supply and
load disturbances are introduced and their influences on the output power of RGB LEDs are
discussed. Finally, Section IV presents the simulation results and waveforms of the study.
The values of L & C are calculated from the equations below,
D = VO/Vi (1)
L crit = ((1-D) R)/2f (2)
C crit = (1-D)/16f2L (3)
Where:
D: Duty Cycle
VO: Output Voltage (V)
Vi: Input Voltage (V)
R: Load Resistance (Ω)
f: Frequency of operation (Hz.)
328 PREETI BALIGA B.1 AND PINTO PIUS A. J.

L: Inductance (H)
L crit: Critical value of Inductance
C crit: Critical value of Capacitance

2. ARCHITECTURE
2.1 Buck Converter
Also known as Step-Down Converter, the buck converter converts its input voltage into a
lower output voltage. When the switch (MOSFET) is closed, inductor current increases,
diode is reverse biased. When the switch is open, inductor current falls, diode is forward
biased to carry inductor current.

Fig. 2: Buck converter with ESR

In practical cases, Equivalent Series Resistance (ESR) of capacitor and winding resistance of
inductor must be considered as in fig. 2. ESR value is inversely proportional to capacitor
value.
ANALYSIS OF DIGITALLY CONTROLLED BUCK… 329

2.2 Open Loop System

The switch is turned on and off, by using PWM pulses. A triangular waveform is compared
with a constant value to generate the PWM pulses.
A constant value block as shown in fig. 3 is used with a range 0-5V. Every time a constant/
reference value say 3 is passed to the block to achieve a constant brightness [5, 8] at the
LED load. With a reference voltage of 3, the duty cycle obtained is 60%. Average output
voltage is 5.5 which is almost equal to 6V.

Fig. 3: Open Loop with PWM control

2.3 Closed Loop System

As current mode control is preferred in this project, a small value of current sense resistor is
added in series with the load. Voltage across it which is proportional to the output current is
taken as feedback. The amplification factor is calculated and found to be 20.4 as shown in fig.
4. The current sense resistor together with the amplification factor constitutes H(s) of the
control system.
330 PREETI BALIGA B.1 AND PINTO PIUS A. J.

Fig. 4: Closed Loop with PID control

3. DISTURBANCES
3.1 Supply Voltage Disturbance
Fig. 5 shows a signal builder block which is used to input different supply voltages. It is seen
that that output current is maintained at a constant value with only spikes at points where
supply voltage changes. Hence inspite of supply disturbance the output current (LED
brightness) is maintained constant.
ANALYSIS OF DIGITALLY CONTROLLED BUCK… 331

Fig. 5: Variation in supply using signal builder block

3.2 Load Disturbance

Fig. 6: Variation in load

When two different values (1 ohm and 4 ohm) of load resistors are applied, with supply
voltage at a constant value, output current remains constant at a value 2.98A. Hence the
control system designed works properly. This is shown in Fig 6.
332 PREETI BALIGA B.1 AND PINTO PIUS A. J.

4. SIMULATION RESULTS
All the simulation results have been recorded to demonstrate the proper working of the
circuits. The main focus of measurement is the output current that has to remain constant at
any given time to ensure constant brightness of LED. The results and waveforms have been
verified based on the literature survey done.
4.1 Buck Converter Waveforms

4.2 Waveforms of Open Loop System

ANALYSIS OF DIGITALLY CONTROLLED BUCK… 333

The above graph indicates output voltage, inductor current and output current in the
respective order.
4.3 Simulation of Closed Loop System

The above graph indicates output voltage, inductor current and output current in the
respective order.

Table 1 : Range of values for reference voltage and output current

Vref (V) Iout (A)

0 0
1 0.49
2 0.98
3 1.47
4 1.96
5 2.45
334 PREETI BALIGA B.1 AND PINTO PIUS A. J.

Table 2 : Design values/ Specifications

Vin (dc source) 10 V

Vout (output) 5.7V
Iout 1.44 A
Vref 3V
T (time period) 0.1m s
L 0.75m H
Winding resistance 0.05 ohm
C 100u F
ESR 0.9 e-2 ohm
VF of 2 diodes in series 5.2 V
R(current sensing resistor) 0.1 ohm
H(s) multiplication factor 20.4
Kp 10(trial and error)
Ki 1000(trial and error)

4.4 Supply Disturbance

ANALYSIS OF DIGITALLY CONTROLLED BUCK… 335

The above graph shows different supply voltages, output voltage, inductor current and output
current in the respective order.
Table 3 : Design values/ Specifications
Vin (dc source) 20V, 10 V
Vout (output) 5.3V

4.5 Load Disturbance

The above graph shows different load resistance values, constant supply voltage, output
voltage, inductor current and output current in the respective order.

CONCLUSION & FUTURE WORK

This paper has presented a buck converter driving an array of nonlinear load. It focuses on
current mode control rather than constant voltage, thus providing constant light output,
controlled brightness and enhanced reliability. The PI controller for the system has been
designed using trial and error method. Simulations of supply disturbance and load
disturbance have been included which indicates that the control system designed is working
336 PREETI BALIGA B.1 AND PINTO PIUS A. J.

satisfactorily. Hardware implementation of the system is an ongoing process which will

include digital control of the converter and load using DSP. Accordingly a multi-coloured
lamp source using RGB LEDs which is the main aim of the project, will be designed
keeping in mind the efficiency and size of the prototype, which could be used in traffic light
systems, for decorative purposes, architectural lighting or industrial panel lighting systems.
An attempt will be made to derive the small signal model of the above proposed system.

REFERENCES

[1] Chattopadhyay Souvik and Das Somshubhra, “A Digital Current-Mode Control Technique for
DC–DC Converters” IEEE Transactions On Power Electronics, Vol. 21, No. 6, November 2006
[2] Broeck Heinz van der, Sauerlander Georg, Wendt Matthias “Power driver topologies and control
schemes for LEDs” IEEE Publication 2007
[3] Calleja A.J., Torres A., Garcia J., Secades [Link], Ribas J., Martínez J Ángel “Evaluation of
Power LEDs Drivers with Supercapacitors and Digital Control” IEEE Publication 2007
[4] Lopez Carlos, Doval Jesu's, Pereira Moises, Perez Sergio, Dios Javier, Lopez Oscar ” DMX512
Controller for High Brightness RGB LED Matrix” IEEE Publication 2007
[5] Tseng S. Y., Lin S. C. and Lin H. C. “LED Backlight Power System with Auto-tuning
Regulation Voltage for LCD Panels” IEEE Publication 2008
[6] Doshi Montu and Zane Regan “Reconfigurable and Fault Tolerant Digital Phase Shifted
Modulator for Luminance Control of LED Light Sources” IEEE Publication 2008
[7] Cha Sanghyun, Park Deukhee, Lee Yuenjoong, Lee Changseok, Choi Joongho, Lee Jaeshin and
Lee Hyobum “AC/DC Converter Free LED Driver for Lightings” IEEE International
Conference on Consumer Electronics (ICCE) 2012
[8] Sun Tai-Ping and Wang Chia-Hung, “Specially Designed Driver Circuits to Stabilize LED Light
Output without a Photodetector” IEEE Transactions on power electronics, Vol. 27, No. 9,
September 2012
[9] Cheung Y. F. and Choi H. W. “Color-Tunable and Phosphor-Free White-Light Multilayered
Light-Emitting Diodes” IEEE Transactions on electron devices, Vol. 60, No. 1, January 2013
[10] Albert T. L. Lee, Johnny K. O. Sin and Philip C. H. Chan,” Scalability of Quasi-Hysteretic
FSM-Based Digitally Controlled Single-Inductor Dual-String Buck LED Driver to Multiple
Strings” IEEE Transactions on power electronics, Vol. 29, No. 1, January 2014
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 337-350

AUTOMATED ELECTRICAL DRIVES

USING EMBEDDED SYSTEM

DHANUSHA PRABHU AND SANGEETA MAHADDALKAR

Dept. of Electrical and Electronics,
Goa College of Engineering, Ponda, India

Abstract
Digitization is a fast growing technology which gives an edge for improvement in the conventional
systems. Electrical drives are used in the applications right from domestic sector to industrial sector,
retrofitting such drives with the embedded systems will definitely be a cost-effective, time and labor
saving option. This paper presents a way to modernize the conventional drives with a better response
and efficiency. For this purpose a PIC microcontroller is considered and simulations done with
MPLAB software are presented. Applications considered are control of pump and coal mills.

-----------------------------------
Keywords : Conventional drive, Microcontroller, Power Electronic Converters (PECs).
© http: //[Link]
338 DHANUSHA PRABHU AND SANGEETA MAHADDALKAR

1. INTRODUCTION
In modern electrical drive system, about 50% of the total energy is consumed by the drives
which may be either constant speed drives or variable speed drives [1]. Electric motors are
around us everywhere. Generators in power plant are connected to a three-phase power grid
of alternating current (AC), pumps in your heating systems, refrigerator and vacuum cleaner
are connected to a single phase ac grid and switched on or off by means of a simple
contactor. In cars a direct current (dc) battery is used to provide power to the starter motor,
windshield wiper motors and other utilities, this motors run on dc and in most cases they are
activated by a relay switch without any control [2].
Electric drives systems have very wide field of applications with some strong points.
 Large power range available: actuators and drives are used in a very range of
applications from wrist watch level to machines at the multi mega-watt level i.e. has
used in coal mines and steel industry.
 Electrical drives are capable of full torque at standstill hence no clutches are
required.
 Electrical drives can provide a very large speed range, usually gearboxes can be
omitted
 Clean operation, no oil spills to be expected.
 Safe operation possible in environments with exclusive fumes (pumps in oil
refineries) [2].
However there are limitations towards the use of conventional electric drives for which a
modernised approach is required. The conventional electric motor drive system with variable
speed is depicted in figure 1. In conventional drives for the speed control a bulky system is
incorporated which is inefficient and inflexible to the change parameters which increases the
power loss as well the cost [3].

Fig 1. Conventional electric motor drive system [3].

AUTOMATED ELECTRICAL DRIVES USING EMBEDDED SYSTEM 339

In modern electric drives power electronic converters with the control unit is employed as in
figure 2 which is compact and flexible as compared to the bulky conventional system.

Fig 2. Modern electric drives employing power electronic converters [4].

The control unit which we are going to discuss in this paper is the PIC microcontroller for
which a brief comparison is done with the dedicated hardware structure as follows:
Flexibility-Microcontroller has ability to reprogram using Flash, EEPROM or EPROM allows
straight forward changes in the control law used.
Circuit Complexity-Microcontroller based Control required less hardware so circuit
Complexity less
Power Consumption-With respect to dedicate hardware microcontroller consume less power
Cost-In Microcontroller based control Cost savings come from several locations.
Development costs are greatly decreased because of the design/flexibility advantages.
Easy to Use-Just program and go! Microcontrollers often only require a single 5V supply as
well which makes them easier to power and use.
Self-Monitoring-Microcontroller based system have self-monitoring and self-checking
feature.
Reliability-Microcontroller based control more reliable than dedicated hardware due to its self
checking capability.
Accuracy-It is digital system so accuracy is very high Testing And debugging time is very
less. Very less time will be required to rectify the problem [5].
This paper is organized as follows: section II briefs about the the control scheme for
modernize drive, section III discuss about the applications to include the control scheme,
section IV and V are results and conclusion drawn respectively.
340 DHANUSHA PRABHU AND SANGEETA MAHADDALKAR

2. CONTROL SCHEME
The control scheme consists of the microcontroller under study and the power electronic
converters. A microcontroller (sometimes abbreviated μC, uC or MCU) is a small
computer on a single integrated circuit containing a processor core, memory, and
programmable input/output peripherals. Program memory in the form of NOR flash or OTP
ROM is also often included on chip, as well as a typically small amount of RAM.
Microcontrollers are designed for embedded applications. Microcontrollers are used in
automatically controlled products and devices, such as automobile engine control systems,
implantable medical devices, remote controls, office machines, appliances, power tools, and
other systems. Microcontroller used here is PIC16F72 [6] this is a basic model which is used
to demonstrate the automation which can be employed in the applications taken up for
discussion. Its features are as follows.
The program memory contains 2K words, which translate 2048 instructions, since each 14-
bit program memory word is the same width as each device instruction. The data memory
(RAM) contains 128 bytes. It has high performance RISC CPU with only 35 single word
instructions to learn. All single cycle instructions except for program branches, which are
two-cycle. The Operating speed is DC - 20 MHz clock input DC 200 ns, 128 x 8 bytes of
Data Memory (RAM). Interrupt capability Eight-level deep hardware stack Direct, Indirect
and Relative Addressing modes. It has High Sink/Source Current: 25 mA, Timer0: 8-bit
timer/counter with 8-bit prescaler, Timer1:16-bit timer/counter with prescaler, can be
incremented during SLEEP via external Crystal/clock, Timer2: 8-bit timer/counter with 8-bit
period register, prescaler and postscaler. Capture, Compare, PWM (CCP) module Capture is
16-bit, max. Resolution is 12.5 ns - Compare is 16 bit, max. Resolution is 200 ns PWM max.
resolution is 10-bit 8- bit, 5-channel analog-to-digital converter Synchronous Serial Port
(SSP) with SPI™ (Master/Slave) and I2C™ (Slave) Brown-out detection circuitry for
Brown-out Reset (BOR) [7].
Now the power electronic converters which drives the motor has several topologies each one
for specific application. The power conversion systems can be classified according to the
type of the input and output power as follows [8]:
AC to DC (rectifier) in Fig. 3
DC to AC (inverter) in Fig. 4
DC to DC (DC-to-DC converter) in Fig. 5
AUTOMATED ELECTRICAL DRIVES USING EMBEDDED SYSTEM 341

AC to AC (AC-to-AC converter) in Fig. 6

Fig. 3 Rectifier [8]

Fig. 4 Inverter [8]

Fig. 5 DC to DC converter [8]

342 DHANUSHA PRABHU AND SANGEETA MAHADDALKAR

Fig. 6 AC to AC converter [8]

3. MICROCONTROLLER APPLICATIONS
Two applications are discussed in this paper 1. Control of DC motor for pump, 2. Control of
Coal processing in coal mines. The hardware and software design of the microcontroller
based control scheme is presented in this section.
3.1 Control of DC motor for pump
Water systems are designed for the “worst case” situations. Most of the time they have
excess capacity. Controlling flow below its maximum saves energy and improves system
operation as shown in figure 7, which is achieved through control unit and power electronic
converters (PECs).

Fig. 7 Difference in energy consumption using a Throttling device v/s Power Electronic
Converters (PEC) [9].
3.2 Hardware Design
The figure 8 shows the block diagram of the speed control system for the dc motor drive. The
microcontroller
AUTOMATED ELECTRICAL DRIVES USING EMBEDDED SYSTEM 343

Fig.8 DC motor control hardware block

unit (MCU) controls the dc motor by processing its controlling parameters that are speed and
current signals. There are two push buttons one for starting the motor and the other for
stopping the motor. Reference speed set is 1500 rpm. There are two push buttons provided
one for increasing speed and other for decreasing speed. Speed changes are in steps of 10
rpm. The PWM section of PIC16F72 will generate PWM signal for desired DC voltage. The
voltage control is achieved by varying the duty cycle of PWM. The optocouplers are used to
isolate the high power circuit from the low power controlling signals. The assembly program
developed for the control purpose includes monitoring overload and under load conditions
accordingly change the duty cycle to generate PWM signals to chopper circuit which in turn
would control the DC motor to run at rated speed.
3.3 Software Design
The microcontroller is being programmed in such a way that the reference parameters are
compared with the analog data collected by the unit through the sensors. This analog data is
then converted to digital by the in built ADC channels of PORTA (RA0, RA1, RA2, RA3 or
RA5). If the sensed quantity is not matched then there is a change in duty cycle from the
referenced duty cycle which is explained in TABLE I. Accordingly the firing signals are
344 DHANUSHA PRABHU AND SANGEETA MAHADDALKAR

generated for the chopper which achieves the speed control of DC motor. The complete
flowchart for the system software design is given in figure 9.
Table 1. Duty cycle variation with speed
Speed Duty cycle CCPR1L CCP1CON
Rated 50% 7C 2C
Above rated 25% 3E 1C
Below rated 75% BA 3C

In this case PWM signal is generated for speed control at 5 kHz frequency with fosc=20 MHz
and Prescalar value
PS = 2 .Formula used to find the time period is given in Equation 1
 f 1 
PR 2   osc    1 ................Equation 1
 f PWM 4  PS 

Fig.9 Flow chart for speed control of DC motor

AUTOMATED ELECTRICAL DRIVES USING EMBEDDED SYSTEM 345

For the conversion of analog-to-digital the ADC channel has to be configured by setting the
last bit of ADCON0, bit addressable register. For ADC conversion ADCON0.2 is set and
cleared once the conversion is complete. RC2 pin is cleared to set as the output pin.
3.4 Control of Coal mill processing

Fig. 10 Coal processing block diagram

Coal preparation is both a science and an art. It deals with taking raw coal and producing a
saleable product that meets contract specifications by removing impurities. Coal processing
steps from extraction of coal to coal usability are shown in Figure 12. After extracting coal
from coal mines it is sent to coal preparation plant where coal handling is done which
includes a) stockpiling, b) stacking and c) reclaiming. Once coal handling is done coal is sent
for sampling under this process coal properties are checked like calorific value, ash, moisture
content etc. Following process is washability where minerals are removed from coal by
various washing processes so that the end product is nearly mineral free [10]. Further
overall top size of the coal is reduced by crushing so that it can be easily handled and
processed in coal preparation plant. Then screening is done which groups the process
particles into certain ranges. After screening moisture from the coal is removed by a process
called dewatering using filters and centrifuges [11]. At the end of dewatering process coal
from coal preparation plant is transported to the customer. Now depending upon the
346 DHANUSHA PRABHU AND SANGEETA MAHADDALKAR

application customer decides whether to send it for further processing or to use it as it is.
Further process is coal milling which uses a coal mill or pulverizer [12]. Coal mill is the
equipment which grinds the coal particles into a fine powdered coal. There are three types of
coal mills a) low speed, b) medium speed and c) high speed.
In this paper Ball and tube mill which is a low speed mill is discussed which is shown in
figure 11.

Fig. 11 Ball and Tube mill [10].

3.5 Hardware Design

Coal mill consists of horizontal rotating cylinder (grinder) which rotates about its axis and
contains steel balls, pebbles or rod inside it. Coal, hot air and cold air is input to the mill
from one end since the grinder is rotating it will grind the coal with steel balls or pebbles and
the pulverized fuel exits from the same end. Output of the pulverized fuel should be
maintained at desired temperature and this is achieved by supplying proper combination of
hot and cold air at the input. Output of the pulverizer is maintained at 75 microns [14]. The
micro-controller has been incorporated for checking the quantity of the input fuel, the
temperature which should be maintained inside the ball mill and for checking the output of
the pulverizer. Figure 12 shows the pin out of the control scheme.
AUTOMATED ELECTRICAL DRIVES USING EMBEDDED SYSTEM 347

Fig. 12 Pin-out of the control scheme.

3.6 Software Design

Complete flowchart for the coal mill control process is shown in the figure. 13. Coal that has
to be processed inside the grinder, temperature inside the grinder and fineness of the
processed coal are the three parameters sensed by the inbuilt ADC. Once the required
quantity of coal is sensed conveyor motor starts and coal is transferred into coal mill. Then
the temperature inside the grinder is checked, if it is lesser than the set value heater is put on
otherwise cooler starts its function.

Fig. 13 Flowchart for coal mills

348 DHANUSHA PRABHU AND SANGEETA MAHADDALKAR

4. RESULTS
Figure 14 shows the result of speed control of dc motor simulations in MPLAB software, it
depicts that as in TABLE I the reference speed is at 50% duty cycle. Initially the motor runs
at 50% duty cycle when the load is suddenly decreased the speed had to proportionally
decrease, the speed and current is sensed and according to reference value a 25% duty cycle
PWM is generated so also once when the load is increased chopper is fired at 75% duty
cycle. Thus in an efficient way the speed control is achieved rather than installing the
throttling valve for speed control which causes frictional losses and adds to power loss.

Fig.14 PWM signal generation

Fig. 15 shows the result for coal preparation process, first graph is output across RB4 pin.
RB4 is high when sensor is sensing for the coal quantity to be equal to capacity of the mill.
Once the set value is reached RB4 goes low and RB0 goes high which depicts the starting of
the conveyor motor which takes the coal to the grinder. Along with the coal, hot and cold air
is also input to the grinder, by supplying it in proper proportion temperature of the grinder is
maintained. Temperature is sensed by second sensor, if temperature is more than the set
value cooler (RB3) is put on otherwise heater (RB2) (which is depicted in graph third and
fourth respectively). After setting the temperature grinder is put on with high on RB1. Third
sensor simultaneously checks for the fineness once the fineness of 75 microns is achieved
grinder is put off with low on RB1.
AUTOMATED ELECTRICAL DRIVES USING EMBEDDED SYSTEM 349

Fig. 15 Output of a coal milling process

CONCLUSION
A brief overview was carried out of electrical drives, the merits and demerits were found. As
a result of demerits the solution of retrofitting with a microcontroller is suggested
eliminating the bulky system in conventional drive. Two applications were studied in which
use of modernized drive to achieve energy and labour saving, cost effective solution and
easy operation.

REFERENCES

[1] Castabnet T. and Nicolai J., "Digital Control for Brush DC Motor," IEEE Transaction On
Industry Application, Vol, 30, No 4, July/August 1994

[2] Fundamentals of electrical drives Available: http:// [Link]

[3] Preeti, Dogra Sandeep, Jain Rashmi,”DC Drives: Microcontroller Based Control” International
Journal o f Engineering and Advanced Technology (IJEAT), Volume-2, Issue-2, December
2012

[4] Introduction to electric drives Available: http:// [Link]/

[5] Microcontroller v/s dedicated hardware Available: http:// [Link]/

[6] Shrivastava Shruti, Rawat Jageshwar and Agrawal Amit, ”Controlling DC Motor using
Microcontroller (PIC16F72) with PWM”, International Journal of Engineering Research,
Volume No.1, Issue No.2, Dec 2012
350 DHANUSHA PRABHU AND SANGEETA MAHADDALKAR

[7] “PIC16F72 data sheet,” Microchip

[8] Power Electronic Available: [Link] [Link] /

[9] Tom Neuberger and Steven B. Weston,”Variable frequency drives: energy savings for pumping
applications,”Eaton Corporation, November 2012

[10] Coal mining process Available:http:// [Link]

[11] Subbarao P M V, “Performance analysis of coal mills”.

[12] Albrecht M. C., “Coal preparation plant”, Kaiser Engineers Inc.

[13] Li Yingde, “Study on Whole Process Quality Control in Coal Production Based on Industry
Engineering”.

[14] Mr. Pujari R.N., Mr. Mohanta G.C., “ Dynamic Operating Regimes of Ball and Tube Mill With
Varying Coal Quality- An Analysis”, NTPC/Talcher Super Thermal Power Station
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 351-364

ANALYSIS OF CASCADE DUAL BUCK INVERTER

AND PHASE SHIFT CONTROL SCHEME FOR
DC TO AC POWER APPLICATION BY USING
PULSE WIDTH MODULATION TECHNIQUE

POOJA CHINIVAR T.S. AND MANJULA. C.

Dept. of Electrical and Electronics Engineering
The Oxford College f Engineering, Bangalore, India

Abstract
This paper introduced, a new type of cascade inverter and phase shift control. Advantages by using
proposed cascade dual buck inverter with the help of using power MOSFETs are high system
reliability, no shoot through worries and lower switching losses. The advanced cascade dual buck
inverter along with phase shift control gives all the merits of dual buck inverters and overcome their
disadvantages. If AC loads are to be supplied, an inverter is required, transforming the DC power
from the Photovoltaic array to AC power at prescribed voltage and frequency. By using Phase shift
control scheme we can theoretically eliminate current zero crossing distortion. By the combination of
both cascade topology and phase shift scheme, we can increase the switching frequency and reduce
current ripple. A cascade dual buck inverter has been designed and tested to illustrate advantages of
the system by comparing single unit dual buck inverter.

------------------------------------
Keywords : Phase shift control scheme, Single dual buck inverter, Cascade inverter.
© http: //[Link]
352 POOJA CHINIVAR T.S. AND MANJULA. C.

1. INTRODUCTION
In high power applications, multilevel voltage source inverters (VSI) are raising as a new
breed of power converters. Multilevel inverter having the characteristics of high efficiency,
high voltage, high power without using transformer. To improve the efficiency, and also to
enhance the performance and reliability, various circuit topologies and different types of
control methods have been proposed.
With modular design, without transformer connection, extended power output and voltage
output, less maintenance and higher fault tolerance, the cascade inverters are having very
good features for low cost utility interface of several renewable sources. In different levels,
there is a voltage unbalance that is the major drawback of using multilevel voltage inverter.
The technique to balance the voltage is capacitor charge control or voltage clamping. The
most commonly used and available multilevel voltage source inverters are the cascade H-
bridge inverter, diode clamped inverter, and flying capacitor inverter [1]-[2]. One of the
advantage by using cascade type inverters are capable of reaching higher output voltage level
by using lower voltage devices and also it has modular design structure which results less
maintenance [2]. Cascade H-bridge multilevel inverter can be demonstrated by using single
DC power source and capacitor. Without using transformer, we can use the battery or fuel cell
stack [3]-[4]. The major failure of VSI is shoot through problem, it occurs only if cascade
inverters are connected in series having two active devices in one leg. By using the power
MOSFETs we can improve the efficiency and also switching speed.
The half bridge or full bridge inverter is a typical voltage source inverter (VSI), it has two
active switches in one phase leg. It requires dead time to avoid shoot through worries between
the switches in one leg. By using dead time effect, we can get distorted output waveform and
there is a great reduction of equivalent transferred energy of pulse-width modulation (PWM).
At some fault conditions, even though by using the topology of dead time schemes, still there
is shoot through problems, that will be the dominant failure of the circuit.
This paper proposes a new type of cascade inverter. It is an efficient and highly reliable. Dual
buck type inverters do not have shoot through worries, but still they are VSI with unique
features and operation principle. It has much system reliability compared to other inverters.
Including this cascade dual buck inverter does not have dead time issues. In this paper phase
shift control plays a very important role, it can be easily implemented with digital controller
and voltage ripple and output current ripple. So by using phase shift control scheme
ANALYSIS OF CASCADE DUAL BUCK INVERTER AND… 353

theoretically we can eliminate the zero crossing distortion in cascade dual buck inverter from
zero to full load. The main aim of this paper is to demonstrate the feasibility of the system by
comparing single unit dual buck. In this paper closed-loop control for cascade dual buck
inverter has been designed and implemented. This paper mainly shows the basic operation
principles and also different methodology of the proposed cascade dual buck inverters.

2. OPERATION PRINCIPLE
Basically single-unit dual buck inverter has two kind’s namely dual buck half bridge inverter
and dual buck full bridge inverter. According to proposed cascade dual buck inverter, further
there is a classification namely cascade dual buck half-bridge inverter, shown in Fig. 1, and
cascade dual buck full-bridge inverter, shown in Fig. 4. The main focus of this paper is to
testing, designing and analyzing the cascade dual buck half bridge inverter to demonstrate the
advantage of cascade dual buck inverter. By using series output concept we can get higher
voltage.
To obtain higher voltage, we are using the control strategy for two dual buck half bridge
inverters in series output fashion [5]. In this strategy, two dual buck inverter shares the same
dc power supply, having two sets of filter inductor and capacitor, in this fashion the
connection is only effective for two units. In this paper, the proposed inverter explains a
different series connection concept and also cascading fashion, for each cell there is a separate
DC power supplies and extended to N unit connection.
2.1 Analysis of dual buck inverter
The proposed cascade dual buck half bridge inverter shown in Fig. 1, In this inverter each unit
simultaneously increases the switching frequency. Main feature of phase shift control is to
reduce the output consisting of two MOSFETs and two diodes. The iN port and (i+1)P port
are connected together of the ith unit and (i+1)th unit respectively. It is having a separate DC
source while each source located with a single phase inverter and AC terminals are connected
in series. It was first invented by Richard .H. Baker.
354 POOJA CHINIVAR T.S. AND MANJULA. C.

Fig.1. cascade dual buck half-bridge inverter.

Fig.2. Full H-bridge Configuration.

According to Fig. (2) we can get three voltage levels, by using single H-Bridge inverter [6].
But the output voltage levels of cascaded full H-Bridge inverter are given by 2n=1 and
voltage step of each level is Vdc/n, where n is the number of H-bridges connected in
cascaded.
The switching table is given in Table (1) and (2)
Table 1. Switching table for H-Bridge
Switches Turn ON Voltage Level
S1,S2 Vdc
S3,S4 -Vdc
S4,D2 0

Table 2. Switching table for Cascaded H-Bridge

Switches TurnOn Voltage Level
S1, S2 Vdc
S1,S2,S5,S6 2Vdc
ANALYSIS OF CASCADE DUAL BUCK INVERTER AND… 355

Switches TurnOn Voltage Level

S4,D2,S8,D6 0
S3,S4 -Vdc
S3,S4,S7,S8 -2Vdc

Even after adding dead time, at some fault condition shoot through problems are occur in the
circuit. By using power MOSFETs so many benefits are their like lower switching loss,
voltage drop, and fast switching speed there by we reduce the ripple current. By using single
dual buck inverter we may face one drawback that is current zero crossing distortion. To over
come this drawback one technique has to be implemented that is PWM technique.
The standard half-bridge or full-bridge must be a typical VSI consisting of two active
switches in one phase leg. Operation modes of single unit dual buck half bridge inverter are
shown in Fig. 3 [5]. In this modes of operation Sip and Dip are a working pair, these working
pair is operating under positive half cycle. Another working pair Sin and Din are operating
under negative half cycle. The standard half-bridge or full-bridge inverter is a typical voltage
source inverter (VSI) with two active switches in one phase leg. To prevent shoot through
problems, it requires dead time, but by using dead time effect there is a disturbance in output
wave form and greatly reduction in PWM
In the single dual buck inverter, we have to switch all the units, if phase shift control is not
adopted. This leads to the zero crossing distortion problem of single unit into cascade
topology. There are two main advantages by using phase shift control namely increasing
switching frequency and reduction in output current ripple. So by using phase shift control
method we can greatly reduce the zero crossing distortion of single unit type and thereby
increase the switching frequency adopted.

Fig. 3. Modes of single-unit dual buck half-bridge inverter.

356 POOJA CHINIVAR T.S. AND MANJULA. C.

In dual buck full bridge inverter, potentially the efficiency can increases because there
is a flow of freewheeling current through the diodes. It also increase the utilization of
rate of input voltage and minimizes the voltage stress of diodes compared to dual buck
half bridge inverter.

Fig. 4. (a) Single-unit dual buck full-bridge inverter.

(b) Cascade dual buck full-bridge inverter.

3. ANALYSIS OF PHASE SHIFT CONTROL

According to single dual buck inverter, selective switch is based on the output current. From
Fig. 3, we can observe the four modes of operation (a) positive current, S1p turned ON (b)
positive current, D1p free-wheeling (c) Negative current, S1n turned ON (d) Negative current,
D1n free-wheeling. The main condition is when i1 is positive, S1p and D1p are the working
pair and when i1 is negative, S1n and D1n are the working pair. By this operation we may
come across one drawback called current zero crossing distortion. By turning on both the
switches S1p and S1n near to the zero crossing distortion, it can be easily eliminate the
drawback which is mentioned above. But this solution is reverse to the operation principle of
the dual buck type inverter. For high reliable and best feature of dual buck inverter it should
not allow to turn on both the active switches at the same time. The original aim of the dual
buck inverter is only one switch can operate at a given time. But as mentioned above both
switches are turning on at the same time, this leads to higher switching loss. Thanks to
cascade topology, by using phase shift control it solves the issues of zero crossing distortion.
By using phase shift control it can be reduce the passive filter component there by increasing
the equivalent switching frequency by N times that of single dual buck inverter. It can be
ANALYSIS OF CASCADE DUAL BUCK INVERTER AND… 357

easily eliminate the current zero crossing distortion theoretically by using phase shifted [7]
PWM.
To demonstrate the phase shift control scheme, we have to analyze the single unit half bridge
inverter and 2-unit cascade half bridge dual buck inverter. Fig. 5 represents the equivalent
circuit of single unit half bridge dual buck inverter, the condition is when S1p is on [8]. Fig. 6
shows the corresponding gate signal of S1p and the current through output inductor i1. In Fig.
6 there is a shaded area corresponds to the operation mode shown by Fig. 5. From the Fig. 5
and Fig. 6 the current ripple of i1 can be derived as follows:
Δ i1 = ((0.5 Vdc − vo) DsTs) / (L1p + Lf) (1)
Where, Ds is the duty cycle of the switch S1p , 0.5 ≤ Ds ≤ 1
and fs is switching frequency( Ts =1/fs).

Fig. 5. Equivalent circuit of single-unit half-bridge dual buck inverter.

Fig. 6 Gate signal of single-unit half-bridge dual buck inverter.

Current ripple at zero crossing region is not zero since Ds is equivalent to 0.5 at zero crossing
period. The switching frequency components are removed by capacitor, and there by the
current io1 gets its average component. Since current averages of both the half cycles are not
zero, there is jump from negative to positive average that is exactly known as current zero
crossing distortion.
358 POOJA CHINIVAR T.S. AND MANJULA. C.

Fig. 7. Equivalent circuit of 2-unit cascade half-bridge dual buck Inverter.

By using cascade dual buck topology we have many advantages that are already explained in
early stage. Addition to that a united control scheme is proposed for the individual voltage
balancing control. Fig. 7 shows the equivalent circuit of the 2-unit cascade half-bridge dual
buck Inverter. Generally the phase shift angle defined as (360º/N), here for 2-unit, the phase
shift angle is 180º. Fig. 8 shows the gate signals and current through output inductor. The
shaded area of Fig. 8 refers to the operation mode shown by Fig. 6.
Δi2 = (0.25 Vdc + 0.25 Vdc − vo )( Ds − 0.5) Ts
L1p + L2p + Lf (2)
If L1p = L2p , (2) can be write it as,
Δi2 = (0.5 Vdc − vo )( Ds − 0.5) Ts
2L1p + Lf (3)

Fig. 8. Gate signals and current i2 through output inductor of 2-unit cascade half-
bridge dual buck inverter.
ANALYSIS OF CASCADE DUAL BUCK INVERTER AND… 359

The process will be similar for getting the equation from 3-unit, 4-unit, 5-unit upto n-unit. The
current ripple for n-unit cascade half bridge inverter is mentioned below:
Δin = (([(n + 1) / 2] / 2n) Vdc − vo ) (Ds −([(n −1)2]/n)) Ts
nLnp + Lf (4)
To obtain the same output voltage, the 2-unit cascade inverter requires two dc sources along
with 0.5Vdc each. At zero crossing period, Ds is very nearer to 0.5 this can be notice from
(4). It can be observe that, there is a great reduction in ripple current. The main concept to
reduce the distortion at zero crossing point is to maintain the even number for n, because
distortion will zero at [(n-1) /2] /n is equal to 0.5. Now it can be conclude that current ripple
reduces to zero when the number of units increases.

4. DESIGN OF CLOSED LOOP SYSTEM

Most of the control system utilizes feedback. The operating frequency of a control system is
fixed. Closed loop is derived to illustrate the advantage of cascade dual buck inverter shown
according to Fig. 9. According to the Fig. 10 the average model of N-unit cascade half bridge
dual buck inverter is shown, dj (j = 1, . . .n) is the duty cycle of each unit and Lj (j =1. .. . , n)
is the output inductor of each unit.
Lj = Ljp : i > 0, Lj = Ljn : i > 0 (5)

Fig. 9. The average model of N-unit cascade half-bridge dual buck inverter.
According to Fig (9) we have the following relation
(d1 + d2 + … + dn ) Vdc/2= d1 + d2 + … + dn NVdc
N 2 (6)
360 POOJA CHINIVAR T.S. AND MANJULA. C.

It is necessary to have equal dj, to ,monitor the power balance of each unit. Hence d1 = d2 =
… = dn = d, (6) can be altered as
(d1 + d2 + … + dn ) Vdc= d . NVdc (7)
2 2
It can be derived the equivalent model of N-unit cascade dual buck inverter as per from (7)
and Fig. 8 which is in Fig. 9:

Fig. 10 The equivalent average model of N-unit cascade half-bridge dual buck inverter.
ΣLj = L1 + L2 + …+ Ln .
From Fig. 10 we can write the equation like
d(t) . NVdc − vo (t) = L di(t) (8)
2 dt
i(s) = 1 (d(s) . NVdc − vo (s)) (9)
sL 2
Transfer functions from duty cycle d to current i :
Gid (s) = i(s) = (NVd c/ 2)
d(s) sL (10)
Transfer functions from voltage vo to current i :
Giv (s) = i(s) = 1 .
vo (s) sL (11)
where, Gid (s) is the control-to-output transfer function
Giv (s) is an uncontrolled feed-forward term
According to Fig. 10, it adopts two loop design, the inner current loop and the outer voltage
loop, if it go through a simple proportion controller GP(s) in inner loop design it can achieve
fast dynamic response. If it go through PR controller GPR(s) in outer voltage loop it can
greatly reduce the steady-state voltage error. In this control diagram, the bandwidth ωc
should be very smaller, so that, we can get higher selective bandwidth
ANALYSIS OF CASCADE DUAL BUCK INVERTER AND… 361

Fig. 11 Control block diagram of N-unit cascade half-bridge dual buck inverter.

To obtain a high loop bandwidth , and to reduce current steady state error current loop is
designed in a dual loop system. For single unit inverter system equivalent dc bus voltage
is Vdc and admittance compensation term is reciprocal of Vdc. But according to Fig. 11,
for cascade dual buck unit inverter system, equivalent dc bus voltage is (NVdc) /2. Hence
the admittance compensation is GAC(s) = 1/ (N Vdc / 2).

Fig. 12. Generation of PWM for all switches of N-unit cascade half-bridge dual buck
inverter with phase-shift control.
362 POOJA CHINIVAR T.S. AND MANJULA. C.

5. EXPERIMENTAL RESULTS

Fig. (a) Gate pulse are shown as below in the PWM generator for all 1-unit, 2-unit, 3-unit
inverter topology.

Fig. (b) The experimental results of output current and output voltage in Single dual buck half
bridge inverter.

Fig. (c) The experimental results of output current and output voltage in Cascade dual buck half
bridge inverter.

Fig. (d) The experimental results of output current and output voltage in Cascade dual buck full
bridge inverter.
ANALYSIS OF CASCADE DUAL BUCK INVERTER AND… 363

Fig. 13. Voltage waveforms across split capacitors for 3-unit cascade dual buck half-bridge
inverter system.

According to Fig. 13, Split capacitors are needed for cascade topology, here the voltage is
balanced naturally. If there is any unbalance by using capacitors, ESR, then voltage balance
compensator might be used. Finally it can conclude that by using cascade dual buck full
bridge inverter voltage unbalance issues can solved.

CONCLUSION
Cascade dual buck inverter having all the merits of traditional cascading inverters, by using
this we can eliminate the shoot through problems and lower switching losses. Mainly by
using phase shift control scheme we can solve the current zero crossing distortion. By the
combination of both cascade topology and phase shift scheme, we can increase the switching
frequency and reduce current ripple. A cascade dual buck inverter has been designed and
tested to illustrate advantages of the system by comparing single unit dual buck inverter. To
get better efficiency and good output, it is better to go for proposed cascade dual buck
inverter, by using this there is a great enhancement of system reliability.

REFERENCES

[1] Lai J.S and Peng F.. “Multilevel converters—A new breed of power converter, “IEEE Trans.
Ind. Electron., vol. 32, no. 3, pp. 509 -517, n May/jun. 1996.

[2] Rodriguez J., Lai J.S., and Peng, “Multilevel Inverters: A survey of topologies, controls, and
applications,” IEEE Trans. Ind Electron., vol. 49, no. 4, pp. 724–738, Aug. 2002.

[3] Lezana P. and Ortiz G., “Extended operation of cascade multicell converters under fault
condition,” IEEE Trans. Ind. Electron., vol. 56, no. 7, pp. 2697–2703, Jul. 2009.
364 POOJA CHINIVAR T.S. AND MANJULA. C.

[4] Ertl H., Kolar J.W., and Zach F. C., “A novel multicell dc–ac converter for applications in
renewable energy systems,” IEEE Trans. Ind. Electron., vol. 49, no. 5, pp. 1048–1057, Oct.
2002.

[5] Liu J. and Yan Y., “A novel hysteresis current controlled dual buck half bridge inverter,” in
Proc. IEEE PESC, 2003, pp. 1615 –1620.

[6] Gupta R., Ghosh A., and Joshi A., “Switching characterization of cascaded multilevel-inverter-
controlled systems,” IEEE Trans. [Link]., vol. 55, no. 3, pp. 1047–1058, Mar. 2008.

[7] Enjeti P. N., Ziogas P. D., and Lindsay J. F., ―Programmed PWM techniques to eliminate
harmonics: A critical evaluation, IEEE Trans. Ind. Appl., vol. 26, no. 2, pp. 302–316, Mar 1990.

[8] Lee J. Y. and Sun Y. Y., ―Adaptive harmonic control in PWM inverters with fluctuating input
voltage,‖ IEEE Trans. Ind. Electron., vol. IE-33, no. 1, pp. 92–98, Feb. 1986.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 365-374

ENERGY HARVESTING FROM SPEED BREAKER USING

PIEZO ELECTRIC MATERIAL

BASANTI. KALADAGI, RAJANI. G. M, SANDHYARANI. D. S.,

SEEMA. BANDI AND SOUMYA. ASANGIMATH
EEE Department, [Link] Institute of Technology, Belgaum

Abstract
Energy is the basic need for the development of the modern world. For meeting up the regular demand
of energy, there is a need to design a system that will produce electricity without destroying the
nature. Researches show that the world has already had its enough shares of its energy resources.
Fossil fuels pollute the environment. Nuclear energy requires careful handling of both raw as well as
waste material. The focus now is shifting more and more towards the renewable sources of energy,
which are essentially, non-polluting. Here is an attempt to show how energy can be produced, stored
and used using the road transport pressure or any kind of pressure. One of the promising electricity
generation methods is by using the piezoelectric material. Strain on piezoelectric material causes
charge separation which produces electricity. The number of vehicles passing over the speed breaker
on roads is increasing day by day. It is possible to tap the energy and generate power by making the
speed breaker as a power generation unit. Instead of taking traffic as a problem we can take it as an
opportunity to produce energy. In addition to energy used to move vehicle forward, part of energy is
wasted on deformation in the asphalt, which is due to weight of vehicle, it can be transformed into
electricity via piezoelectric effect by a road with embedded piezoelectric crystals. The energy is
converted into electricity and transferred to battery on the side of the road. Potential benefits of this
solution are; the vehicle does not waste any additional fuel, there are no moving parts. The generated
power can be used for low and medium voltage applications, such as street lighting (LED), powering
the signal indicator etc near the speed breakers and this will be a great boon for the mankind.
-----------------------------------
Keywords: Piezo electric material, Speed breaker, Rectifier, Battery.
© http: //[Link]
366 BASANTI. KALADAGI, RAJANI. G. M, SANDHYARANI. D. S., SEEMA. BANDI AND SOUMYA. ASANGIMATH

1. INTRODUCTION
The energy crisis is a great bottleneck in the supply of energy resources to an economy. The
recent fluctuations in the price of petroleum and coal have affected world economy which
has enforced increase in the price of other items including food, shelter and clothing. This
shows that we are too dependent on petroleum and coal as sources of electrical power.
Besides, petroleum and coal as a source of electrical energy they have also contributed to
severe air pollution problem. Therefore, some alternative method to produce electricity has
to be put in place. Other solutions which can be explored are nuclear energy, ocean thermal
energy, and hydroelectric power generators. However, these options require huge financial
capacity to run and to maintain. Besides, not many countries are allowed to use nuclear
power generator due to world political scenario. Thus, ambient energy harvesting such as
solar cells and wind turbines have been the popular choices and these renewable energy
sources are gaining more attention. However, they are expensive and not affordable to many
countries to acquire them. As a consequence, other possible energy sources must again be
explored to fight with these problems.
The process of acquiring the energy surrounding a system and converting it into usable
electrical energy is termed as ambient energy harvesting. In ambient energy harvesting one
of the promising options for power generation is by using piezoelectric material. Vibrations
from machines, mechanical stress, and strain from high-pressure motors, manufacturing
machines, and waste rotations can be captured and used as ambient mechanical energy
sources for electricity generation. Piezoelectric material can be used as a mechanism to
transfer ambient vibrations into electrical energy. This energy can be stored and used to
power up electrical and electronics devices. Usually this is used in energy harvesting for low
power and small autonomous devices, such as wireless sensor networks, and portable
electronic equipments. The application of piezoelectric material as a power generator can be
extended to operate daily low power electrical appliances such as tuner, light bulb, mobile
phone, street light and so on. It is recognized that proper arrangement of piezo crystals can
generate enough electrical density that can be stored in a rechargeable battery for later usage.
The search for a clean, reusable source of energy has caused a spike in interest in the
exploration of piezoelectricity. Further research into its uses has significantly increased in
the last few years. Currently, smaller projects have been exploring new applications for
piezoelectric energy. Numerous piezoelectricity nightclubs in Europe have sprung up in
ENERGY HARVESTING FROM SPEED BREAKER USING… 367

recent years, advertising their use of piezoelectricity-charged batteries to power their

establishments. There are also efforts to create mobile energy sources with piezoelectric
fabrics, allowing for energy to be collected from more than just footsteps-- including
heartbeats, ambient noises, and airflow. Furthermore, researchers are experimenting in
harvesting piezoelectricity for bigger projects and could even turn ordinary highways and
roadways into power stations.
The context of this paper is to convert the pressure of vehicles over the speed breaker into
electrical energy, which otherwise would go waste. Since then the idea caught our working
team and decides to develop such a project that will produce more power and can be stored
when it is not necessary to use.
We referred “Analysis of piezoelectric devices” by Jiashi Yang to study the properties,
structure and types of piezoelectric material. There are naturally occurring piezoelectric
crystals like berlinite, sucrose, quartz, salt, topaz and [Link] artificial piezoelectric crystal
like Lead zirconate titanate. The types of piezoelectric materials are hard doped and soft
doped. Hard doped piezoelectric materials are doped with niobium pent oxide which have
high mechanical and electric stress ,and soft doped materials are doped wit nickel and iron
which have high relative permittivity and large electro mechanical coupling factor. We
referred book “Energy harvesting from speed breakers” by Alper Erturk and Daniel [Link]
to study how mechanical energy can be converted to electrical energy.

2. PIEZOELECTRIC MATERIAL
Piezo-electric materials are the crystals that generate electricity when compressed or
vibrated. A piezoelectric material is made up of both positively and negatively charged
particles arranged in such a way that all the positively charged particles and all the negatively
charged particles are grouped about the same central point. If two opposite faces of crystal are
placed under pressure, the crystal can be slightly flattened and distorted, and the charged
particles making up the crystal are pushed together and spread out sideways. The change is
such that the average position of the negatively charged particles shifts slightly with respect to
the average position of the positively charged particles. This means there is in effect, a
separation of positive and negative charges and a potential difference is therefore created
between the two faces of the crystal causing the output voltage and power is directly
proportional to the pressure applied or in other words weight of the vehicle passing through it
368 BASANTI. KALADAGI, RAJANI. G. M, SANDHYARANI. D. S., SEEMA. BANDI AND SOUMYA. ASANGIMATH

and the time the vehicle is passing through it. The energy harvesting via piezoelectricity uses
direct piezoelectric effect. The phenomenon will be clear from the below figure

Figure1
3. EXISTING APPROACH

Figure 2

3.1 Working
 Here the reciprocating motion of the speed breaker is converted into rotary motion using
the rack and pinion arrangement.
 The axis of the pinion is coupled with the sprocket arrangement.
 The axis of the smaller sprocket is coupled to gear arrangement.
 Finally the gear arrangement is coupled with the generator.
ENERGY HARVESTING FROM SPEED BREAKER USING… 369

3.2 Advantages
 Pollution free power generation.
 No manual work required during generation.
 Energy is available all the year.
 No consumption of fossil fuel.
3.3 Disadvantages:
 Bulky in construction.
 Needs maintenance.
 More friction.

4. METHODOLOGY
The piezoelectric phenomenon was first described by the Curie brothers, who observed that
certain materials generate electric current when they are deformed. The piezoelectric effect
exists in two domains: the first is the direct piezoelectric effect that describes the material’s
ability to transform mechanical strain into electrical charge; the second form is the converse
effect, which is the ability to convert an applied electrical potential into mechanical strain
energy. Thus, they can be used as either actuators or sensors.
The direct piezoelectric effect alters mechanical energy into electrical energy by straining a
piezoelectric material. Strain or deformation of a piezoelectric material causes charge
separation across the device, producing an electric field and consequently a voltage drop
proportional to the stress applied. The voltage produced varies with time and strain,
effectively producing an irregular AC signal on the average, which is then converted into
direct current (DC). The converted current is then used to charge a capacitor or a battery,
which can store the energy for later use.
When vehicles are running on the specialized Speed Breaker (With piezoelectric panel). This
will create pressure on the piezoelectric material which is kept under this speed breaker. As a
result, deformation of the material takes place which in turn produces electrical charge
across its surface resulting in alternating voltage. This electricity can be stored by a
rechargeable battery or is used for lighting lamp during night time on the road side. The type
of piezoelectric crystals used here are PZT (lead zirconate titanate).PZT has high mechanical
strength, it is insoluble in water,non-toxic and easily available.
370 BASANTI. KALADAGI, RAJANI. G. M, SANDHYARANI. D. S., SEEMA. BANDI AND SOUMYA. ASANGIMATH

4.1 Circuit Diagram

Figure3
4.2 Working Operation
When stress is applied to a piezo electric material this produces ac voltage .the produced ac
voltage is converted to pulsating dc by bridge rectifier. This voltage is stored into capacitor
which gets charged to a pre-determined value at which switch closes and capacitor
discharges through diode. In this way energy can be stored in the capacitor. And can be
discharged when required.
The units are installed on the road bed and epoxy is used as a filler to permanently affix the
units in place. Approximately 2” (5 cm) of asphalt is overlaid. Saw cut channels to run
electric connections are required to route wiring to the roadside, and these must also be
buried. Because the asphalt layer and road bed are not perfectly rigid, some energy is
absorbed by the surrounding layers of material and therefore a larger unit with force-
multiplying components is required to harvest significant energy.
The average power available on the road due to the moving vehicles can be estimated using
the following equation,
𝑁𝐹𝐷 𝐹𝐷
Pavg= 𝐷𝑇 = ∆𝑇
where N is the number of wheels passing through, D is the deflection , F is the normal force
exerted by the wheel on the road, T is the total time taken by the vehicle to pass by, and DT
is the average time for each wheel to pass by.
Series connection of crystals:
ENERGY HARVESTING FROM SPEED BREAKER USING… 371

Figure4
Piezo crystals sandwiched between wooden plates:

Figure5
4.3 Practical Observation

Figure6
372 BASANTI. KALADAGI, RAJANI. G. M, SANDHYARANI. D. S., SEEMA. BANDI AND SOUMYA. ASANGIMATH

When bike is passed over the sandwiched piezo crystals following result is obtained.

Figure7
Here the weight of the vehicle in kilograms is converted to pressure by using the relation
FORCE=MASS*ACCELERATION......N
Where acceleration is 9.81m/sec2
And to convert force into pressure we need to divide the above equation by area on which
the force is exerted
i.e. PRESSURE=FORCE/AREA.....N/m2 or Pascal

5. RESULTS
Where the following table shows the experimental results:
Table1. Output voltage with respect to weight and pressure.
Sl No. Type of Weight in kg Pressure in Voltage in
Vehicle (of one wheel) N/m2 Volts

1 Bicycle 10 9.146 0.4

2 Bike 100 91.46 1.3
3 Truck 500 457.3 2.16

5.1 Advantages
 Parasitic energy of busy roads can be converted into useful electrical energy.
 Energy can be produced without using fuel.
 Pollution free energy can be produced.
ENERGY HARVESTING FROM SPEED BREAKER USING… 373

5.2 Disadvantages
 Use of asphalt layer increases the cost of installation and operation.
 Time and labour is needed for the installation.
 Maintenance is difficult.

5.3 Applications
 Street lightning.
 In Traffic signals and road side signals.
 In security alarm.
 Energy can be stored in battery and can be used for further applications.

6. FUTURE SCOPE
The proposed work portrays the concept of Piezoelectric Energy Harvesting and the results
obtained after the implementation are very encouraging. Future work of the proposed idea
encompasses further amplification of the crystal output to a greater extent. Future lies in the
inclusion of advanced material used to design the piezoelectric crystal which further
amplifies the crystal output in terms of voltage as well as current. A study could be carried
out from the variety of piezoelectric crystals and after comparing the results, the choice of
the optimum material for the best performing crystal could be devised. More amount of
energy can be harvested by placing piezoelectric material throughout the road and beneath
railway tracks.

CONCLUSION
Energy is the basic need for the development of the modern world. For meeting up the
regular demand of energy we need to design a system that will produce electricity without
destroying the nature. As the number of vehicles passing over the speed breaker in roads is
increasing day by day. It is possible to tap the energy and generate power by making this
specialized speed breaker as a power generation unit. The generated power can be used for
low and medium voltage applications.
374 BASANTI. KALADAGI, RAJANI. G. M, SANDHYARANI. D. S., SEEMA. BANDI AND SOUMYA. ASANGIMATH

REFERENCES

[1] Shakun Srivastava, Ankitasthana, “Produce electricity by the use of speed breakers”
Journal of Engineering Research and Studies, Vol.2, No.1 April-Jun 2011.
[2] [Link] Islam1, Syed Khalid Rahman2, Jakeya sultana Jyoti3, “International Journal of
Engineering Science and Innovative Technology (IJESIT)” Volume 2, Issue 3, May 2013
[3] Fraden J., “Handbook of Modern Sensors: Physics, Designs, and Applications”(Springer, 2004).
Proceedings of the 4th National Conference; INDIACom-2010.
[4] Sodano HA and Inman DJ “A review of power harvesting from vibration using piezoelectric
materials.”2004
[6] Arjun A.M., Ajay Sampath, Sandhya Thiyagarajan, and Arvind V “A Novel Approach to
Recycle Energy Using Piezoelectric crystal” International Journal of Environmental Science and
Development, Vol. 2, No. 6, December 2011
[7] “Large scale vibration energy harvesting” Journal of intelligent material systems and structures
2013
[8] Imran Patel “Ceramic Based Intelligent Piezoelectric Energy Harvesting Device” British
University in Egypt, Suez Desert Road.
[9] Donald L. Halvorsen “Piezoelectric Polymer Axle Sensors”National Traffic Data Acquisition
Conference Albuquerque, New Mexico,May 1996
[10] Andriopoulou Symeon “a review on energy harvesting from roads” E. Lefeuvre, D. Audigier, C.
Richard and D. Guyomar IEEE Trans. Power Electron. 22, 2018 (2007)
[11] Micka¨el Lallart,_ Shashank Priya, Scott Bressers and Daniel J. Inman “Small-scale
Piezoelectric Energy Harvesting Devices Using
[12] Low-energy-density Sources” Journal of the Korean Physical Society, Vol. 57, No. 4, October
2010
[13] Faruk Yildiz, ”Potential Ambient Energy-Harvesting Sources and Techniques”, The Journal of
Technology Studies., Sensors Peterborough NH (2009),vol. 7288, March 2009, pp. 1-7.
[14] Henry A. Sodano, Daniel J. Inman and Gyuhae Park, ”A Review of Power Harvesting from
Vibration using Piezoelectric Materials”, Sage Publications ,The Shock and Vibration Digest,
Vol. 36, May 2004, pp. 197-205.
[15] Singh, U.K. and Middleton, R.H., Piezoelectric power scavenging of mechanical vibration
energy. Australian Mining Technology Conference: Smart Technologies for Overcoming the
Skills Shortage, Swan Valley, WA, 2-4 October 2007, pp. 111-118
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 375-388

PEAK CURRENT MODE CONTROL FOR PHASE

SHIFTED FULL BRIDGE ZERO VOLTAGE
TRANSITION DC-DC CONVERTER

MANASA H S1, CHAITHRA L2, SURYANARAYANA K3

AND N M RENUKAPPA4
1,4
Dept. of Electronics and Communication Engineering,
SJCE, Mysore, India
2
HEXMOTO Controls Pvt. Ltd, Mysore, India
3
Dept. of Electrical and Electronics engineering,
NMAM Institute of Technology, Karkala, India

Abstract
This paper deals with the modeling and design of a peak current mode control for phase shifted full
bridge zero voltage transition (PSFB ZVT) dc-dc converter. Peak current mode control provides
cycle-by-cycle current limiting, transformer’s flux balancing and inherent input voltage feed-forward.
The peak current mode control is developed using the transformers primary current. A digital type 2
compensator is used for the implementation of the peak current mode control in digital signal
controller. A 1.2kW converter is designed and developed for cathodic protection of gas pipelines and
the experimental observations are verified against simulation results.

----------------------------------
Keywords : Full Bridge, ZVT converter, phase shifted 2p2z compensator.
© http: //[Link]
376 MANASA H S, CHAITHRA L, SURYANARAYANA K AND N M RENUKAPPA

1. INTRODUCTION
The phase shifted full bridge zero voltage transition dc-dc converter topologies are widely
used in many applications because of their combined hard switching and soft switching
technologies. These converters use transformer’s primary leakage inductance and switch
parasitic capacitance of the converter circuit to achieve zero voltage switching (ZVS). The
ZVS action reduces the switching losses of the converter. Along with the advantage of
reduced switching loss, the PSFB ZVT converter come up with other benefits like high
efficiency due to combined switching techniques, moderate device stress due to ZVS action
and ease of design. The main drawback of these types of converters is phase shift controller
design [1].
The dead beat controller proposed in [2] suggests achieving fast response, as the controlled
variable reaches reference in one or two switching periods. This controller requires a prior
knowledge of loss of duty cycle interval in order to implement. The loss of duty cycle interval
depends upon the converter parameters and load, possibly making the converter analysis
difficult. An average current mode control or double loop control algorithm in [3] suggests a
reduced output current ripple characteristics in order to increase the battery life.
This paper presents the modeling and control of phase shifted full bridge zero voltage
transition converter using transformer’s primary current for peak current mode. The output
voltage is regulated to the desired level using peak current mode control (PCMC). This
control algorithm provides benefits of cycle by cycle current limiting, inherent input voltage
feed forward [4, 5]. The control algorithm is used with a digital type 2 compensator. The
implementation of peak current mode control for PSFB ZVT converter requires pulse width
modulated waveform generation with precise timing control. In PCMC, the power
semiconductor switch is turned ON by a fixed clock frequency and is turned OFF when the
current feedback reaches the reference defined by the voltage loop. The inductor current or
the primary current in PSFB ZVT converter reaches its peak twice in every switching cycle.
So, peak current is detected in every half switching cycle.
The small signal model proposed by V Vlatkovic et al [6, 7] for the phase shifted full bridge
converter is used to obtain the transfer function required for control circuit design of the
PSFB ZVT converter with peak current mode. The designed converter control mechanism is
simulated using MATLAB/SIMULINK and is compared with the experimental observations.
The operating principle of PSFB ZVT converter is discussed in section II. The detail of
PEAK CURRENT MODE CONTROL FOR PHASE SHIFTED… 377

control loop design is given in Section III. Section IV gives the simulation results and section
V presents the obtained experimental observations. The last part of this paper, section VI
gives the conclusion.

2. OPERATING PRINCIPLE
The phase shifted full bridge zero voltage transition converters employs four power
semiconductor switches forming a full bridge inverter circuit, a diode bridge rectifier circuit, a
high frequency transformer, and an output filter circuit comprising of inductor and capacitor.
The circuit diagram of PSFB ZVT converter using IGBT’s is shown in Figure 1. The
corresponding waveform of primary voltage and primary current is shown in Figure 2. All the
switches in the PSFB ZVT converter operate with a fixed 50% duty cycle. The gate signals of
the left leg switches𝑀1 and 𝑀4 are fixed and complementary, whereas for the right leg 𝑀2 and
𝑀3 , they are phase shifted with respect to the former. This provides zero voltage switching
operation for all switches during turn ON. This zero voltage switching or soft switching
operation provides high efficiency with reduced switching losses and stresses, and it also
eliminates the need of snubber circuit at primary side.
The operation of the PSFB ZVT converter is explained with the following modes.

Lo

M1 D1 C1 M3 C3
D3 DR1 DR3

Lr Co
Vdc Rload
RESR

M4 D4 C4 M2 C2 DR4 DR2
D2

Figure 1: Circuit diagram of PSFB ZVT converter

378 MANASA H S, CHAITHRA L, SURYANARAYANA K AND N M RENUKAPPA

Mg1
t

Mg2
t

Mg3
t

Mg4
t
Vdc

Vpri
D t

Ip(slope) Deff
-Vdc
Ipk
Ip2
Ip1

Ipri
to t1 t2 t3 t4 t5 t6 t7 t8 t9 t10 t

t r
M1 M2 D3 M3 M3 D1
M2 D4 D4 M4 D1 D2

Figure 2: Switching waveform and transformers primary voltage

and current waveform

2.1 Power transfer mode (𝒕𝟎 − 𝒕𝟏 )

When two of the diagonal switches,𝑀1 and 𝑀2 or 𝑀3 and 𝑀4 are ON, the power transfer
takes place. The secondary diodes 𝐷𝑅1 and 𝐷𝑅2 or 𝐷𝑅3 and 𝐷𝑅4 are conducting during this
period. Considering 𝑀1 and 𝑀2 are turned ON in the interval 𝑡0 and 𝑡1 , the slope of the
primary current𝐼𝑝(𝑠𝑙𝑜𝑝𝑒) is calculated using (1). It depends on the output filter inductance and
the leakage inductance. The leakage inductance is small compared to the output filter
inductor. The ON time or power transfer interval time depends on the phase shift between the
two switches 𝑀1 𝑀4 and 𝑀3 𝑀2 . Considering the gate signals of top two switches 𝑀1 and 𝑀3 ,
a 100% phase shift results in 100% duty cycle value and vice-versa.
𝑉𝑑𝑐 − 𝑉𝑜′
𝐼𝑝(𝑠𝑙𝑜𝑝𝑒) = (1)
𝐿′𝑜
where, 𝑉𝑑𝑐 is the input voltage, 𝑉𝑜′ is the primary referred output voltage, 𝐿′𝑜 is the primary
referred output filter inductor.
PEAK CURRENT MODE CONTROL FOR PHASE SHIFTED… 379

2.2 Left Leg Transition mode (𝒕𝟏 – 𝒕𝟐 )

The mode starts with the turning OFF of switch 𝑀1 . In this mode the primary current charges
capacitor 𝐶1 and discharges capacitor𝐶4 . The energy required to charge 𝐶1 and discharge 𝐶4 is
provided by the energy stored in leakage inductor and output filter inductor. This stored
energy is large in output filter inductor, hence the capacitors are assumed to be charged and
discharged linearly. The mode lasts for short time [6] equal to 𝛿𝑡
4𝐶𝑚𝑜𝑠 𝑉𝑑𝑐
𝛿𝑡 = (2)
𝐼𝑝𝑘
where, 𝐶𝑚𝑜𝑠 = 𝐶1 + 𝐶4, and 𝐼𝑝𝑘 is the peak value of the transformer’s primary current.
2.3 Freewheeling mode (𝒕𝟐 − 𝒕𝟑 )
Once 𝐶4 is discharged, the current in the circuit freewheels through the switch 𝑀2 and the
diode 𝐷4 of switch 𝑀4 .The current during this period reduces the primary current to𝐼𝑝2 with
−𝑉𝑜′
the slope equal to 𝐿′𝑜
. The primary voltage across transformer is zero.

2.4 Right Leg Transition Mode (𝒕𝟑 − 𝒕𝟒 )

This mode starts with the turn OFF of switch𝑀2 , the charging and discharging of parasitic
capacitors𝐶2 and 𝐶3 take place, accordingly. The energy required to charge 𝐶2 and discharge
𝐶3 is provided by the stored energy in the leakage inductor𝐿𝑙𝑘 . This interval lasts for short
duration of time equal to 𝛿𝑟 as in (3). The secondary voltage is zero since all the diodes of
secondary are conducting and the primary voltage drops down from 0 to – 𝑉𝑑𝑐 .
𝜋
𝛿𝑟 = √𝐿𝑙𝑘 (𝐶2 + 𝐶3 ) (3)
2
Since the voltage across switch 𝑀3 is zero, this switch is turned ON with ZVS. The resonant
circuit consisting of leakage inductance and parasitic capacitance provides ZVS. The
condition to achieve ZVS is given in (4)
1 2 1 2
𝐿𝑙𝑘 𝐼𝑝𝑟𝑖 ≥ (𝐶2 + 𝐶3 )𝑉𝑑𝑐 (4)
2 2
Here, 𝐿𝑙𝑘 is the leakage inductance of transformer primary winding. 𝐶2 𝑎𝑛𝑑 𝐶3are the
parasitic capacitors of power semiconductor switches 𝑀2 𝑎𝑛𝑑 𝑀3 .𝐼𝑝𝑟𝑖 is the transformer
primary current and 𝑉𝑑𝑐 is the input voltage. When the stored energy in the leakage inductor is
greater than the energy required to charge and discharge the parasitic junction capacitors of
380 MANASA H S, CHAITHRA L, SURYANARAYANA K AND N M RENUKAPPA

the switches, the energy charges the capacitor 𝐶2 of switch 𝑀2 and discharges the capacitor
𝐶3 of switch 𝑀3 . Thus, 𝑀3 is turned ON with zero voltage across it.
Table 1: Switching modes of PSFB ZVT converter at different time interval
Time IGBT switches along with its parasitic
Modes of Rectifier diode
interv capacitors
Operation
al 𝐌𝟏 𝐌𝟐 𝐌𝟑 𝐌𝟒 𝐃𝐫𝟏 𝐃𝐫𝟐 𝐃𝐫𝟑 𝐃𝐫𝟒
Power Transfer t0
ON ON OFF OFF ON ON OFF OFF
Mode − t1
Left leg t1 C4
C1 charges ON OFF ON ON OFF OFF
transition − t2 discharges
Freewheeling t2
OFF ON OFF D4 ON ON ON OFF OFF
Mode − t3
Right leg t3 C2 C3
OFF D4 ON ON ON ON ON
Transition − t4 charges discharges
t4
Duty Loss Mode OFF OFF D3 ON D4 ON ON ON ON ON
− t5
Power Transfer t5
OFF OFF ON ON OFF OFF ON ON
mode − t6
Left Leg t6 C4
C1 discharges
OFF ON OFF OFF ON ON
Transition − t7 charges
Freewheeling t7
D1 ON OFF ON OFF OFF OFF ON ON
Mode − t8
Right leg t8
D1 ON C2 discharges
C3 charges OFF ON ON ON ON
Transition − t9
t9
Duty Loss Mode D1 ON D2 ON OFF OFF ON ON ON ON
− t10

2.5 Duty Loss mode(𝒕𝟒 − 𝒕𝟓 ):

In this mode the diode 𝐷3 of switch𝑀3 and diode 𝐷4 of switch 𝑀4 are conducting in one half
cycle and diode 𝐷2of switch 𝑀2 and diode 𝐷1of switch 𝑀1 are conducting in other half cycle.
The loss of duty value depends upon the transformers primary leakage inductance, external
resonance inductance(𝐿𝑟 )and external resonance capacitance (𝐶𝑟 )connected to the circuit.
The transition time 𝑡𝑟 is calculated using (5)
PEAK CURRENT MODE CONTROL FOR PHASE SHIFTED… 381

𝑇𝑟
𝑡𝑟 = (5)
4
where, the resonance period 𝑇𝑟 = 1/(2𝜋√𝐿𝑟 𝐶𝑟 )
Equation (6) gives the effective duty cycle 𝐷𝑒𝑓𝑓 of the converter
𝐷𝑒𝑓𝑓 = 𝐷 − ∆𝐷 (6)
The loss of duty cycle is expressed as (7) [6],
𝐼𝑝1 + 𝐼𝑝2
∆𝐷 = 𝑉𝑑𝑐 (7)
𝑇
𝐿𝑙𝑘 𝑠

The output voltage in steady state is expressed as (8)

𝑁𝑠
𝑉𝑜 = 𝑉 𝐷 (8)
𝑁𝑝 𝑑𝑐 𝑒𝑓𝑓
The sequence will be repeated in the second half cycle. The operating modes of the PSFB
ZVT converter are given in Table 1.

3. DESIGN OF CONTROL LOOP

The modeling of the PSFB ZVT converter can be done by using state space averaging method
[8] or by PWM switch model method. Implementation of state space averaging method
requires the derivation of state equations for each mode and later taking the average.
Therefore, it would be better to use PWM switch model [9] to derive the small signal model
for the converter circuit. Small signal model proposed in [7, 10] is used to derive the transfer
function of the PSFB ZVT converter. The small signal model is shown in Figure 3.
The open loop control to output transfer function is given in (9).
𝑣̂𝑜 (𝑠) 𝑛𝑉𝐷𝐶 𝑅(1 + 𝑠𝐶𝑅𝐶 )
𝐺̂𝑣𝑑 = = 2 (9)
̂
𝑑(𝑠) 𝑠 𝐿𝐶(𝑅 + 𝑅𝐶 ) + 𝑠(𝐿 + 𝑅𝑅𝐶 𝐶) + 𝑅 + 𝑅𝑑 [1 + 𝑠𝐶(𝑅 + 𝑅𝐶 )]
The open loop control to inductor current transfer function is as in (10),
𝑖̂(𝑠) 𝑛𝑉𝐷𝐶 (1 + 𝑠𝐶(𝑅 + 𝑅𝐶 ))
𝐺̂𝑖𝑑 = = 2 (10)
𝑑̂(𝑠) 𝑠 𝐿𝐶(𝑅 + 𝑅𝐶 ) + 𝑠(𝐿 + 𝑅𝑅𝐶 𝐶) + 𝑅 + 𝑅𝑑 [1 + 𝑠𝐶(𝑅 + 𝑅𝐶 )]
Equation (9) and (10) facilitates to derive the closed loop control to output transfer function
with peak current mode control and is expressed as (11)
𝐴𝑖 = 𝐾𝐹𝑝 (𝑠)𝐹ℎ (𝑠) (11)
382 MANASA H S, CHAITHRA L, SURYANARAYANA K AND N M RENUKAPPA

where,
𝑅 2 𝑇𝑠 𝑚(1 − 𝐷)𝑅𝑑
𝐾 =1+ ( )+ (12)
𝐿 𝑇𝑠 (𝑚(1 − 𝐷) − 0.5) 2𝐿
𝑅(1 + 𝑠𝐶𝑅𝑐 )
𝐹𝑝 (𝑠) = 𝑇𝑠 𝑚(1−𝐷)𝑅𝑑 (13)
1 + 𝑠𝐾𝐶𝑅 (1 + 2𝐿
)

𝑠 𝑠2
𝐹ℎ (𝑠) = 1 + + 2 (14)
𝜔𝑛 𝑄 𝜔𝑛

̂𝒊 + 𝒅
̂ 𝒏𝑽𝑫𝑪(𝒅 ̂ 𝒗) Lo
𝒏𝑽𝑫𝑪𝒅
+

+
-

Co
𝒏𝑽𝑫𝑪 𝒏𝑽𝑫𝑪
̂ 𝒗 ) Rload
̂𝒊 + 𝒅
(𝒅
̂ 𝑫𝑪
𝒏𝑽 ̂
𝒅 𝑹
𝑹 RESR

1:Deff

Figure 3: Small signal model of the PSFB ZVT

converter

Lo

M1 D1 C1 M3 C3
D3 DR1 DR3

Lr Co
Vdc Rload
Hv
High RESR
frequency
Transformer
M4 D4 C4 M2 C2 DR4 DR2
D2

M1 M2 M3 M4 Ri

Gate Drive -
Logic Verr
+ + Gcz -
+
-

External ramp Vref

Ve(pk-pk)

Figure 4: The PSFB ZVT converter with peak current mode

control waveform of the transformer
Equation (12) is the gain of the system,(13) is the transfer function of power stage of the
converter and (14) is high frequency transfer function.
Figure 4 shows the PSFB ZVT converter system with peak current mode control, consisting
of outer voltage loop and inner current loop. The transformer primary current is used as
feedback current. In peak current mode control, the measured output voltage is compared with
the reference voltage, generating an error voltage. The generated error voltage is compensated
with compensator to obtain a current reference. The current reference is then compared with
PEAK CURRENT MODE CONTROL FOR PHASE SHIFTED… 383

the measured primary current. The signal GENERATED is given to the gate drive logic. The
gate drive logic enables turn ON and turn OFF of the power semiconductor switches. When
peak current is detected, it turns OFF the corresponding switch 𝑀3 𝑜𝑟 𝑀2 while turning ON
its complementary switch 𝑀2 𝑜𝑟 𝑀3 of the right leg pair. This gives the phase shifted signals
for the right leg pair 𝑀2 and 𝑀3 with respect to the left leg switches 𝑀1 and𝑀4 . The left leg
switches𝑀1 and 𝑀4 are operated with fixed 50% duty cycle. Both switch leg pairs are given
sufficient dead time in order to avoid the shoot through of IGBT switches during turn ON.
To avoid sub-harmonic oscillations at duty ratio greater than 50 %, slope compensation is
required. It comprises of addition of an external ramp to the compensator output. Addition of
external ramp, modifies the naturally sampled gain 𝐹𝑚 as given in (15)
2
𝐹𝑚 = (15)
𝑚𝑆𝑛 𝑇𝑠
𝑇𝑠 is the switching period, 𝑆𝑛 is the on slope of the primary [Link] value of 𝑚 is equal
𝑆
to(1 + 𝑆𝑒 ). The natural frequency𝜔𝑛 and damping factor 𝑄in (14) are expressed as
𝑛

𝜔𝑛 = 2𝜋⁄𝑇 (16)
𝑠
1
𝑄=
𝜋(𝑚(1 − 𝐷) − 0.5) (17)
Equating 𝑄 = 1 for a critically damped system, and substituting the value of 𝑚, the value of
the external ramp𝑆𝑒 is calculated.
1 1
𝑆𝑒 = 𝑆𝑛 (( + 0.5) − 1) (18)
𝜋 1−𝐷
The peak-to-peak value of external ramp 𝑉𝑒(𝑝𝑘−𝑝𝑘) is calculated using (19).
𝑇𝑠
𝑉𝑒(𝑝𝑘−𝑝𝑘) = 𝑆𝑒 ( ) (19)
2
4. SIMULATION RESULTS
The parameters of the converter under full load condition are as follows:
Input voltage 𝑉𝑑𝑐 70 to124V
Output voltage 𝑉𝑜 24V
Load 𝑅 0.48Ω
Switching frequency𝑓𝑠 20KHz
384 MANASA H S, CHAITHRA L, SURYANARAYANA K AND N M RENUKAPPA

Output filter inductor𝐿𝑜 24μH

Output Filter Capacitor𝐶𝑜 4100μF
Equivalent series resistor 𝑅𝑒𝑠𝑟 2.6mΩ
Resonant Inductor𝐿𝑟 8.22μH
Resonant capacitor 𝐶𝑟 44nF
current sense gain 𝑅𝑖 0.0187V/A
Transformer turns ratio𝑁𝑝 /𝑁𝑠 7:4
A type 2 compensator (𝐺𝑐) is used in the voltage loop with an integrator. The location of
poles and zero of the type 2 compensator is such that the zero of the compensator is placed to
achieve suitable phase margin and a pole to coincide with ESR zero of (10).
For the control loop design the crossover frequency is chosen as 1/7th of the switching
frequency. The obtained phase margin is 114 degrees at a gain crossover of 446Hz. The gain
margin is 18.2dB. The closed loop bode plot of the system with the transfer function of power
stage with peak current mode control is given in Figure 5.
The closed loop transfer function with peak current mode control is calculated using (20)
𝐺𝑐 𝐴𝑖 𝐻𝑣
𝐺𝐶(𝑐𝑙𝑜𝑠𝑒𝑑) = (20)
1 + 𝐺𝑐 𝐴𝑖 𝐻𝑣

where, 𝐻𝑣 is the voltage feedback gain.𝐺𝑐 is the type 2 compensator transfer function in s
domain. The discrete form is obtained by using bilinear transformation to the continuous time
domain transfer function. This gives a digital type 2 compensator also termed as two pole two
zero (2p2z) compensator with transfer function 𝐺𝑐𝑧 expressed as (21),
𝐵0 𝑍 0 + 𝐵1 𝑍 −1 + 𝐵2 𝑍 −2
𝐺𝑐𝑧 = (21)
𝐴0 𝑍 0 − 𝐴1 𝑍 −1 − 𝐴2 𝑍 −2
The step response of the closed loop compensated system is shown in Figure 6. It can be seen
the time taken by the system to reach its steady state is 8 milliseconds. Figure 7 shows the
primary voltage and primary current waveform of the transformer.
A. Step Response due to Change in Load:
PEAK CURRENT MODE CONTROL FOR PHASE SHIFTED… 385

Figure 5: Frequency Response of the uncompensated (blue)

compensated (red) of the PSFB ZVT converter

Figure 6: Step Response of the compensated PSFB ZVT

converter

In this case a load transient from 100 % to 30 % load is simulated using SIMULINK. The
output voltage under this condition is maintained to 24V using peak current mode control.
This is shown in Figure 8.
386 MANASA H S, CHAITHRA L, SURYANARAYANA K AND N M RENUKAPPA

Figure 7: Transformer Primary voltage and current

waveform obtained in MATLAB/SIMULINK software

Figure 8: Output voltage for different load condition

Figure 9: Transformer Primary voltage (blue) and

current (yellow) waveform at 40% load.
PEAK CURRENT MODE CONTROL FOR PHASE SHIFTED… 387

Figure 10: Load Voltage (blue) waveform and Load

current waveform (yellow)

Experimental Verification
A hardware setup is developed and tested for the mentioned specifications. A digital signal
controller, MC56F84789 by Freescale having CPU clock frequency of 100MHz is used for
the digital control of the converter. Figure 9 shows the waveform of the transformer’s primary
current and primary voltage at 40% load. Figure 10 shows the load voltage and current
waveform. The output voltage is maintained at constant 24 V with lower ripple voltage of
0.7% of full load.

CONCLUSION
In this paper, the peak current mode control for phase shifted full bridge converter is
described. A 1.2 kW converter is designed and the stability of the system with peak current
mode control is verified. The control algorithm maintains a constant output voltage of 24V
during load variation. The overall output voltage ripple and output current ripple is less than
1%.
388 MANASA H S, CHAITHRA L, SURYANARAYANA K AND N M RENUKAPPA

REFERENCES

[1] Richard Red1, Laszlo Balogh, David W. Edwards, “Optimum ZVS Full-Bridge DC/DC
Converter with PWM Phase-Shift Control: Analysis, Design Considerations, and Experimental
Results”, Applied Power Electronics Conference Page(s):159 – 165 vol.1 and Exposition, 1994,
APEC `94, Conference Proceedings 1994, Ninth Annual.
[2] Cleber Zanatta, José Renes Pinheiro,“Dead-Beat Current Mode Control Scheme For The Zero-
Voltage-Switching Phase-Shift-Modulated Full-Bridge Dc-Dc Converter”,
[3] Tae-Hoon Kim, Seung-Jun Lee, and Woojin Choi, “Design and Control of the Phase Shift Full
Bridge Converter for the On-board Battery Charger of Electric Forklifts”, Power Electronics and
ECCE Asia (ICPE & ECCE), 2011 IEEE 8th International Conference Page(s):2709 – 2716
[4] “Digital Peak Current Mode Control With Slope Compensation Using the TMS320F2803x”,
Texas Instruments, April 2012
[5] “Implementing peak current mode control of a switch-mode power supply with a single
microcontroller” – Brett Novak, Texas Instruments.
[6] J. A. Sabate, V. Vlatkovic, R. B. Ridley, F. C. Lee, and B. H. Cho, “Design consideration for
high-voltage high-power full-bridge zero voltage-switched PWM converter,” in Proc. IEEE
APEC, 1990, pp. 275-284.
[7] V. Vlatkovic, J. A. Sabate, R. B. Ridley, F. C. Lee, and B. H. Cho, “Small-signal analysis of the
phase-shifted PWM converter,” IEEE Trans. Power Electronics., vol.7, no.1, pp.128-135, Jan
1992.
[8] Ping Wang, Can Liu, Lin Guo, “Modeling and Simulation of Full-bridge Series Resonant
Converter Based on Generalized State Space Averaging, “Proceedings of the 2nd International
Conference on Computer Science and Electronics Engineering (ICCSEE 2013).
[9] Vatché Vorpérian, “Simplified Analysis of PWM Converters Using The Model of The PWM
Switch, Parts I (CCM) and II (DCM) ”, Transactions on Aerospace and Electronics Systems, Vol.
26, N°3, May 1990.
[10] Xu-xiaofeng, “Small-Signal Model for Current Mode Control Full-bridge Phase-shifted ZVS
Converter
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 389-405

REVIEWING VOLTAGE PROFILE, POWER & REACTIVE

POWER COMPENSATION IN EHVAC & UHVAC LINE AT
POWER & FRACTIONAL FREQUENC

B S MOHAN AND SHANKARALINGAPPA C. B

EEE Department
Acharya Institute of Technology, Bangalore, India

Abstract
This paper analysis the voltage profile Power transfer capacity, active & reactive power flow in line
at 50Hz and 50/3Hz. It’s known that higher the transmission voltage lower are the voltage losses and
degree of compensation [1]. In this paper, 765kV and 1200kV levels are taken as transmission
voltages for a line length of 1200Km. For each transmission voltage load is varied from 0% to 150%
of surge impedance load to measure foresaid parameters. While studying the parameters at fractional
frequency transmission system (ffts), loads are connected to 50Hz system through step-up
cycloconverter.

----------------------------------
Keywords : Voltage profile, reactive power, fractional frequency transmission system (ffts),
frequency converters, long-distance power transfer. © http: //[Link]
390 B S MOHAN AND SHANKARALINGAPPA C. B

1. INTRODUCTION
Increasing transmission distance and capacity is always the motivation to advance power
industry technologies. In the history of the ac transmission system, increasing distance and
capacity mainly depends on raising voltage level of transmission lines. At present, the
highest voltage level of the ac power transmission line in India is 765 kV. To further
upgrade, the voltage level encounters difficulties of material and environment issues.
The flexiable ac transmission system (FACTS) has been used to improve power system
performance and has become a very hot research field [2]. The FACTS exploits power
electronic techniques to regulate the parameters of the ac transmission, which can raise
transmission capacity to some degree.
Power generated is transferred through 1200Km transmission length at 765kV and 1200kV
voltage level. Standard operating frequency is either 50Hz/60Hz. In India and other parts of
world 50Hz is being set as standard operating frequency. In this paper voltage profile at
receiving end, active & reactive power is noted down at 50Hz and 50/3Hz.
The structure of this paper is as follows. Section II discusses power transmission system.
Section III discusses the model of transmission system at 50Hz. Section IV presents the
principle and structure of FFTS, generation. Section V discusses the model of transmission
system at 50/3Hz and Result analysis and Section VI Conclusion.

2. POWER TRANSMISSION SYSTEM

As Power resources near the country's major load centers become exhausted, the
transmission system will enlarge its interconnection as part of the long range expansion
strategy. The energy resources to be exploited in the near future are located, where there is
great hydroelectric potential or any other type of resources. Major power markets on the
other hand are expected to be in the nearby region. The distances between the generating
plants and the main load centers ranging from thousands of km. Therefore a great challenge
will have to be faced in future, namely the transmission of large amounts f energy from very
remote sources.
Generally speaking, there are three factors limiting transmission capability, i.e., the thermal
limit, stability limit, and voltage drop limit. For the long-distance ac transmission, the
thermal limitation is not a significant impediment. Its load ability mainly depends on the
stability limit and voltage drop limit [3].
391

The stability limit of an ac transmission line can be approximately evaluated by

Pmax= V2/X (1)
where is the normal voltage, and is the reactance of the transmission line. We can see from
the above equation that transmission capacity is proportional to the square of the normal
voltage and inversely proportional to the reactance of the transmission line. The voltage drop
% can be evaluated by
∆V= (QX/V2)*100 (2)
where is the reactive power flow of transmission line. Thus, the voltage drop is inversely
proportional to the square of voltage and proportional to the reactance of the transmission
line. Therefore, in order to raise transmission capability, we can either increase the voltage
level or decrease the reactance of the transmission line. The reactance is proportional to
power frequency f.
X= 2πfL (3)
where is the total inductance of the transmission line. Hence, decreasing the electricity
frequency can proportionally increase transmission capability. This paper analysis the
voltage drop and power transfer capability of transmission system at 765kV and 1200kV for
1200Km at 50Hz and 50/3Hz.

3. MODEL OF TRANSMISSION SYSTEM AT 50HZ

Transmission system, for EHVAC & UHVAC .i.e. 765kV and 1200kV transmission voltages
at 50Hz is built in matlab for which the power needs to be transmitted at 1200 Km. Since this
paper is evaluating the transmission system alone, the required transmission voltage is
simulated without incorporating a generator and a step-up transformer.
Fig.1 shows the matlab model for 765kV transmission system.

Fig.1 Matlab model for 765kV transmission system

392 B S MOHAN AND SHANKARALINGAPPA C. B

Surge impedance loading at this voltage is 2186.4 MW. Three phase voltage source in
modeled such that the output voltage 765kV and operating frequency being 50Hz.
Transmission line parameters for 765kV line are entered [1] in per Km and the transmission
line length is modeled to 1200Km. Three phase load is varied from no-load to 150% of surge
impedance loading (SIL) in sets of 25%, and the receiving end voltage, active and reactive
power across load are noted for [Link].2 shows the matlab model for 1200kV,
1200Km transmission system below.

Fig.2 Matlab model for 1200kV transmission system

SIL at this voltage is 5333.3MW [4]. The procedure is repeated as mentioned above in 765kV
transmission model. Load is varied from 0%-150% of SIL and the parameters are noted
down. Table 1 shows the receiving end voltage profile for 765kV and 1200kV transmission
voltage, Active and reactive power at receiving end of line in pre unit.

Table.1 Measured voltage, active & reactive power at receiving end

50Hz 0pu .25pu .5pu .75pu 1pu 1.25pu 1.50pu
765kv 5 1.58 0.86 0.5875 .445 0.36 0.3
1200kv 10.585 1.746 0.9 0.6 0.5 0.4 0.35
P 13.86e-9 0.8061 0.4774 0.3341 0.2562 0.2076 0.1744
Q 8.0039e-9 0.4654 0.2756 0.1928 0.1479 0.1198 0.1007
P 2.7038 0.97759 0.5074 0.3415 0.2572 0.2063 0.1722
Q 1.5609 0.5644 0.2929 0.1982 0.1485 0.1191 0.0994

It can be observed from above table that the receiving end voltage is 5 times 765kV for load
being zero pu, and the voltage is 1.58pu for .25pu load of SIL and as load increases the
receiving end voltage profile decreases. For 1.50pu load receiving end voltage is .3pu. The
393

active and reactive power at receiving end for .25pu load is .8061pu and .4654pu in case of
1200kV system its .977pu and .5644pu.
It can be observed from table.1 that voltage profile is better when compared to 765kV system
at .25pu of SIL. Receiving end voltage is 1.58pu in case of 765kV system and its 1.746pu in
case of 1200kV system. At 1pu of SIL receiving end voltage is .44pu in case of 765kV system
and its .5pu in case of 1200kV system. Thus it can be said that the voltage drop in case of
UHAV line is lesser that 765kV system. Thus as we increase transmission voltage or upgrade
it to UHVAC system the voltage drop is reduced and the active power at receiving end is
increased.

4. PRINCIPLE AND STRUCTURE OF FFTS

The fractional frequency transmission system FFTS is a very promising long distance
transmission approach, which uses lower frequency (50/3) to reduce the electrical length of
the AC power line, and thus its transmission capacity can be increased several fold [5].
The AC electricity supplied by utilities has two basic parameters: voltage and frequency.
After the transformer was invented, different voltage levels could be used flexibly in
generating, transmitting, and consuming electricity to guarantee efficiency for different
segments of the power systems. In the history of electrical transmission, besides of 50_60 Hz,
many frequencies were used, such as 25, 50/3and 133 Hz. In 1896, the first two generators
and the transmission line from Niagara to Buffalo, NY were put into the operation A 25 Hz
electric system had been chosen as the winning.
The principle of FFTS can also view from another perspective. It is well known that the
velocity of electricity transmission is approximately equal to the light velocity, 300000 km/s.
When electricity frequency is 50Hz, the wavelength is 6000km; for 50/3Hz, the wave length
enlarges to 18000km. Thus, when frequency is 50 Hz, a transmission line of 1200km
corresponds to one fifteenth of the wave length. Therefore, the “ELECTRICAL LENGTH”
decreases to one third. This is the essential reason why the FFTS can increase transmission
capability several fold and remarkably improve is performance.
Therefore, in order to raise transmission capability, we can either increase the voltage level or
decrease the reactance of the transmission line. The reactance is proportional to power
frequency f, X=2πfl
394 B S MOHAN AND SHANKARALINGAPPA C. B

Where, L is the total inductance of the transmission line. Hence, decreasing the electricity
frequency can proportionally increase transmission capability. The FFTS uses fractional
frequency to reduce the reactance of the transmission system; thus, its transmission capacity
can be increased several fold. For instance, when frequency is 50/3Hz, the theoretically
transmission capability can be raised three times.
The frequency changer is the key equipment in FFTS, which can be either the saturable
transformer [7] or the power electronics ac-ac frequency changer, such as the cycloconverter
[8].
The basic structure of FFTS is illustrated in fig.2a. The hydropower generator in the figure
generates ac power of fractional frequency (say 50/3 Hz), which is then stepped up by a
transformer and transmitted to the receiving end of the transmission line where the fractional
frequency ac power is stepped up to the industrial frequency.

Fig.2a Structure of FFTS

Point A is the 50/3 generation point, point B is step-up transmission voltage, point B to C is
transmission line, point C-D is three phase step-up cycloconverter, point D-E is step-down
transformer and further to point E is load.
This paper analysis the parameters like, voltage profile, active and reactive power at receiving
end, thus system is modeled such that 765kV and 1200kV are modeled at operating frequency
50/3Hz and power is transferred to a distance of 1200Km. At the end of 1200Km a three
phase step-up cycloconverter is placed to step-up the frequency to 3 times the source
frequency and load being connected to other side of cycloconverter.
4.1 Step-up cycloconverter- single phase
Single phase step-up cycloconverter, which converter 50/3 Hz to 50Hz is realized using
power electronics devices. Using IGBT/GTO/MOSFET in cycloconverter to facilitate
frequency conversion, it’s important to check the output voltage across the load after
conversion and drop across each power electronics devices. Thus a matlab model is simulated
to compare the voltage drop and output voltage across load for IGBT, GTO and MOSFET are
tabled in table2.
395

It can be observed from table 2 that output voltage across load for GTO is much better
compared to other gates. Fig.3 shows the matlab model for single phase step-up
cycloconverter, source 100V at 50/5Hz [6]. Fig.4 shows the voltage source at 50/3HZ and
gate pulses to step-up the source frequency to 50Hz. Load is connected to operate at 50Hz.
Fig.5 shows the voltage source and output across load for IGBT, GTO and MOSFET
respectively.

Fig.3 Single-phase to single-phase cycloconverter

Fig.4 Gate pulses and source voltage at 50/3Hz.

396 B S MOHAN AND SHANKARALINGAPPA C. B

Fig.5 Source voltage at 50/3Hz and output across load for IGBT,GTO and Mosfet
Table.2 Output voltage across load for different power electronic devices
100 V at 50/3Hz
Power Electronics Device Voltage across Load
IGBT 34V
GTO 90V
MOSFET 6V

5. THE MODEL OF TRANSMISSION SYSTEM AT 50/3HZ

In section IV, single phase to single phase step-up cycloconverter is been discussed. Three
phase to three phase step-up cycloconverter is being discussed under this section. 3phase to
3phase bridge cycloconverter, also called a 6-pulse cycloconverter or a 36-GTO based
cycloconverter [6]. Fig.6 shows the block diagram for a 6pulse step-up cycloconverter. There
are two modes of operation either Blocked Mode and Circulating Current Mode[7]. The
output is more efficient in case of Blocked Mode operation. Here when the load current is
positive, the positive converter supplies the required voltage and the negative converter is
disabled. On the other hand, when the load current is negative, then the negative converter
supplies the required voltage and the positive converter is blocked. This operation is called
the blocked mode operation, and the cycloconverter using this approach are called blocking
mode cycloconverter.
397

Fig.6 6-pulse, 3-phase to 3 phase cycloconverter

Fig.6 employs thirty six thyristors and is called 6-pulse, 3-phase to 3 phase cycloconverter. In
this circuit, each phase group consists of a 3-phase dual converter. The load phases, shown in
Fig.3, must not be interconnected. If it is done, then the positive group of one output phase
and negative group of other output phase would be joined together without load impedance
which is undesirable.
Since the voltage source and transmission line parameter only change for modeling the
transmission system for both the transmission voltages fig.7 shows the block diagram for
step-up 6-pulse 3 phase to 3 phase step-up cycloconverter.
5.1 Phase A operation
The output frequency is three times higher than that of the input frequency. The model input
and output wave forms are given below. From figure the time axis is separated by electrical
degrees with respect to input voltage, in Fig.8. Phase A operation is tabulated in table.3 [8]

Table .3 Phase A operation of step-up cycloconverter

Phase A Gate Most positive Most negative 50Hz o/p

pulses in degree phase phase cycle

0-30 C-3 B-5 Positive

398 B S MOHAN AND SHANKARALINGAPPA C. B

Phase A Gate Most positive Most negative 50Hz o/p

pulses in degree phase phase cycle

30-60 A-1 B-5 Positive

60-90 A-10 B-8 Negative
90-120 A-10 C-9 Negative
120-150 A-1 C-6 Positive
150-180 B-2 C-6 Positive
180-210 B-11 C-9 Negative
210-240 B-11 A-7 Negative
240-270 B-2 A-4 Positive
270-300 C-3 A-4 Positive
300-330 C-12 A-7 Negative
330-360 C-12 B-8 Negative

Fig.7 Matlab model for cycloconverter at receiving end of transmission line to synchronize
with grid frequency
399

Fig.8 Phase input Voltage Model wave form for phase – A

5.2 Generation of Gate Pulse
Using signal builder [9] to generate square wave for trigger GTO gates; this requires Two
square waves with different duty cycles. Fig.9 shows these square waves used for gate pulse
generation. Fig.10 shows the required gate pulses for Phase A 50Hz output from a 50/3Hz
source.
Gate pulses generated with signal builder ‘signal 4’ and ‘signal 2’ are tabulated in table.4
below.
Table.4 Signal builder data for generating required gate pulses
Signal 4 Signal 2
Frequency 50/3 Frequency 50/3
Amplitude 1 Amplitude 1
Offset 1 offset 1
Duty cycle 8.33% Duty cycle 16.67%
400 B S MOHAN AND SHANKARALINGAPPA C. B

Fig.9 Output from signal builder for generation of gate pulses

Output of signal builder along with delay signal block the required gate pulses are generated
for Phase A conversion.

Fig.10 Gate pulses required for phase A operation for 50Hz

401

Fig.11 Gate pulses to GTO Phase A only

Fig 11,12and 13 shows gate pulses required for GTO to step-up input frequency, three times
the source frequency.
The output at load terminal is set to operate at 50Hz. The output of cycloconverter is shown
in below fig 14.
Similarly, gate pulses are generated to trigger GTO respectively to convert the 3 time the
source frequency to get the required Phase B and Phase C conversions.
The output for Phase B and Phase C are shown in fig.15 and fig.16 respectively. Three phase
to three step-up cycloconverter output voltage is shown in Fig.17.

Fig.12 Gate pulses to GTO Phase B only

402 B S MOHAN AND SHANKARALINGAPPA C. B

Fig.13 Gate pulses to GTO Phase C only

Fig.14. Cycloconverter output for Phase A

403

Fig.15 Cycloconverter output for Phase B

Fig.16 Cycloconverter output for Phase C

404 B S MOHAN AND SHANKARALINGAPPA C. B

Fig.17 Output of 6-pulse 3phase to 3phase cycloconverter

Table.5 Voltage profile at receiving end of cycloconverter

50/3Hz .25 .75 1.25 1.50
0pu .5pu 1pu
pu pu pu pu
765kv 2.28 1.44 1.03 0.81 0.66 0.56 0.49
1200kv 2.17 1.09 1.11 0.94 0.83 0.74 0.65

Thus the SIL for transmission voltages are varied in steps of 25% to 150% and there voltage
profile are tabulated in table 5.
It can be observed for table 5 that the no-load voltage across cycloconverter is reduced to
2.28per unit in case of 765kV and 2.17per unit in case of 1200kV system. The voltage
profile is flatter in case of 1200kV system compared to 765kV system when power is been
transferred at fractional frequency transmission system and converting the frequency to three
times the source frequency through step-up cycloconverter.

CONCLUSION
From table1 we get the voltage profile of the transmission system without compensated for
the reactive power flow in line for varying the Surge impedance loading from zero percent to
150 percent in steps of 25% for both transmission voltages.
405

Table 5 provides the voltage profile for a transmission system at fractional frequency .i.e.
50/3Hz for a line length of 1200Km, the frequency is stepped up thrice the source frequency
.i.e. to 50Hz using GTO as the power electronics device through step-up cycloconverter.
Voltage profile is noted down by varying Surge impedance load in steps of 25% from 0% to
150%.
Comparing voltage profile for 50Hz and 50/5Hz system from table.1 and table.5 it can be said
that voltage profile at receiving end of transmission line is more flatter by reducing the
transmission frequency at sending end side and using a step-up cycloconverter at receiving
end to synchronize it to grid frequency without using any compensation system in
transmission line.
Total active power and reactive power consumed at reduced frequency needs to be calculated
and compared. Compensating for the reactive power consumed by line for both 50Hz and
50/3Hz system, and tabulating the total volt-ampere required to have a flat voltage profile has
to be carried out.

REFERENCES

[1] Shankaralingappa C. B and Suresh. H. Jangamashetti, Senior member, IEEE." FACTS

Controllers to Improve Voltage Profile and Enhancement of Line Loadability in EHV Long
Transmission Lines"
[2] Nan Qin Student Member, IEEE, Shi You, Zhao Xu, Member, IEEE and Vladislav Akhmatov,"
Offshore Wind Farm Connection with Low Frequency AC Transmission Technology "
[3] Wang Xifan, Senior Member, IEEE, Cao Chengjun, and Zhou Zhichao" Experiment on
Fractional Frequency Transmission System" IEEE TRANSACTIONS ON POWER SYSTEMS,
VOL. 21, NO. 1, FEBRUARY 2006
[4] Wang X. and Wang X.,“Feasibility study of fractional frequency transmission system,”IEEE
Trans. Power Syst., vol. 11, no. 2, pp. 962–967, May 1996
[5] Kovalev V ., Panibratets A., Volkova O., Koryavin A., "T HE EQUIPMENT FOR THE AC
1150 kV TRANSMISSION LINE" International Workshop on UH V AC Transmission
Technology, April 25-28, 2005
[6] Hao Chen, Michael H. Johnson, Student Member, IEEE, and Dionysios C. Aliprantis, Senior
Member, IEEE," Low-Frequency AC Transmission for Offshore Wind Power"IEEE
TRANSACTIONS ON POWER DELIVERY
[7] Rashid M. H., 2001, Power Electronics Handbook. New York: Academic, Burak Ozpineci, Leon
M. Tolbert, Department of Electrical and Computer Engineering,
[8] University of Tennessee-Knoxville" Cycloconverters" B. K. Bose, 2002, “Modern Power
Electronics and AC Drives”, Englewood Cliffs, NJ: Prentice-Hall, 2002a
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 407-418

THREE PHASE INTERLEAVED BUCK CONVERTER

FOR REDUCED BATTERY CURRENT RIPPLE

VISHWAS K1, CHAITHRA L2 AND SURYANARAYANA K3

1
Dept. of Electronics and Communication Engineering, SJCE Mysore, India
2
HEXMOTO Controls Pvt. Ltd., Mysore, India
3
Dept. of Electrical and Electronics Engineering, NMAMIT Nitte, Karkala, India

Abstract
The ripple content in the charging current of battery has significant role in determining the battery life.
This paper involves analysis of three phase buck converter in battery charging for current ripple
reduction. The converter is analyzed for constant battery current charging in boost charging mode.
Control loop design for boost charging is described. Simulation results and practical observations are
presented for a 22kW battery charger employing three phase buck converter.

----------------------------------
Keywords : Buck converter, average current mode control, ripple, multiphase, compensator.
© http: //[Link]
408 VISHWAS K, CHAITHRA L AND SURYANARAYANA K

1. INTRODUCTION
Buck converter is the commonly used topology for obtaining a non-isolated output voltage
lower than the input voltage. For low power applications, a single phase converter works
good. For applications demanding high current, the power dissipation in the semiconductor
switches has a noticeable impact on the efficiency. The stress on the switch can be reduced
by paralleling the converters. If the gate pulses to the switches are in phase, then the current
ripple content in the input and output of the converter is high. Large filters have to be used to
maintain desired voltage regulation. This problem can be addressed by using multiphase
topology wherein the gate signals are interleaved [3][4].
A full bridge rectifier followed by buck converter is used for battery charging from AC
mains provided that the average rectifier output voltage is greater than the battery voltage.
To minimize the ripple in the rectifier output voltage, a large capacitor is required. But for
large power requirements, the ripple cannot be removed completely. This causes a 300Hz
ripple on the battery side voltage and current for a 50Hz three phase AC input. For a single
phase input, a second order harmonic component appears in battery voltage and current. The
battery life is affected by overcharge, temperature, current ripple [1]. The current ripple and
the internal impedance of the battery generate a reactive power [2]. This reactive power
results in an increase of battery temperature, which reduces the battery lifetime.
Passive power filters (PPFs), active power filters (APFs), proportional resonant controller
(PR) and digital algorithms are proposed in literature to reduce harmonics in grid connected
converters. PPFs and APFs have the disadvantage of extra hardware requirements, high
operational cost, grid frequency dependent performance [2].
In this paper, three phase interleaved buck converter for battery charging from mains supply
is proposed. Rectified input to the converter is provided from three phase 50Hz AC supply.
Multiphase topology results in reduced input current ripple, size reduction in filter
components, reduced stress on power switches and improved performance [3][4][5]. A
feedforward control is used to reduce the 6th harmonic content of the AC input in charging
current. The system immunity towards electromagnetic interference is improved with
reduced current ripple. The operating principle of multiphase buck converter is discussed in
section 2. Small signal modeling of three phase buck converter is described in section 3.
Converter control for battery charging is described in section 4 followed by simulation and
experimental results in section 5 and conclusion in section 6.
THREE PHASE INTERLEAVED BUCK CONVERTER… 409

2. MULTIPHASE BUCK CONVERTER

Multiphase converters use the concept of interleaving, where the power switches of the
paralleled power stages are switched with fixed delay in gate pulses. A three phase buck
converter is shown in Fig.1. For a three phase converter, the delay is 1200 or 1/3rd of the
switching period. A three phase buck converter has three modes of operation depending
upon the duty cycle of the converter
Mode 1: 0 < 𝐷 < 1/3
Mode 2: 1/3< 𝐷 < 2/3
Mode 3: 2/3 < 𝐷 < 1

T1 T2 T3

L1 iL1

L2 iL2 iL i
Vin Cin +
L3 iL3 ic
+
Rc
R v
D1 D2 D3 vc
C
_
_

Fig. 1. Three Phase Buck Converter

MODE 1
G1
t
G2
t
G3
t
MODE 2
G1
t

G2
t

G3 t
MODE 3

G1
t

G2 t

G3 t
Ts/3 2Ts/3 Ts

Fig. 2. Gate pulses for Mode 1, 2 and 3 operation

410 VISHWAS K, CHAITHRA L AND SURYANARAYANA K

G1

t
G2

G3

t
iL3 iL2 iL1
Inductor
Currents

t
iL
Load
Current

Ts t

Fig. 3. Inductor Current and load current for Mode 3 operation

Mode 1 operation is similar to a single phase buck converter operating at thrice the switching
frequency. In Mode 2 operation, one or two switches are conducting at any instant. Mode 3
is characterized by two or three switches conducting at any instant. Gate pulses for the three
modes are shown in Fig.2. Mode 1 and Mode 2 are subsets of Mode 3. Analysis of Mode 3
will be sufficient to obtain the model of the three phase buck converter.
For the three phase buck converter in Fig.1, the three switches T1,T2 and T3 are turned ON
with a phase delay of 120 degrees or 1/3rd of switching period. When the switch is turned
ON, the inductor current increases in that phase and falls linearly when the switch is turned
OFF. Volt-second balance is achieved during steady state and the peak and valley current in
the inductor are same in every switching cycle. The load current is same as sum of current of
three phases. The ripple in current before output capacitor is thrice the switching frequency
as in Fig.3. The inductor currents and load current waveforms without output capacitor in
Fig.3 are shown for Mode 3 operation. . The ripple current approaches zero for duty cycle of
1/3 and 2/3.

3. SMALL SIGNAL MODEL OF THREE PHASE BUCK CONVERTER

State space averaging technique and PWM switch model are the two widely used techniques
for converter modeling. The small signal transfer function obtained by modeling is required
THREE PHASE INTERLEAVED BUCK CONVERTER… 411

to study the converter dynamics. These transfer functions are also a basis for designing
feedback compensators.
State space technique is used for obtaining mathematical model of three phase buck
converter in this paper. Equivalent series resistance (ESR) of the output capacitor is
considered in the model since it has an important role in the converter dynamics. The
inductance value in each phase is same and DC resistance of the inductor is neglected. The
four independent energy storing elements viz. three inductors and an output capacitor imply
that there are four state variables involved in the modeling. The state variables are the
inductor currents of three phases and the output capacitor voltage. One switching cycle will
have six instants over which the state space average of the state variable is to be taken.
The inductor voltage when the switch of phase 1 is turned ON and OFF is given by (1) and
(2) respectively. Equation (1) and (2) hold good for phase 2 and phase 3.

𝑑𝑖𝐿1 (𝑡)
𝐿 = 𝑉𝑖𝑛 − 𝑣(𝑡) when T1 is ON (1)
𝑑𝑡
𝑑𝑖𝐿1 (𝑡) (2)
𝐿 = 𝑣(𝑡) when T1 is OFF
𝑑𝑡
The capacitor current at any instant is given by
𝑑𝑣𝐶 (𝑡)
𝐶 = 𝑖(𝑡) − 𝑖𝐿1 (𝑡) − 𝑖𝐿2 (𝑡) − 𝑖𝐿3 (𝑡) (3)
𝑑𝑡
The capacitor voltage and the output voltage are related by
𝑑𝑣𝑐 (𝑡)
𝑣(𝑡) = 𝑣𝑐 (𝑡) + 𝑅𝐶 𝐶 (4)
𝑑𝑡

Equations (1) to (4) are used to obtain the state matrices for the six time durations. To obtain
the complete behavior of the three phase boost converter, the state space average [3][7] is
taken over one switching cycle. To obtain the DC gain of the converter, the differential terms
in the averaged state space model are set to zero. The dynamic behavior is obtained by
adding small perturbations in (5) to the averaged state variables as
〈𝐷’〉 = 𝐷’ + 𝑑̂1
〈𝑣〉 = 𝑉 − 𝑣̂
〈𝑖𝐿1 〉 = 𝐼𝐿1 + 𝑖̂𝐿1 (5)
〈𝑖𝐿2 〉 = 𝐼𝐿2 + 𝑖̂𝐿2
〈𝑖𝐿3 〉 = 𝐼𝐿3 + 𝑖̂𝐿3
412 VISHWAS K, CHAITHRA L AND SURYANARAYANA K

Variables with a cap denote small signal perturbations and the ones in 〈 〉 denote the
averaged variables[3]. Adding the small signal perturbations and neglecting the second order
non-linear terms, the small signal transfer functions are obtained.
The output voltage to duty cycle transfer function is given by
𝑣̂(𝑠) 𝑉𝑖𝑛 𝑅(1 + 𝑠𝐶𝑅𝐶 )
= 2 (6)
̂
𝑑 (𝑠) 𝑠 𝐿𝐶𝑅 + 𝑠(𝐿 + 3𝑅𝑅𝐶 𝐶) + 3𝑅

The inductor current of each phase to duty cycle transfer function is given in (7) where
n=1,2,3 denotes the phase.
𝑖̂𝐿𝑛 (𝑠) 𝑉𝑖𝑛 (1 + 𝑠𝐶(𝑅 + 𝑅𝐶 ))
= 2 (7)
𝑑̂ (𝑠) 𝑠 𝐿𝐶𝑅 + 𝑠(𝐿 + 3𝑅𝑅𝐶 𝐶) + 3𝑅

4. THREE PHASE BUCK CONVERTER IN BATTERY CHARGING

A three phase buck converter in a battery charging from grid supply is shown in Fig.4. The
three phase AC supply from grid is rectified by a three phase diode bridge rectifier. The
rectified AC is filtered using capacitors. Three phase buck converter at the rectifier output
will bring down rectified output to battery charging voltage level.

T1 T2 T3

L1 iL1

L2 iL2 iL
Cin +
L3 iL3

Cout R Vo
D1 D2 D3

Fig.4. Three phase buck converter in battery charging

Initially the battery is charged with constant current reference. When the battery voltage
reaches float voltage level, a constant voltage is maintained across the battery with constant
reference voltage.
4.1 Constant Current Mode of Charging
Average current mode control is used for buck converter control in both constant current
mode and constant voltage mode. With current mode control, current balancing in the three
THREE PHASE INTERLEAVED BUCK CONVERTER… 413

phases is ensured and stress on each switch is uniform. For constant current charging, two
current loops are involved as in Fig.5. The outer loop involves battery charging current.

SWR1 + G1
Ich + Gc
iref -+ Gci
- +
- -
Comparator

kf iL1 SWR2 + G2
iout
-+ Gci -
Comparator
Vin
GFF iL2 SWR3 + G3
Vo -+ Gci -
SWR Vp Comparator

Ts iL3

Fig.5. Control diagram of three phase buck converter in constant current mode
for battery charging

The charging current Iout is compared with the set current reference Ich. The error is
amplified by a type II compensator Gc. The compensator output is subtracted by feedforward
gain Kf. This gain is dependent on the instantaneous converter input and output voltage and
is given by
𝑣𝑂
𝑣𝑂 (1 − ⁄𝑣𝑖𝑛 )
𝐾𝑓 = (8)
2𝐹𝑆 𝐿
In (8), 𝑣𝑂 and 𝑣𝑖𝑛 are the instantaneous output and input voltages respectively. The output
from sum block Iref is the reference for the inner current loop formed by inductor current
feedbacks. The reference is same for the loops of three phases. The current reference (i ref) is
compared with inductor current and the error is amplified by the current compensator G Ci.
The output of GCi is the control signal for comparator. The gate pulses are generated by
comparison of control voltage and a sawtooth ramp (SWR). The ramp has a period T S and
peak to peak voltage of VP. For other phases, the current reference is same whereas sawtooth
ramp SWR2 and SWR3 are phase shifted from SWR1 by 120 degrees and 240 degrees (or
Ts/3 and 2Ts/3) respectively.
4.2 Constant Voltage Mode of Charging
For constant voltage charging, average current mode control with outer voltage loop is used.
The control block diagram is shown in Fig.6. The float voltage value is given as reference to
414 VISHWAS K, CHAITHRA L AND SURYANARAYANA K

the voltage loop. The reference value is compared with the instantaneous output voltage and
the error is amplified by a type II compensator. The compensator output is the reference for
current loop as in constant current mode.
The current compensator GCi is a PI controller with a pole at origin and a zero at power stage
resonant frequency of (7).

SWR1 + d1
Vfloat iref
+ Gcv -+ Gci -
- Comparator

iL1 SWR2 + d2
Vo
-+ Gci -
SWR Comparator

Vp iL2 SWR3 + d3
Ts -+ Gci -
Comparator

iL3
Fig.6. Control diagram of three phase buck converter in constant
voltage mode for battery charging

The voltage compensator is a type II compensator. The compensator has an integrator, a pole
coincident to ESR zero of (6) and a zero to set the desired phase margin [4]. The crossover
frequency should be near to (1/10)th of the switching frequency. The compensators are
designed so as to have a system gain margin greater than 10dB and phase margin greater
than 600. For digital implementation, the transfer function of GCV is discretized using bilinear
transformation at a sampling rate equal to converter switching frequency.

5. SIMULATION AND EXPERIMENTAL RESULTS

Three phase buck converter is used for battery charging application wherein the power is
derived from the grid. Three phase, 125V, 50Hz AC supply is rectified by a diode bridge
rectifier and filtered using 2000uF capacitors. The filtered DC is the input to the three phase
buck converter. The rectifier output has nominal voltage of 175V DC. The charger has a
capacity of 22kW and is designed for an output of 120V, 180A. The converter is operating at
20 kHz. The inductance per phase is 100µH. The output capacitor has a value of 200µF. The
battery bank is made up of 76 Ni-Cd batteries in series to form a total rating of 120V 850Ah.
THREE PHASE INTERLEAVED BUCK CONVERTER… 415

The system is simulated in SIMULINK. The current reference for constant current charging
is set for 130A. The loop frequency response obtained from MATLAB for constant current
mode charging is shown in Fig.7. The charging current waveform with and without
feedforward is shown in Fig.8 and Fig.9 respectively. It can be seen that with feedforward
gain, the ripple content has reduced from 11.5% to 3.8%.

Fig.7. Frequency response of the loop in constant current mode

Fig.8. Charging current with feedforward gain

416 VISHWAS K, CHAITHRA L AND SURYANARAYANA K

Fig.9. Charging current without feedforward gain mode

The hardware is developed with diode bridge rectifier, IGBT module, inductors and
capacitors. The antiparallel diode of IGBTs is used in rectification mode. MC56F84789VLL,
Freescale make Digital Signal Controller (DSC) is used to generate the desired PWM pulses
and to measure the feedbacks using inbuilt 8 channel ADC. The inductor currents, output
current are measured using Allegro Microsystem make CTs. The measured feedback signals
are signal conditioned to meet the requirements of DSC. The detailed hardware block
diagram is shown in Fig.10. The measured charging current for a current reference set to
130A is shown in Fig.11.

POWER BOARD

MAIN T1 T2 T3
CONTACTOR
L1 SKKD380/16
R R IPH1 IBATP DC+
LA100 LA
125V
L2 200P
50Hz Y Y
IPH2
LA100
3 PH AC B B L3
IPH3
LA100
CHARGING
CONTACTOR

LA IBATN DC-
ACPL FF300R12KE4 BASED BUCK 200P
SKKD101/16 BASED ACPL ACPL
C7840 CONVERTER
RECTIFIER C7840 C7840
VDCBUS
VCHOP VBAT
VDCBUS
VCHOP

IBATN
VBAT

IBATP

IGBT DRIVERS
NTC FB
IPH2
IPH1

IPH3

(ACPL332J)
ISOLATED
OUTPUTS

SMPS UNIT
T3
T1

T2

(UCC28C44)
ADC PWM OUTPUTS

MC56F84789VLL

Fig.10. Frequency response of the loop in constant current mode

THREE PHASE INTERLEAVED BUCK CONVERTER… 417

Fig.11. Measured charging current for current reference set to 130A

CONCLUSION
Three phase buck converter is designed for battery charging application. Multiphase
topology helps in reducing the current ripple at input and output. Feedforward technique is
employed to reduce the DC link ripple which reflects in battery voltage and current. With
reduced ripple in battery voltage and current, the battery life is improved.

REFERENCES

[1] Ruddell, A.J.; Dutton, A.G.; Wenzl, H.; Ropeter, C.; Sauer, D.U.; Merten, J.;
Orfanogiannis, C.; Twidell, J.W.; Vezin, P “Analysis of battery current microcycles in
autonomous renewable energy systems”, Power Sources 2002, 112, 531–546.
[2] Jin-Hyuk Park, Hae-GwangJeong and Kyo-Beum Lee, “Output Current Ripple
Reduction Algorithms for Home Energy Storage Systems”, Journal on Energies, Vol. 6
Issue 10, p5552-5569, October 2013.
[3] Suryanarayana K, [Link], “Multiphase Boost Converter Model for Input Ripple
Current Reduction”, National conference on Emerging Trends in Engineering,
Technology and Management”, January 2014.
[4] Vishwas K, Suryanarayana K, N.M Renukappa, L.V Prabhu, “Modeling of multiphase
boost converter for solar battery charging system”, IEEE Students Conference on
Electrical, Electronics and Computer Science, February 2014.
418 VISHWAS K, CHAITHRA L AND SURYANARAYANA K

[5] “Benefits of multi-phasing buck converters”, Tim Hegarty, National Semiconductors.

[6] “Analog Circuit Design - A Tutorial Guide to Applications and Solutions” By Linear
Technologies, First Edition 2011 (Edited by Bob Dobkin and Jim Williams)
[7] Erickson R.W, Maksimovic D, “Fundamentals of power electronics”, 2nd edition,
Kluwer Publishers, 2001.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 419-428

ENABLING HAND AS A 3-D MOUSE AND CONTROLLING

DESKTOP APPLICATIONS USING MEMS TECHNOLOGY

SHIVALILA A HANGARAGI1 AND NAYANA P SHETTY2

1
Dept of E&E, Microelectronics and Control System NMAMIT,
NITTE, Karnataka, India
2
Asst. Prof. Dept of E&E, Microelectronics and Control System
NMAMIT, NITTE, Karnataka, India

Abstract
There are different gesture recognition models which are capable of recognizing seven hand gestures,
i.e., up, down, left, right, tick, circle, and cross, based on the input signals from MEMS 3-axes
accelerometers. The accelerations of a hand in motion in three perpendicular directions are detected
by three accelerometers respectively and transmitted to a PC via Zigbee wireless protocol. An
automatic gesture segmentation algorithm is developed to identify individual gestures in a sequence.
To compress data and to minimize the influence of variations resulted from gestures made by different
users; a basic feature based on sign sequence of gesture acceleration is extracted. Finally, the gesture
is recognized by comparing the gesture code with the stored templates. Using this application cursor
movement can be controlled in the desktops. Eye blink sensors are used to control the desktop
applications. When eye is blinked it accounts the obstacle and based on its digital signals are sent to
control the applications such as file opening and closing operations.

---------------------------------
Keywords: Gesture recognition, Zigbee, Eye blink sensor.
© http: //[Link]
420 SHIVALILA A HANGARAGI AND NAYANA P SHETTY

1. INTRODUCTION
THE increase in human-machine interactions in our daily lives has made user interface
technology progressively more important. Physical gestures as intuitive expressions will
greatly ease the interaction process and enable humans to more naturally command
computers or machines. For example, in telerobotics, slave robots have been demonstrated to
follow the master’s hand motions remotely. Other proposed applications of recognizing hand
gestures include character-recognition in 3-D space using inertial sensors gesture recognition
to control a television set remotely, enabling a hand as a 3-D mouse and using hand gestures
as a control mechanism in virtual reality. Moreover, gesture recognition has also been
proposed to understand the actions of a musical conductor. In our work, a miniature MEMS
accelerometer based recognition system which can recognize seven hand gestures in 3-D
space is built. The system has potential uses such as a remote controller for visual and audio
equipment, or as a control mechanism to command machines and intelligent systems in
offices and factories. Many kinds of existing devices can capture gestures, such as a
“Wiimote,” joystick, trackball and touch tablet. Some of them can also be employed to
provide input to a gesture recognizer. But sometimes, the technology employed for capturing
gestures can be relatively expensive, such as a vision system or a data glove.
To strike a balance between accuracy of collected data and cost of devices, a Micro Inertial
Measurement Unit is utilized in this project to detect the accelerations of hand motions in
three dimensions. There are mainly two existing types of gesture recognition methods, i.e.,
vision-based and accelerometer and or gyroscope based. Since heavy computation burden
will be brought if gyroscopes are used for inertial measurement our current system is based
on MEMS accelerometers only and gyroscopes are not implemented for motion sensing.
Existing gesture recognition approaches include template-matching, dictionary lookup
statistical matching, linguistic matching, and neural network. For sequential data such as
measurement of time series and acoustic features at successive time frames used for speech
recognition, HMM (Hidden Markov Model) is one of the most important models .It is
effective for recognizing patterns with spatial and temporal variation. There are three
different gesture recognition models, which are: 1) sign sequence and Hopfield based gesture
recognition model; 2) velocity increment based gesture recognition model; and 3) sign
sequence and template matching based gesture recognition model. In this paper sign
ENABLING HAND AS A 3-D MOUSE AND CONTROLLING… 421

sequence and template matching based gesture recognition model is used among these three
models, in order to find a simple and efficient solution to the hand gesture recognition
problem based on MEMS accelerometers, the acceleration patterns are not mapped into
velocity, displacement or transformed into frequency domain, but are directly segmented and
recognized in time domain and it is more accurate than other two models [1]. By extracting a
simple feature based on sign sequence of acceleration, the recognition system achieves high
accuracy and efficiency without the employment of HMM.

2. GESTURE MOTION ANALYSIS

Gesture motions are in the vertical plane (as defined by the x-z plane in Fig. 1(a)) or the
projection of the motions is mainly in the vertical plane, so the accelerations on x- and z-
axes are adequate to distinguish each gesture. Therefore, the acceleration on y-axis is
neglected to reduce computational requirement. We propose that the exact shape of the
acceleration curves is not critical, but only the alternate sign changes of acceleration on the
two axes are required to uniquely differentiate any one of the 7 gestures: up, down, left,
right, tick, circle, and cross. This is the basis of the recognition algorithms discussed in this
paper. For instance, the gesture up has the acceleration on z-axis in the order: negative —
positive— negative (positive z direction points downward) and nearly has no acceleration on
x-axis; for a circle gesture, on x axis: positive-negative-positive and on
Z-axis: negative-positive-negative-positive. Experiments showed that each of these gestures
has a special order of sign changes, and a kinematics analysis also proves this. For example,
a simple up gesture can be decomposed into several acceleration and deceleration periods.
As shown in Fig. 1(b) an up gesture is actual consist of motion from point 1 to point 2, and
then back to point 1. The velocity at the starting point 1, midpoint 2 and end point 1 are all
zeros. For the convenience of analysis, point 3 is the point between point 1 and point 2
where acceleration changes sign and point 4 is the point between point 2 and point 1 where
acceleration changes sign. Then the acceleration changes can be described as:
1→ 3: acceleration on z-axis is negative (since positive z direction is downward); velocity
changes from zero to a maximum value at 3; acceleration at point 3 is zero.
3 →4: acceleration on z-axis is positive; velocity changes from negative to positive and is
maximum at point 4, where acceleration becomes zero.
422 SHIVALILA A HANGARAGI AND NAYANA P SHETTY

4→ 1: acceleration on z-axis is negative; velocity changes from positive to zero. Also,

acceleration and velocity become zero at point 1.
The analysis above is illustrated by Fig. 1(c). Fig. 1(d) is the real acceleration plot for the
gesture up in which the dotted line is the acceleration on z-axis and solid line is the
acceleration on x-axis. From Fig. 1(d), we note that noise exists from sensor measured data.
However, the noise does not influence the trend of the acceleration curves, and hence, the
analysis of gestures based on the above method still works without adding computational
burdens on a CPU by using a noise-filtering algorithm. Comparing the predicted acceleration
pattern in Fig. 1(c) with the real acceleration plot in Fig. 1(d), it is concluded that the trend
of the real acceleration is the same with the prediction. After analyzing the other gestures, it
was found that they all have unique acceleration patterns for classification. Gesture down is
similar to up but with changes in directions, left and right and also similar, but the changes in
motion axes information. Tick, circle and cross are more complex since they have
accelerations on both x- and z- axes simultaneously, but the accelerations on the two axes
can be separated and decomposed, then the motion trend becomes similar to the above
example. The uniqueness of each gesture trend makes the recognition algorithm possible,
and the algorithms presented in this paper are based on this basic motion feature of the seven
gestures.

3. OVERVIEW
3.1 Sensor Description
The sensing system utilized in our experiments for hand motion and eye blink data
collection is shown in Fig. 2 and is essentially a MEMS 3-axes acceleration sensing chip and
EYE BLINK sensor integrated with data management and Zigbee wireless data chips. The
algorithms described in this paper were implemented and run on a PC.
ENABLING HAND AS A 3-D MOUSE AND CONTROLLING… 423

Fig. 1. Gesture up motion analysis. (a) Coordinate system. (b) Gesture up motion
decomposition. (c) Predicted velocity and acceleration in the z-axis of the gesture up.
(d) Real acceleration plot of the gesture up. Solid and dotted lines are accelerations on x- and z-
axis, respectively

a b
Fig [Link] of the components of the sensing system used for hand gesture recognition.
[Link] based eye blink sensor
424 SHIVALILA A HANGARAGI AND NAYANA P SHETTY

3.2 System Work Flow

When the sensing system is switched on, the accelerations in three perpendicular directions
are detected by the MEMS sensors and transmitted to a PC via zigbee protocol. The gesture
motion data then go through a segmentation program which automatically identifies the start
and end of each gesture so that only the data between these terminal points will be processed
to extract feature. Subsequently, the processed data are recognized by a comparison program
to determine the presented gestures. The work flow of this system is shown in Fig. [Link]
the EYE BLINK sensor is switched on, this Eye Blink sensor is IR based. The Variation
Across the eye will vary as per eye blink if the eye is closed means the output is high
otherwise output is low. This to know the eye is closing or opening position. This output is
give to logic circuit by which the desktop applications are controlled. In this paper, I have
implemented the system using Embedded C and Java programming.

4. GESTURE SEGMENTATION
4.1 Data Acquisition
To collect reliable hand gesture data for the sensing system, the experimental subject should
follow guidelines below during the data acquisition stage:
• The MEMS sensing device should be held horizontally during the whole data collection
process (i.e., the x-y plane of the sensor chip in Fig. 2(a) pointing towards the ground).
• The time interval between two gestures should be no less than 0.2 seconds so that the
segmentation program can separate each one of the gestures in sequential order.
• The gestures should be performed as shown below.

UP LEFT DOWN
ENABLING HAND AS A 3-D MOUSE AND CONTROLLING… 425

Fig. 3. Workflow of gesture recognition system using MEMS.

426 SHIVALILA A HANGARAGI AND NAYANA P SHETTY

4.2 Gesture Segmentation

1) Data Pre-processing: Raw data received from the sensors are pre-processed by two 2
processes:
a) vertical axis offsets are removed in the time-sequenced data by subtracting each data
points from the mean value of a data set; hence, a data set shows zero value on the
vertical axes when no acceleration is applied.
b) A filter is applied to the data sets to eliminate high-frequency noise data.
2) Segmentation: The purpose of the segmentation algorithm is to find the terminal points of
each gesture in a data set of gesture sequence. The algorithm checks various conditions of all
the data points and picks out the most likely data points as the gesture termination points.
The conditions of determining the gesture terminal points in our algorithm are a) amplitude
of the points (y-coordinate value of a data point); b) point separation (the difference between
the x-coordinates of the two points); c) mean value (mean of y-coordinates of points on left
and right sides of a selected point); d) distance from the nearest intersection (quantifies how
far is a selected point away from an “intersection point”, i.e., a point where acceleration
curve crosses from negative to positive or vice versa”); e) sign variation between two
successive points. After examining all the points by checking these 5 different conditions,
the terminal points can be generated for the motion data on each axis. Since, all these five
conditions are checked separately on x- and z- axes acceleration data, two 2xn matrices are
generated for each of gesture sequence data

t11 t12....t1n t11 t12...t1n

ax= t21t22.....t2n az= t21 t22....t2n

The element on the first row is the starting point of gesture and the element in the second
row in same column is the end points of the same gesture. Comparing the columns of the two
matrices, if the pair of terminal points on one axis is close to a pair on the other axis, one pair
of the terminal points will be eliminated. A final determination on if a given set of pairs of
points are indeed terminal points is made by comparing the maximum acceleration between
them with the mean value of the maximum accelerations between all pairs of points. If the
former is too small, then that pair of points will be eliminated.
ENABLING HAND AS A 3-D MOUSE AND CONTROLLING… 427

After obtaining the terminal points of each gesture, the number of gestures becomes obvious
since every gesture has one starting point and one end point, i.e., the number of the columns
of the final terminal points matrix is the number of gestures.

5. GESTURE RECOGNITION BASED ON TEMPLATE MATCHING.

The algorithm is based on the feature of acceleration sign changes which is generalized from
gesture motion analysis; it is not limited to specific users. Therefore, there is no requirement
to train the system by specific users before using it. The gesture motion data then go through
a segmentation program which automatically identifies the start and end of each gesture so
that only the data between these terminal points will be processed to extract feature.
Subsequently, the processed data are recognized by a comparison program to determine the
presented gestures. The work flow of this system is shown in Fig. [Link] range of values is
stored in the program for every gesture. Whenever any gestures are performed comparison is
done between obtained value and the stored value and gestures performed are recognized and
according to that cursor moves on the desktop.

6. EYE BLINK SENSORS FOR CONTROLLING DESKTOP APPLICATIONS.

This Eye Blink sensor is IR based. The Variation Across the eye will vary as per eye blink. If
the eye is closed means the output is high otherwise output is low. This to know the eye is in
closing or opening position. This output is given to the logic circuit for controlling desktop
applications .such as file opening and closing operations

CONCLUSION
This paper describes the system which is very helpful for disabled persons. When the
sensing system is switched on, the accelerations in three perpendicular directions are
detected by the MEMS sensors and transmitted to a PC via zigbee protocol. Using Gesture
recognition based on Template matching model gestures are recognised and according to the
gesture performed cursor moves on the [Link] based eye blink sensors are used for
controlling desktop applications such as file opening and closing operations. So this paper
proposes efficient system for disabled persons and physically challenged people to operate
the system easily.
428 SHIVALILA A HANGARAGI AND NAYANA P SHETTY

REFERENCES

[1] MEMS Accelerometer Based Nonspecific- User Hand Gesture Recognition Ruize Xu, Shengli
Zhou, and Wen J. Li, Fellow, IEEE .
[2] Zhou S., Dong Z., Li W. J., and Kwong C. P., “Hand-written character recognition using
MEMS motion sensing technology,” in Proc. IEEE/ASME Int. Conf. Advanced Intelligent
Mechatronics, 2008, pp.1418–1423.
[3] ZigBee™ Wireless Transceiver Engineering Options- Cirronet, Inc. Summer 2005
[4] Speeter T. H., “Transformation human hand motion for telemanipulation,” Presence,
vol. 1, no. 1, pp. 63–79, 1992.
[5] Zhou S., Dong Z., Li W. J., and Kwong C. P., “Hand-written character recognition using MEMS
motion sensing technology,” in Proc. IEEE/ASME Int. Conf. Advanced Intelligent Mechatronics,
2008, pp. 1418–1423.
[6] Hand gesture recognition using MEMS for specially challenged people- International Journal
of VLSI and Embedded Systems-IJVES ISSN: 2249 – 6556
[7] u Wave: Accelerometer-based Personalized Gesture Recognition and Its Applications- Jiayang
Liu, Zhen Wang, and Lin Zhong
[8] Newman W. M. and Sproull R. F., Principles of Interactive Computer Graphics. New York:
McGraw-Hill, 1979
[9] Rubine D. H., “The Automatic Recognition of Gesture,” Ph.D dissertation,Computer Science
Dept., Carnegie Mellon Univ., Pittsburgh, PA, Dec. 1991.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 429-440

EVALUATION OF REACTANCES AND TIME CONSTANTS

OF SYNCHRONOUS GENERATOR

SHAHEENA KHANUM1 AND K. L. RATNAKAR2

1
PG Student, 2 Professor, Department of Electrical and Electronics Engineering,
Sri Siddhartha Academy of Higher Education, Maralur, Tumkur 572105, Karnataka, India.

Abstract
For any power equipment the control system and protection system plays an imperative role as the
dependency on the power equipment in any industry will be very high. Thus for the Generator the
Reactances and Time Constants becomes controlling parameter and hence protecting it is necessary
for prolonged usage. In general reactance shall be defined as “non-resistive component of impedance
in an AC circuit, arising from the effect of inductance or capacitance or both and causing the current
to be out of phase with the electromotive force causing it”. In other word it can also be defined as the
imaginary part of the impedance in any power circuit. There are various Reactances and Time
Constants which contributes during the selection of control systems and protection scheme for any
Generator. Therefore practical evaluation of those parameters shall also be super critical same as those
are by calculation. More accurate measurement or the calculation of reactance shall certainly give us
more stabilized error free system and hence ensuring a healthy control system. As there are different
tests or different methods to measure and calculate the reactances and time constants, in this context
we are considering Line to Line fault (L-L fault), Double line to Ground fault and Sudden three phase
short circuit fault to determine the Negative Sequence reactance, Zero Sequence reactance, Sub-
transient and Transient reactance and time constants respectively. The fault condition will be
demonstrated with the 40MW, 11KV, 2624A, 4pole, 50 Hz synchronous generator and above
mentioned parameters are calculated based on references available.

---------------------------------
Keywords : Reactances, Time Constants, Brushless Alternator.
© http: //[Link]
430 SHAHEENA KHANUM AND K. L. RATNAKAR

1. INTRODUCTION
The synchronous machine is an ac machine in which the rotor moves at a speed which bears
a constant relationship to the frequency of currents, in the armature winding. A synchronous
machine is one of the important types of electric machines. Large ac networks operating at
constant frequency of 50 Hz or 60 Hz rely almost exclusively on synchronous generators,
also called the alternators, for the supply of electrical energy, and may have synchronous
compensators at key points for control of reactive power. Private stand-by and peak load
plants with diesel or gas-turbine prime movers also have synchronous generators. Non-land-
based synchronous plants are found on oil rings, on large air-crafts and constant speed
industrial drives with the possibility of power factor correction, but are not often built in
small ratings for which the induction motor is cheaper. Synchronous machines are generally
constructed in larger sizes. Small size alternators are not economical. The modern trend is to
build alternators of very large sizes capable of generating 500MVA or even more. The
synchronous motor is rarely built in small sizes owing to superior performance
characteristics and economical construction of induction motors. Synchronous machines,
according to their applications, may be synchronous generators, synchronous motors or
synchronous compensators. A synchronous generator is a synchronous machine which
receives mechanical energy from a prime-mover to which it is mechanically coupled and
delivers electrical energy. A synchronous motor receives electrical energy from ac supply
main and drives mechanical load. Synchronous compensator is a synchronous machine
designed to operate on no-load with its shaft connected neither to a prime-mover nor to a
mechanical load and is used to control reactive power in power supply networks.
An alternator works on the principle of Faraday’s laws of Electro-Magnetic Induction.
“Whenever a conductor/coil cuts the magnetic flux an EMF is induced in that conductor. The
magnitude of EMF induced in the conductor/coil is directly proportional to the rate of
change of flux and the number of conductors”. The main component which effects the
operation of alternator is resistance and reactances. The effective resistance of the armature
winding is somewhat greater than the conductor resistance, called the dc resistance, as
measured by direct current. This is due to additional loss over the purely I2R loss, inside and
sometimes outside the conductor, owing to alternating current. The main sources of this
additional loss are
 Eddy current in the surrounding material,
EVALUATION OF REACTANCES AND TIME CONSTANTS… 431

 Magnetic hysteresis in the surrounding material and

 Eddy currents or unequal current distribution in the conductor itself.
In many cases it is sufficiently accurate to measure the armature resistance by direct current
and increase it to a fictitious value, called the effective resistance, Re which is large enough
to take care of these additional losses. Effective resistance, Re can vary widely from 1.25 to
1.75 or more times the dc resistance, depending upon design. The EMF setup due to
armature reaction MMF Fa is always in quadrature with the load current I and is proportional
to it. Thus it is equivalent to an EMF induced in an inductive coil and the effect of armature
reaction can, therefore, be considered equivalent to reactance drop IX a where Xa is the
fictitious reactance which takes care of the armature reaction effect. The armature winding
possesses a certain leakage reactance Xl. The sum of leakage reactance Xl and fictitious
reactance Xa is called the synchronous reactance.

2. REACTANCES
It can be defined as the imaginary part of the impedance in any power circuit. There are
various Reactances and Time constants which contribute during the selection of control
systems and protection scheme for any Generator.

Fig.1 Axes of Alternator.

432 SHAHEENA KHANUM AND K. L. RATNAKAR

The axis of symmetry of the north magnetic poles of the rotor is called the direct axis or d-
axis. The axis of symmetry of the south magnetic poles is the negative d-axis. The axis of
symmetry halfway between adjacent north and south poles is called the quadrature axis or
q-axis.
 Direct axis reactance (Xd): The quotient of the sustained value of that fundamental AC
component of armature voltage, which is produced by the total direct-axis primary flux
due to direct-axis armature current and the value of the fundamental AC component of
this current, the machine running at rated speed.
 Quadrature axis reactance (Xq): The quotient of the sustained value of that
fundamental AC component of armature voltage, which is produced by the total
quadrature-axis primary flux due to quadrature-axis armature current, and the value of
the fundamental AC component of this current, the machine running at rated speed.
 Direct-axis transient reactance (Xd’): The quotient of the initial value of a sudden
change in that fundamental AC component of armature voltage, which is produced by
the total direct-axis primary flux, and the value of the simultaneous change in
fundamental AC component of direct-axis armature current, the machine running at
rated speed and the high decrement components during the first cycles being excluded.
 Direct axis sub-transient reactance (Xd”): The quotient of the initial value of a
sudden change in that fundamental AC component of armature voltage, which is
produced by the total direct-axis armature flux, and the value of the simultaneous
change in fundamental AC component of direct-axis armature current, the machine
running at rated speed.
 Negative sequence reactance (X2): The quotient of the reactive fundamental
component of negative sequence armature voltage, due to the presence of fundamental
negative sequence armature current at rated frequency, by the value of that component
of current, the machine running at rated speed.
 Zero sequence reactance (Xo): The quotient of the reactive fundamental component of
zero sequence armature voltage, due to the presence of fundamental zero sequence
armature current at rated frequency, by the value of that component of current, the
machine running at rated speed.
EVALUATION OF REACTANCES AND TIME CONSTANTS… 433

3. TIME CONSTANTS
• Direct-axis transient short-circuit time constant ( d): It is the time, in seconds,
required for the transient alternating component of the short-circuit current to decrease
to 1/ε0 or 0.368 times its initial value.
• Direct-axis sub-transient short-circuit time constant ( d): It is the time, in
seconds, required for the sub-transient alternating component of the short-circuit current
to decrease to 0.368 times its initial value.
• Direct- do): It is the time, in seconds,
required for the transient alternating component of the open-circuit current to decrease
to 0.368 times its initial value.

4. TESTING OF SYNCHRONOUS GENERATOR

In this paper, study of different types of testing is going to be done for an alternator. Testing
is the practice of making objective judgments regarding the extent to which the product
meets else fails to meet the specified objectives. In other words for any manufactured
product or equipments it is essential that the evaluation of the product with reference to the
specification and the applicable standards. It is also indispensable for any manufacturer to
ensure their product is meeting all the requirements as per their design and by their
processes. Testing of a Generator or an Alternator also play a vital role before it is being put
in to continuous operation. As the manufacturing of Generator involves various crucial
processes either it may be carried by machinery or by man or may be both put together and
hence it becomes important to assess the work done at the end. Though each component and
part of the Generator has under gone various tests it is necessary that the complete assembled
Generator also shall undergo at least Routine (described later) tests to understand the basic
performance.
Generator testing shall be of two types i.e. Type test and Routine test. Type test is inclusive
of all routine tests along with some additional tests. Routine test is compulsory to understand
the elementary performance of the Generators which comes under serial production type.
434 SHAHEENA KHANUM AND K. L. RATNAKAR

4.1 Sudden Short Circuit Test

4.1.1 Circuit

Fig.2 Sudden short circuit test setup.

4.1.2 Procedure
1. Makes the connections as per the above diagram.
2. Run the generator to its rated speed.
3. Build up the voltage to 10% of the rated voltage.
4. 4. Close the circuit breaker & check the waveforms. Check the waveform for the Clear
identification of the amplitudes. (Should be 3 current & 1 voltage waveform)
5. If ok, then raise the voltage to 30% of the rated voltage then close the breaker, record
the current and voltage waveforms.
6. Note down the steady state current, record & print waveforms.
4.2 Zero Sequence Test (Xo)
4.2.1 Circuit

Fig.3 Zero sequence test setup.

EVALUATION OF REACTANCES AND TIME CONSTANTS… 435

4.2.2 Procedure
1. Make the connections as per above circuit.
2. Run the machine to its rated speed
3. Increase the armature current in steps of 10%, 20%, 30 %, 40% and 50% of rated
current.
4. Record the values of current (I), voltage (V) and power (P).
5. Xo shall be calculated using measured values.
4.3 Negative Sequence Test
4.3.1 Circuit

Fig.4 Negative sequence test setup.

4.3.2 Procedure
1. Make the connections as per above circuit.
2. Run the machine to its rated speed
3. Increase the armature current in steps of 10%, 20%, 30 %, 40% and 50% of rated current.
4. Record the values of current (I), voltage (V) and power (P).
5. X2 shall be calculated using measured values.

5. CALCULATION OF REACTANCES AND TIME CONSTANTS

5.1 Three phase sudden short circuit test and determination of characteristic
reactance and time constants.
The synchronous reactance (Xd), transient reactance (X'd) and sub-transient reactance (X''d),
and the transient short-circuit time constant (t'd) and sub-transient short circuit time constant
(t''d) are used to describe and machine's behavior on sudden short-circuit. This can be done in
436 SHAHEENA KHANUM AND K. L. RATNAKAR

accordance with the following equation for the AC RMS components of current following a
three-phase short-circuit from no load neglecting armature-circuit resistances and assuming
constant exciter voltage.
I (t) = (E/Xds) + ((E/Xd’ ) – (E/Xds)) ԑ (-t/Ʈd’ )+ (E/Xd’’ –E/ X’ d) ԑ (-t/Ʈd’’ )
Where,
I (t): is the AC RMS short- circuit current, p.u.
E: Is the AC RMS voltage before short circuit, p.u.
t: is the time in seconds, measured from the instant of short circuit.
In this expression, it is assumed that the current is composed of a constant term and two
decaying exponential terms where the third term of the equation decays very much faster
than the second. By subtracting the first (constant) term and plotting the remainder on semi-
logarithmic paper as a function of time, the curve would appear as a straight line after the
rapidly decaying term decreases to zero. The rapidly decaying portion of the curve is the
sub-transient portion, while the straight line is the transient portion.
5.2 Evaluation
(a) Measure the voltage from wave form recorded (pk-pk) & converts the same to RMS
value.
Es = Epk-pk*(PT ratio) / (2*sqtr (2))
(b) Measure the mV drop of steady state current from the waveform recorded & convert the
same to amperes (RMS value)
Is = mVpk-pk*(Shunt ratio) / (2*sqtr (2))
(c) Peak to Peak value (mV) of the currents from the waveform recorded is measured &
RMS value of the currents are calculated to first 50 to 60 cycles.
(d) Subtract the steady state current from point (c) & plot the graph for the component
currents (I'k+I''k) in the semi log sheet with current in on log scale & time on linear
scale.
(e) To separate transient & sub-transient components, extrapolation to zero time from the
straight portion by neglecting the rapid variation of current in first few cycles will give
the initial values of corresponding current.
(f) From the semi log sheet
I'k: is the AC RMS short-circuit transient current component in ampere (RMS).
I''k: is the AC RMS short-circuit sub-transient current component in ampere (RMS).
EVALUATION OF REACTANCES AND TIME CONSTANTS… 437

Es : is the voltage before short circuit in RMS.

Is : is the AC RMS short-circuit steady state current in ampere (RMS).
5.3 Determination of Direct axis Sub-transient reactance & short circuit Time
constant (Xd'' & Td'')
To calculate Xd'' in %
Xd'' in % = (Es*100)/(1.732*Zs*(I'k+I''k+Is))
Where,
Zs (Ω): Synchronous impedance of the machine = Rated Voltage / (1.732*Rated Current)
To calculate Td'' in seconds
Td'' is the time required to decrease the sub-transient component to 1/e i.e. 0.368 of its initial
value.
OR Time for Id'', where Id''=0.368*I''k
5.4 Determination of Direct axis Transient reactance & short-circuit Time constant
(Xd' & Td')
To calculate Xd' in %
Xd' in % = (Es*100)/(1.732*Zs*(I'k+Is))
Where,
Zs (Ω): Synchronous impedance of the machine = Rated Voltage/(1.732*Rated Current)
To calculate Td' in seconds
Td' is the time required to decrease the transient component to 1/e i.e. 0.368 of its initial
value.
OR Time for Id', where Id'=0.368*I'k
5.5 Determination of Direct axis synchronous reactance (Xd)
To calculate Xd in %
Xd in % = (1/SCR)*100
Where, SCR: Short circuit ratio of the machine
SCR = ( Ifoc/Ifsc)
Ifoc: Field Current for rated voltage.
Ifsc: Field Current for rated current.
5.6 Determination of Direct axis Transient open circuit Time constant (Tdo')
To calculate Tdo' in seconds , Tdo' in seconds = (Xd*Td'/Xd')
438 SHAHEENA KHANUM AND K. L. RATNAKAR

Fig.5.1 Analysis of ac component of short-circuit current (for one of three phases)

Fig.5.2 Transient & Sub-transient component of short-circuit

6. EXPERIMANTAL RESULTS

Fig.7 Voltage and Current waveforms.

EVALUATION OF REACTANCES AND TIME CONSTANTS… 439

The above graph shows the current and voltage waveforms for the sudden 3 phase short
circuit and then the below values of Reactances and Time constants are calculated and the
values are as below.
Table 1 Values of Reactances
Time Design Calculated
Constants value value
’
Td 0.82 0.84
Td’’ 0.035 0.035
Td0’ 4.69 4.408

Table 2 Values of Time Constants

Reactances Design value Calculated value
Xd 166 161.3
X d’ 29 30.7
Xd’’ 23 25.2
X2 30 30.08
X0 17 18.15

As per the IEEE standards used the acceptable tolerance allowed is ±15% or 30% on one side.

CONCLUSION
The above procedure and the calculation shall be used as a generalized method for
calculation of Transient, sub-transient reactance X2, X0 and Time Constants for both
cylindrical and salient pole type alternator of any rating. Though the Transient behavior
involves tedious steps of calculation and deep analysis of the results, the above procedure
helps in calculating the various parameters in easiest way with minimum calculation and
time.

ACKNOWLEDGEMENT
The authors express their sincere thanks to Mr. Ramesh K. N, Quality Assurance, TD Power
Systems Limited, Dabaspet, Bangalore, for his many valuable suggestions in connection
with this paper.
440 SHAHEENA KHANUM AND K. L. RATNAKAR

One of the author is also indebted to Dr. B. Rajesh Kamath, Head of the Dept. E&EE, SSIT,
Tumkur for his many helpful suggestions.

REFERENCES

[1] IEEE115 standards for “Test procedure for synchronous machines”.

[2] IEC60034 standards for “Methods for determining synchronous machine quantities from tests”.
[3] Marksmen, A. L. and Morsztyn, K., "Analysis of Synchronous Machine Transient Tests Using a
Microcomputer," Proceedings IEE, 124(4), pp. 377–380, 1977.
[4] Say M. G., Alternating Current Machines. Pitman Publishing, 1978.
[5] Fitzgerald A.E., Kingsley C., Umans S.D., Electric Machinery, 1983 edition, McGraw-Hill
[6] Canay, I. M., "Causes of Discrepancies on Calculation of Rotor Quantities and Exact Equivalent
Diagrams of the Synchronous Machine," IEEE Transactions on Power Apparatus and Systems,
vol. PAS-88, pp. 1114–1120, July 1969.
[7] “The Transient Reactions of Alternators” by William A Durgin and R. H. Whitehead, a paper to
be presented at the 29th annual convention of the America Institute of Electrical Engineers,
Boston, Mass., June 28, 1912.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 441-449

A THREE PHASE VIENNA RECTIFIER FOR

WIND POWERGENERATION

BHUMIKA S AND SHRI HARSHA J

Department of Electrical and Electronics,
Sri Siddhartha Institute of Technology, Tumkur, India

Abstract
In recent years, wind energy technology has become one of the top areas of interest for energy harvesting
in the power electronics world. Wind energy conversion system is a system that converts kinetic energy
of the wind into electricity. The primary focus of this project is to design and implement the power
electronics converter (AC_DC) of a wind energy conversion system. The aim is to develop an interface
between a three-phase AC generator operating at variable speed (e.g. wind generators) and a constant
voltage DC-bus. The interface is required to ensure high energy efficiency by reducing reactive power
consumption, as well as maintaining a constant DC-bus voltage. This paper deals with the design and
evaluation of power electronics topology commonly called the “Vienna rectifier” that can be used for
improved variable-speed wind energy conversion. The main advantage for the VIENNA converter is that
PFC allowed more power extraction from the turbine, increasing both electrical and mechanical
efficiency.

----------------------------------
Keywords : AC_DC Converter, Vienna Rectifier, Wind energy Conversion system.
© http: //[Link]
442

1. INTRODUCTION
AC-DC converters find application in everyday-life as a front-end to DC-DC and DC-AC
converters. In low power with low cost applications, the AC to DC conversion is very often
merely a diode bridge rectifier with capacitor voltage filter. However, bridge rectification
inherently draws non-sinusoidal current from the mains, which make it inadequate for high
power applications due to the strict regulations on conducted EM (electromagnetic) energy, as
well as the high current stress on components. For high power applications, the sinusoidal
current must be actively shaped by using either a boost type front-end converter or by complex
EM filtering at the input. Research and development of the latter has ceased mainly due to the
cost and size associated with EM filters.
As wind generators as an energy source became more popular, the need arises to efficiently
convert the energy provided to a usable source and the same time conserve energy by reducing
reactive power consumption. The interface developed as part of this paper will serve as a
possible solution for fulfilling this need.
The rectifier must thus ensure that a power factor of close to 1 is achieved at the source input.
This implies that the input current is both sinusoidal and in-phase with the input voltage,
assuming that the input voltage is also sinusoidal. The interface is to be based on the VIENNA
rectifier

2. METHODOLOGY
The wind turbine along with the Vienna rectifier deals with the reduction of line harmonics and
switching losses are reduced. In this paper novel scalar control technique is used for controlling
the switching and hence DC output will stabilize. V ienna rectifier is three phases, three switches,
three level rectifiers where power factor correction is done using a better scalar controlling
scheme. As the pulse number increases, the harmonics present in the input decreases and the
total harmonic distortion reduces. The output of the wind turbine varies according to the wind but
while connecting to load we have to maintain constant voltage so to step up & step down the
voltages in rectifier section a three switch Vienna rectifier is implemented. The wind turbine
converts the kinetic energy present in the wind into mechanical energy. The output of the wind
turbine is connected to Synchronous generator. The generator converts the mechanical energy
into electrical energy. The output of the generator is connected to Vienna rectifier; it converts the
unstable AC voltage into stable DC voltage and input line current is sinusoidal hence unity power
A THREE PHASE VIENNA RECTIFIER FOR WIND POWERGENERATION… 443

factor is obtained. The Vienna rectifier is used to make power factor correction and only three
IGBT switches are used so switching loss is reduced. The output of the Vienna rectifier is given
to battery.

3. VIENNA RECTIFIER
The Vienna rectifier consists of three switches IGBT, it converts the unstable AC voltage into
controlled DC voltage. It can also provide sinusoidal input currents and controlled DC-voltage

Fig.3.1. Circuit Diagram of Vienna Rectifier

The AC voltage from the three phase generator is given to the Vienna rectifier. The current
flows through the three IGBTs and the capacitors in the capacitor bank begin to charge and
when the capacitors are fully charged it compensates the reactive power and hence the power
factor is improved. The topology of the three-phase/three-switch/three-level PWM (“Vienna”)
rectifier is depicted in circuit diagram Fig 3.1. Herein, we consider the electromechanical
system until the dc bus, which is assumed to maintain a constant dc voltage. The switches are
placed and the switching is made in such a way that the numbers of solid state switches are
reduced.
The PWM block is made to generate the gating signals for IGBT. In Vienna Rectifier the output
capacitor is split in two parts as two equal value capacitors, C1 and C2, connected in series.
Across the output capacitors the –Vdc and +Vdc are developed as 3-Phase peak detected outputs.
A switch for each phase is connected, such that when “ON”, it connects the line phase to the
center node of C1 and C2 through a series inductance. For a short switching period, the
444

capacitors charge linearly. This offsets -Vdc and +Vdc. Proper switching control strategy for
switching will boost the Dc voltage to desired level and power factor correction is done so that
input current is sinusoidal and follows voltage phase leading to unity power factor.
3.1 Vienna Circuit Operation (Switching Strategy)
The Vienna rectifier has three switches, and by choosing their (ON\OFF) state considering the
polarity of the phase current in each phase, the voltage for each phase will be determined. So, the
phase voltage is depending on the direction of phase current and switch position State of the
switch (ON/OFF) and the polarity.
Of the line current in each phase determine the rectifier pole voltages (VAM, VBM, and VCM)
at any instant of operation. In order to discuss operation principles of the rectifier, here Phase A
is explained. Faze B and C has the same behavior. If the line current is positive, and the switch
Ta is off, the current flows through diode D11, and the voltage between the converter pole A and
the DC bus midpoint M (i.e. VAM) is VDC/2. The conduction path for this case is illustrated in
Fig 4.1(a). If the polarity of the line current is positive, and the switch Ta is on, the voltage VAM
is 0, in which the conduction path is illustrated in Fig 4.1(b) Similarly, the voltage VAM can be
determined in other states as illustrated in Fig 4.1(c) and (d).This operating principle also can
apply to Phase B and C to determine VBM and VCM.
Here switching is controlled using pi controller such that input current will be made to follow
the sinusoidal voltage waveform thus reducing the harmonics and correcting the power factor
hence input current control. The output Dc voltage is also boosted output and hence using Pi
voltage controller desired Dc voltage is applied

Fig.4.1 Conduction paths for phase-leg A when (a) the line current is positive, and the controlled switch is
off (b) The line current is positive, and the controlled switch is on (c) The line current is negative, and the
controlled switch is off and (d) The line current is negative, and the controlled switch is on.
A THREE PHASE VIENNA RECTIFIER FOR WIND POWERGENERATION… 445

3.2 Advantages of Vienna Rectifier

a. Gives continuous sinusoidal input current
b. No need for a neutral wire.
c. Low number of IGBTs used and hence circuit complexity reduced.
d. Reduction in switching loss of the power semiconductors by almost 40%.
e. Wide voltage range as the Boosting ability is present.
f. Improved power factor and lower THD hence higher efficiency.

4. CONTROL SCHEME FOR SWITCHING

The low-bandwidth outer loop with characteristic (s) is used to keep the output voltage of the
PFC stage constant and to provide the error signal .The high-bandwidth inner loop with
characteristic GH(s) is used to control the input current. A multiplier Vex is used to provide a
reference, which is proportional to the error signal eve and which has a modulating signal with
the desired shape for the input current.

Fig 5.1 Control scheme for PFC using a switching converter operating in CCM.

In this case the modulating signal is Vxy = Vε and it is essentially constant over the line cycle,
because Vε is the control signal from the low-bandwidth output voltage controller. Therefore, the
446

input current is sinusoidal and proportional to input voltage and output voltage at desired level is
obtained

5. SIMULATION RESULTS
The Feasibility of proposed Vienna rectifier with PI Controller is proved with MATLAB
Simulink V2012b. Fig.7.1 shows the Simulink model and results are obtained for input voltage
100VAC and frequency of [Link] waveforms are shown as shown in Fig 7.2 (a) and 7.2 (b).
The same waveform is obtained for varying input condition as controller is tuned to get desired
output
5.1 Simulation Parameters
Simulation parameters are as listed in the table below. The parameters are designed to get the
output of desired level. Simulation time is 10μS and switching frequency or frequency of
repetitive sequence is 10 KHz Kp is tuned to get PI control in PI controller block. For various
input voltages the output is evaluated as in wind power generation we get varying input voltage
based on turbine rotation.

Table.1 List of Simulation Parameters and its Values

Parameter Symbol Value

Simulation time Ts 10μS

Inductance L 4 ηH

Capacitance C 3 mF

Load Resistance R 10 KΩ

Nominal Three phase Input Vin 100 V- 230V

Voltage(Varying)

Nominal Three Phase Input Current Iin 10 A – 30A

Input Frequency F 60 Hz

Switching Frequency Fs 10 KHz

Proportionality Gain Kp 0.018

A THREE PHASE VIENNA RECTIFIER FOR WIND POWERGENERATION… 447

Fig.7.1 Power circuit of Vienna Rectifier

Fig7.2 (a) Phase an input Voltage and Current Waveform

448

Fig.7.2 (b) Output DC Voltage at 670V for input 100 VAC

CONCLUSIONS
This paper proposed a three phase, three switches and three levels Vienna rectifier circuit for wind
power generation that is varying input voltage is simulated. From the simulation results it is
observed that input phase current is in phase with the input phase voltage hence unity power factor
and reduced harmonic levels. It is also observed that it is possible to compensate PF and THD for
varying input voltage level by using the proper switching control scheme. A new technique of
phase locked loop and power factor correction control strategy compensate input imbalances of
wind power generation system.

ACKNOWLEDGMENT
The Authors would like to thank Dr. B. Rajesh Kamath, Head of the Dept. EEE, SSIT, Tumkur for
his constant support and valuable suggestions in carrying out this work. The authors would like to
thank Prof. K L Ratnakar, Professor, EEE Dept. SSIT Tumkur for his guidance and constant
supervision in carrying out this work. The authors highly indebted to L&T Integrated Engineering
Services, Mysore for hosting the work.
A THREE PHASE VIENNA RECTIFIER FOR WIND POWERGENERATION… 449

REFERENCES

[1] Kolar J.W. and Ertl H., "Status of the Techniques of Three-Phase Rectifier Systems with Low Effects
on the Mains", 21st INTELEC, Copenhagen, Denmark, pp.14.1 June 1999.
[2] Tou M., Al-Haddad K., Olivier G. and Rajagopalan V.R., "Analysis and Design of Single-Controlled
Switch Three-Phase Rectifier with Unity Power Factor and Sinusoidal Input Current", IEEE
Transactions on Electronics, Vol. 2, No. 4, pp.856-862, July 1997
[3] Pathak A.D., Locher R. E., Mazumdar H.S.,”3-Phase Power Factor Correction, Using Vienna
Rectifier Approach and Modular Construction For Improved Overall Performance, Efficiency and
Reliability”, Power electronics Conference, CA, USA, 2003.
[4] Radomski G., “Analysis of Vienna Rectifier”, Technical university of Kielce, Poland, Electrical
power quality and utilization, Journal Vol. XI, No.1, 2005.
[5] Chen H., Aliprantis D.C.,” Induction Generator with Vienna Rectifier: Feasibility Study for Wind
Power Generation”, .XIX international conference on electric Machines-ICEM, Rome, 2010.
[6] Ghost Rajesh and Narayanan G., A Single-Phase Boost Rectifier System for Wide Range of Load
Variations, IEEE transactions on power electronics, vol. 22, no. 2, March 2007
[7] Mohamadian Mustafa and Rajaei Amirhossein, Vienna-Rectifier-Based Direct Torque Control of
PMSG for Wind Energy Application, IEEE transactions on industrial electronics, July 2013
[8] Dave Mitulkumar R., Dave K.C. Analysis of Boost Converter Using PI Control Algorithms,
International Journal of Engineering Trends and Technology, July 2012
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 451-462

IMPLEMENTATION OF A SINGLE PHASE Z-SOURCE

BUCK-BOOST MATRIX CONVERTER USING
PWM TECHNIQUE

SHIVANAGOUDA B PATIL1, SWETHA S. G. 2 & VINAY KUMAR3

1 Department of Electrical and Electronics Engineering, V.N.E.C, Shorapur, Karnataka, India
2 Department of Electrical and Electronics Engineering, N.I.T, Raichur, Karnataka, India
3 Department of Electrical and Electronics Engineering, C.M.E ,Pune, Maharashtra, India

Abstract
This paper presents a single phase z-source buck-boost matrix converter using PWM Technique which
can buck and boost the output voltage with step change frequency in this technique both the voltage
and the frequency can be steeped up or steeped down. Also, this converter uses a safe-commutation
technique to conduct current in a continuous path, resulting in the elimination of voltage spikes on
switches without the need for a snubber circuit. It has no energy storage devices, performing the
energy conversion by directly connecting input with output phases through bidirectional switches. For
this reason, it is known as all-silicon power converters. In the proposed scheme, dsPIC30F2010
controller is used to produce PWM signals. The operating principles, analysis and experimental results
of the proposed single phase z-source buck-boost matrix converter using PWM techniwue are
described and the experimental results show that the output voltage can be obtained at three different
frequencies—100, 50, and 25 Hz in the buck–boost amplitude mode. Textronics TDS2024B storage
oscilloscope is used to store the gate pulses and waveforms.

----------------------------------
Keywords : Buck–boost voltage, single-phase matrix converter, Step-up and step-down frequency, Z-
source converter. © http: //[Link]
452 SHIVANAGOUDA B PATIL, SWETHA S. G. AND VINAY KUMAR

1. INTRODUCTION
Matrix converter is an AC-AC converter that can directly convert a fixed AC voltage to
variable voltage, variable frequency AC without a large energy storage element. The first
matrix converter was introduced by Pelly and Gyugyi in 1976. In 1980 Venturni and Alesina
presented the first algorithm capable of synthesizing output sinusoidal reference voltages
from a balanced three phase voltage source connected to the converter input terminals. The
first study of a single phase matrix converter was performed by Zuckerberger on a frequency
step up and fundamental voltage step down converter.
Recent research on matrix converters has extended its operation to inverter, controlled buck
rectifier, boost rectifier and buck-boost rectifier. In other topologies, the AC output voltage
cannot exceed the AC input voltage. Furthermore, it is not possible to turn both the
bidirectional switches of a single phase leg on at the same time, as the current spikes
generated by this action will destroy the switches. Both of these limitations can be overcome
by using Z-source topology. Many researchers have also focused on Z-source AC-AC
converters which mainly finds applications where only voltage regulation is needed. In
primary converter model topology, amplitude of output voltage can’t be greater than
amplitude of input voltage. Z-source converters solve this problem.
The single phase Z-source AC-AC converters has several advantages such as providing large
range of output voltages with the buck-boost mode, reducing inrush and harmonic currents.
In this project work we apply the Z-source concept to a single phase matrix converter to
create a new type of converter called a single phase Z-source buck-boost matrix converter. In
contrast to the existing single phase PWM AC-AC converters, this proposed single phase Z-
source buck-boost matrix converter can provide a wide range of output AC voltages in buck-
boost mode with step up/step down frequencies .
The safe commutation scheme establishes a continuous current path in dead time to
eliminate voltage spikes on the switches without a snubber circuit. Applications of single-
phase matrix converters have been described for induction motor drives, radio-frequency
induction heating, audio power amplification, and compensation voltage sags and swells. It
has been reported that the use of safe-commutation switches with pulse width modulation
(PWM) control can significantly improve the performance of AC-AC converters [4].
IMPLEMENTATION OF A SINGLE PHASE Z-SOURCE… 453

To verify the operation of the proposed converter, we constructed a laboratory prototype

with the lamp load based on dsPIC30F2010, the experimental results show that the output
voltage can be obtained at three different frequencies—100, 50, and 25 Hz in the buck–boost
amplitude mode. Thus, the proposed single-phase Z-source buck–boost matrix converter can
be used for voltage applications that require step-changed frequency or amplitude. In
particular, it can be applied to the starting of an asynchronous motor as well as to the speed
control of an induction motor, which needs a step-changed speed. The various
graphs/waveforms are analyzed and studied on digital Storage Oscilloscope.

2. BLOCK DIAGRAM AND ITS EXPLANATION

2.1 System overview
The block diagram of the proposed single phase z-source buck-boost matrix converter is
shown in figure 1.
The ac voltage is applied to the Z-source network were the input Vi voltage is bucked or
boosted and the output Va is given to the single-phase matrix converter Then, the single-
phase matrix converter varies the frequency of Va depending on the switching sequence .
The output voltage Vo is obtained with a step-changed frequency and a variable amplitude.
Fig. 2 shows the proposed single-phase Z-source buck–boost matrix converter. It has an LC
input filter; a Z-source network, bidirectional switches, and a load. The LC input filter is
required to reduce switching ripple included in input current. All the inductors and capacitors
are small and are used to filter switching ripples. The symmetrical Z-source network, a
combination of two inductors and two capacitors, is the energy storage/filtering element for
the single-phase Z-source buck–boost matrix converter.
Since the switching frequency is much higher than the ac source (or line) frequency, the
requirements for the inductors and capacitors should be low [3]. As shown in Fig. 2, the
proposed single-phase Z-source buck–boost matrix converter requires 4 bidirectional
switches S1j, S2j , S3j , and S4j (j = a, b) to serve as a single-phase matrix converter and one
source bidirectional switch Ssj (j = a, b), where a and b refer to drivers 1 and 2, respectively.
All bidirectional switches are common emitter back-to-back switch cells. The five switches
Ssj, S1j, S2j, S3j, S4j (j=a, b) used in the single-phase Z-source buck–boost matrix converter
are bidirectional switches, as shown in Fig. 2.
454 SHIVANAGOUDA B PATIL, SWETHA S. G. AND VINAY KUMAR

Fig1. Block diagram of the proposed system

The bidirectional switches are able to block voltage and conduct current in both directions.
Because these bidirectional switches are not available at present, they can be substituted for
by combinations of two diodes and two insulated gate bipolar transistors (IGBTs) connected
in antiparallel (common emitter back to back), as shown in Fig. 2. The diodes are included to
provide the reverse blocking capability. The IGBTs are used because of their high switching
capabilities and their high current-carrying capacities, which are desirable for high-power
applications. As indicated in the figure, D refers to the equivalent duty ratio and T is the
switching period.
Implementing the single-phase Z-source buck–boost matrix converter requires different
bidirectional switching arrangements depending on the desired amplitude and frequency of
the output voltage. The amplitude of the output voltage is controlled by the duty ratio D,
while the frequency of the output voltage depends on the switching strategy. In this paper,
the frequency of input voltage fi is assumed to be 50 Hz, and the desired output frequency fo
is synthesized to be 100Hz (step-up frequency), 50Hz (same frequency), or 25Hz (step-down
frequency).
IMPLEMENTATION OF A SINGLE PHASE Z-SOURCE… 455

Fig.2 Proposed Single phase Z-source Matrix converter topology

For example, Fig. 3 illustrates the converter’s switching strategy over one cycle of input
voltage for a 100-Hz output frequency in buck mode. To double output frequency of the
input voltage, the operation of the converter is divided into four stages, as shown in the
figure.
Fig. 4 illustrates stage 1 in the buck mode when both input voltage and output voltage are
positive.
In state 1, as shown in Fig. 5(a), Ssa, s1a and S4b turns on and conducts current flow during
the increasing positive cycle of input voltage; In state 2 as shown in Fig. 5(b) Ssa, S2a and
S3b turn on and conduct negative current flow from source to the load during positive half
cycle, similarly Ssb, S2b and S3a conduct in state 3 and Ssb, S1b and S4a conduct in state 4.
456 SHIVANAGOUDA B PATIL, SWETHA S. G. AND VINAY KUMAR

Fig3. Switching sequence for 100Hz in Buck mode

Fig4. Switches turned ON during each state for 100Hz in Buck mode
IMPLEMENTATION OF A SINGLE PHASE Z-SOURCE… 457

(a) Ssa, S1a and S4a are ON

(b) Ssa, S2a and S3b are ON

Fig 5. Direction of current flow for 100Hz in Buck mode

2.2 Control circuit

The control circuit of the proposed scheme consists of a Digital signal Controller
dsPIC30F2010.A Digital Signal Controller (DSC) is a single-chip, embedded controller that
seamlessly integrates the control attributes of a Microcontroller (MCU) with the computation
458 SHIVANAGOUDA B PATIL, SWETHA S. G. AND VINAY KUMAR

and throughput capabilities of a Digital Signal Processor (DSP) in a single core. The dsPIC
DSC has the “heart” of a 16-bit MCU with robust peripherals and fast interrupt handling
capability and the “brain” of a DSP that manages high computation activities, creating the
optimum single-chip solution for embedded system designs. The dsPIC30F devices contain
extensive Digital Signal Processor (DSP) functionality within high-performance 16-bit
microcontroller (MCU) architecture. It also consists of ten opto-couplers for isolating the
control and power circuits. In this work an optocoupler MCT2E is used to isolate the gate
drive circuit and the IGBT-based power circuit. Ten IGBTs of the power circuit are
controlled by the PWM signals generated by the control circuit.

3. EXPERIMENTAL SETUP AND ITS RESULTS

The implementation of a single phase z-source buck-boost matrix converter is done
successfully and the developed hardware is tested with load. The proposed control system is
implemented by a DSC (dsPIC30F2010) based PWM inverter. C language is used to develop
the program. The device is programmed using MPLAB Integrated Development
Environment (IDE) tool. It is a free, integrated toolset for the development of embedded
applications employing Microchip's PIC and dsPIC controllers. For execution of C-code,
MPLAB compiler is used.
The hardware set is developed and tested in power electronics laboratory and the photograph
of complete setup is shown in fig 6. In the complete experiment the oscilloscope used is
Tektronix TDS2024B Digital Storage Oscilloscope (DSO) to store gate pulses and voltage
waveforms.
Table 1 : Experimental parameters

L-C input filter Li 0.1mH

Ci 6.8µF
L1 =L2 1mH
Z-Source Network
C1 =C2 1µF

Switching Frequency 20kHz

IMPLEMENTATION OF A SINGLE PHASE Z-SOURCE… 459

Experimental results show that the output voltage can be obtained at three different
frequencies 100, 50, and 25 Hz and in the buck–boost amplitude mode. Thus, the proposed
single-phase Z-source buck–boost matrix converter can be used for voltage applications that
require step-changed frequency or amplitude. In particular, it can be applied to the starting of
an asynchronous motor as well as to the speed control of an induction motor, which needs a
step-changed speed.
The input voltage is taken by the normal power supply 230v 50 Hz. The input is stepped
down by the step-down transformer and is given to the LC filter. Depending on the desired
output frequency, the Microcontroller generates ten control signals (four PWM signals to
control four switches S1b, S2b, S3b, S4b, four I/O signals to control four switches S1a, S2a,
S3a, S4a in the single-phase matrix converter, and two PWM signals to control source
bidirectional switch Ssj (j = a, b), where a and b refer to drivers 1 and 2, respectively.
Figs. 7-10 show the input voltage and output voltage for the experiments done in buck–boost
mode with output frequencies of 100, 50, and 25 Hz, respectively.

Fig6. Photograph of complete Experimental setup

460 SHIVANAGOUDA B PATIL, SWETHA S. G. AND VINAY KUMAR

Fig7. Experimental result at frequency of 100 Hz (top: V0 (100 Hz) (In Buck mode); bottom: V0
(100 Hz) (In boost mode).

Fig8. Experimental result at frequency of 25 Hz (top: V0 (25 Hz) (In Buck mode); bottom: V0
(25 Hz) (In boost mode).
IMPLEMENTATION OF A SINGLE PHASE Z-SOURCE… 461

Fig9. Experimental result at frequency of 50 Hz (top: V0 (50 Hz) (In Buck mode); bottom: V0
(50 Hz) (In boost mode).

CONCLUSION
In this project work, we have presented a single-phase Z-source buck–boost matrix converter
using PWM Technique that can do both buck and boost to the desired output voltage with
step-changed frequency. The output of this single-phase Z-source buck–boost matrix
converter PWM Technique produces the voltage in buck–boost mode with a step-changed
frequency, in which the output frequency is either an integer multiple or an integer fraction
of the input frequency. It provides a continuous current path by using a PWM Technique,
which avoids voltage spikes on the switches.
In traditional matrix converters conversion is done in two stages, using two power
controllers one as converter (AC-to-DC) and other as inverter (DC-to-AC), but in this
topology direct ac to ac conversion is done reducing the conversion stages.
We presented here the steady-state Analysis, operational stages and circuit analysis to verify
the performance of this converter; we constructed a laboratory prototype with an input
voltage of 50 Volts/50 Hz based on dsPIC30F2010. The experimental results with a lamp
462 SHIVANAGOUDA B PATIL, SWETHA S. G. AND VINAY KUMAR

load showed that the output voltage can be produced at three different frequencies, 100, 50,
and 25 Hz, and in the buck–boost amplitude mode.
This converter is particularly suitable for controlling the speed of a fan or a pump without
the use of an inverter because for these applications, the input voltage frequency must be
changed to control their speed by stages.

REFERENCES

[1] Nguyen M. K., Jung Y. G., Lim Y. C., and Kim Y. M., “A Single-Phase Z Source Buck–Boost
Matrix Converter,” IEEE Trans. Power Electronics, vol. 25, no. 2,pp. 453-462, Feb. 2010

[2] Deivasundari P. and Jamuna V. “A Z-source single phase matrix converter with safe
commutation strategy,” INTERNATIONAL JOURNAL Volume 2, Issue 3, 2011 pp.579-588

[3] Fang X. P., Qian Z. M., and Peng F. Z., “Single-phase z-source PWM ac-ac converters,” IEEE
Power Electron. Lett., vol. 3, no. 4, pp. 121– 124,2005.

[4] Tang Y., Zhang C., and Xie S., “Single-phase four switches Z-source acac converters,” in Proc.
IEEE APEC 2007, pp. 621–625.

[5] Idris Z., Hamzah M. K., and Saidon M. F., “Implementation of single phase matrix converter as
a direct ac-ac converter with commutation strategies,” in Conf. Rec. IEEE PESC 2006, pp.
2240–2246.

[6] P. Sweety Jose1N. Chandra Deepika [Link] “Applications of single phase matrix converter,”
IEEE Trans. Power Electron., vol. 22, no. 3, pp. 1005–1013, May 2004.

[7] Vargas R., Ammann U., and Rodriguez J., “Predictive approach to increase efficiency and
reduce switching osses on matrix converters,” IEEE Trans. Power Electronics, vol. 24, no. 4,
pp. 894–902, Apr. 2009.

[8] Kolar J. W., Schafmeister F., Round S. D., and Ertl H., “Novel three phase ac ac sparse matrix
converters,” IEEE Trans. Power Electron., vol. 22,no. 5, pp. 1649–1661, Sep. 2007.

[9] Perez J., Cardenas V., Moran L., and Nunez C., “Single-phase ac-ac con- verter operating as a
dynamic voltage restorer (DVR),” in Proc. IEEE IECON 2006, pp. 1938–1943.

[10] Nguyen-Quang N., Stone D. A., Bingham C. M., and Foster M. P., “Single phase matrix
converter for radio frequency induction heating,” in Proc. SPEEDAM 2006, pp. S18-28–S18-32.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 463-474

LED LIGHTING TECHNOLOGY APPLICATIONS: A REVIEW

ANITHA MARINA COLACO1 AND SAVITHA G KINI2

1
Department of Electrical & Electronics Engineering,
NMAM Institute of Technology, Nitte, Karnataka, India.
2
Department of Electrical and Electronics Engineering,
Manipal Institute of Technology, Manipal, Karnataka ,India

Abstract
LED is a benevolent technology that has emerged as supplement for existing conventional lightings.
LED technology has its role in reducing global energy consumption and reducing global warming.
The quality of light is governed by factors such as CRI, CCT, chromaticity coordinates and luminous
flux. The quality of light obtained by LEDs is much superior due to its flexibility in modulating CCT
and CRI. The intuitive control of LEDs has made its smart entry into architectural lighting, hospital
lighting, retail shops, indoor and outdoor applications. The spectral tunability and dynamism in CCT
has made LED lightings in enhancing mood, cheerfulness and reset circadian rhythm. This paper
reviews the development of LED, control strategies and application in various sectors.

------------------------------------
Keywords LED, colour rendering Index, CCT,CRI
© http: //[Link]
464 ANITHA MARINA COLACO AND SAVITHA G KINI

1. INTRODUCTION
The three important aspects of light are vision, perception and wellbeing. The light has
biological impact on humans in their mood and behavior. The figure1 shows the light
conveyed to SCN cells. The rods and cone transmits the visual information to the visual
centre of the brain via optic nerve. The ganglion cells are connected with superior cervical
ganglion in the spinal cord and with Suprachiasmatic nucleus (SCN) retinohypothalamic
tract. The SCN uses the pineal gland and harmones to synchronise the human body to
external world thus regulating circadian rhythm. The intelligent design and implementation
of lightings can enhance the ambience and appearance of the deployed area.

Fig 1. Light transmission to SCN.

Utilizing fire for the light was the trivial method which moved onto candle, then lamp,
incandescent and then to fluorescent, halogen lamps, sodium vapour lamp and mercury
vapour lamp. The figure 2 shows the progression in lighting technology.

Fig 2. Modernization of lighting technology

LED LIGHTING TECHNOLOGY APPLICATIONS A REVIEW 465

The most commonly used incandescent and fluorescent lamp fail in dynamism, intensity
control and spectral tunability. The conventional light sources which are monochromatic
required colored glass filters to render vividness in colour and ambience. The solution to the
dilemma was achieved through solid state lightings which not only gave dynamic colour
renderance but also intensity control. The Solid state lighting burns fifty times longer than
incandescent lamps i.e 1000 hrs and 18000hrs compared to fluorescent lamp. The LEDs first
entered the market in 1962 as an indicator bulb whose intensity was less than 5 lux.
Modernization in technology bestows 8W LED lamp, for example, to deliver the same
amount of luminous flux as a 60W incandescent lamp. Today, LED systems compete even
with fluorescent lamps. The triband fluorescent lamp have intensity control along with
dynamism in CCT apart from it has CRI less than LEDs. The observers prefer LEDs for
reading purpose when compared to CFL. LED luminous efficacy in the past has doubled
about every two years. The longevity, efficiency and high lighting quality of LEDs has
replaced the conventional incandescent lamp. The use of innovative LED solutions raises
lighting quality while at the same time permitting a sustainable reduction of lighting costs
[2], reduction in heat and reduction in global warming]. Savings potential is maximised by
“intelligent” luminaire management. The illumination of each luminaire is adjusted
automatically in response to presence and daylight sensors [3]. The combination of LED
luminaires and lighting management is a lucrative option in corridors, seminar rooms, cove
lightings and technical rooms. LED luminaires producing dynamic colour light set striking
accents in reception areas, corridors and stairwells. Colour and dynamism emphasize areas
with a prestigious character, ambience and also enliven meetings. With a wide spectrum of
colour temperatures, LEDs can even recreate the natural progression of daylight indoors and
thus promote relaxation and activation in line with biological rhythms.

2. LED SPECTRA
White LED light can be produced by various manufacturing methods. A very thin film of
yellow phosphor material is applied to a blue LED chip, which changes part of its blue light
into white. To achieve the light colour required, the concentration and chemical composition
of the phosphor material needs to be very precisely controlled. Today, a variety of white
tones are possible ranging from warm white (colour temperature 2,700 kelvin, K) through
neutral white (3,300 K) to daylight white (5,300k) Fig. 3 shows the spectra of white and
466 ANITHA MARINA COLACO AND SAVITHA G KINI

coloured LEDs. The inner structure of LED is shown in fig 4a and fig4b depicts the
conversion of blue yellow phospor conversion leading to white light. The white spectra is
also achievable by combination of RGB, tetrachromatic or polychromatic LEDs sources.

Fig 3 showing the spectra

Fig 4a shows the structure of LED

LED LIGHTING TECHNOLOGY APPLICATIONS A REVIEW 467

Fig 4b shows the embedded conversion layer

In addition this method results relatively high luminous fluxes and good colour rendering up
to Ra - 90. This method is advantageous in obtaining dynamic colour and good ambience.
RGB solutions are good for dynamic coloured lighting applications. Integrating modern
technology to produce white light permits a switch from warm white to cool white for office
applications. The current technology combines coloured chips with white LEDs resulting
dynamically changing white light with a good colour rendering property Ra 80-90.

3. LUMINOUS EFFICACY
LEDs are extremely efficient light sources. The first LED, produced in 1962, achieved a
luminous efficacy of 0.1 lumen/watt (lm/W). Current study shows, ratings in the region of 50
l/W are standard and high-power LEDs reach an average of 90 lm/W. Incandescent lamps
achieve-around 10 lm/W,
Halogen lamps-20lm/W and LEDs-200lm/w

4. REVIEW ON CONTROL STRATEGIES

The tunability in CCT and luminous intensity require a intelligent control system. The most
common used technique are PWM. The variation in choromaticity shift from the set point
leads poor colour renderance. This problem is addressed by maintaining the thermal stability
using heat sink and constant current control through a feedback. The modern technology
uses DALI, PLC, WIMAX and ZIGBEE’s for automation and bestow good colouring
468 ANITHA MARINA COLACO AND SAVITHA G KINI

rendering index, brightness and user satisfaction. The intelligent control deployed for
efficient control of light is described in the following section: Qingcong Hu and Regan Zane
[4] worked on digitally controlled off line driver based on low voltage series–input converter
modules. The series–input connected modular architecture was adopted to utilize low voltage
high frequency circuits and low profile components in off-line applications. Input current
was regulated to be constant using a simple controller to achieve 0.9 p.f a bidirectional buck
second stage was used to reduce 120 HZ current ripple in the LEDs with minimal energy
storage requirements and high efficiency. The model developed had three modules with 6, 8
and 7 LED string. A HB-LED can be dimmed by reducing the drive current which not the
efficient way to control to overcome this, T. W. Ching [5] worked on low cost
microcontroller to generate PWM dimming signal. Jose A Gutierez [6] implemented power
factor correction converter followed by a coupled inductor fly-back converter. The carried
out was successful in retrofitting fluorescent by the energy efficient HB-LED lighting
systems. [Link] and [Link] [7] predicted Energy saving obtained by daylight responsive
lighting control systems shows differences according to the months and seasons. Alejandro
Fernandez-Montes, Luis Gonzalez_Abril, Juan A. Ortega and Francisco Velasco morente [8]
worked on energy saving in artificial where WSNs are deployed to retrieve data on lighting
conditions
T.P Huynh [9], y. K. Tan and K. J. Tseng proposed the a controllable LED lighting system
embedded with ambient intelligence gathered from a distributed smart wireless sensor
network (WSN) to optimize and control the lighting system to be more efficient and user
oriented. An energy aware communication protocol was implemented to reduce the power
consumption of the WSN for longer operating period. The work conserved up to 10% of
waste light energy without compromising the workplace lighting condition and WSN were
able to operate from 87days to 102 days with about 20% increase in their operational
lifetime.
Yao-jung Wen, Jessica Grandson and Alice M Agogino [2] introduced intelligent day
lighting system to retrofit the existing hardwired building using MEMS based smart Dust
Motes wireless platform because of which authors improved the illumination estimation and
maintenance. The authors also commented that motes functioning as actuators could be
interfaced with dimming ballasts without the need to rewire the building.
LED LIGHTING TECHNOLOGY APPLICATIONS A REVIEW 469

Meng- Shiuan Pan, lun-Wu Yeh, Yen-Ann Chen, Yu Hsuan Lin and Yu –Chee Tseng [10]
developed a WSN based intelligent light control system which considered user activities and
profiles. The experimental set up included two types of lighting devices local lighting and
global lighting devices. The wireless sensor collected the light intensities in the environment
.Illumination decision algorithm and a device control algorithm met the user requirement
aswell as conserved the energy. Future scope of the work suggested to direct relieving the
computation cost of the nonlinear programming and enhance the user interfaces at the
portable sensor nodes.
Lun-Wu Yeh, Che-Yen, Chi-Wai Kao, Yu-Chee Tseng and Chih –wei Yi[11] the authors
developed the autonomus light control system based on the feedback from light sensors
carried by users the authors generated a surface tracking scheme to control the local light.
The model developed adapted dynamically to environmental changes. The user preference is
judged by binary satisfaction model and continuous satisfaction model. Drawback of the
work is user must carry the sensor to measure their current light intensities. The suggested
method to overcome the drawback is to use motion sensor or thermal sensor.
Hong Yang, Jan Q.M Bergmans and Tim C.W. Schenk [12] worked on illumination sensing.
The frequency division multiplexing scheme was used to distinguish the signals from
different LEDS to estimate luminance of all LED simultaneously filter bank sensor structure
was proposed to study the key properties of the FDM techniques. Work predicted that
rectangular function can support maximum LED for the ideal case without frequency
inaccuracies
Jean-Paul [Link], Lorenzo Feri, Hongming Yang Sel B Colak and Tim C.W Schenk
[13] proposed the method to accurately measure and estimate the local light contributions of
a larger set of LEDs. The light emitted by each light source carried a unique identifying
code, invisible to human eye. A sensor in the remote control detects these identifiers,
identified the light sources involved and measured the strength of the contribution of each
LED. This allowed remote control to send a control message to modify the relevant light
settings. The modulation scheme implemented CTDMA PWM and CTDMA-PPM. The
impact of timing accuracy such timing offset and timing jitter were considered. The overlap
hybrid code and time division multiplex access scheme was used to separate the signals from
the different LED, which allows for a large number of light sources to be measured by the
system. The work carried out by the authors was successful to meet the MSE performance
470 ANITHA MARINA COLACO AND SAVITHA G KINI

upto sensor-LED distance of about 7-10m. Future scope was to implement the numerical
analysis of the system taking into account the reflection.
Hongming Yang, Tim C. W. Schenk, Jan W. M Bergmans and Ashish Pandharipande [12]
developed a model to increase the number of LED by exploiting the multiple harmonics in
the sensor. Predicted that the impact of noise in illumination estimation is negligible. The
hurdles that scale down the performance were (i) induced interference that is overlap of
illumination of neighboring LED on LED of interest (ii) frequency offset which made the
control of induced interference much difficult (iii) Duty cycle. To overcome these hurdles
the authors used multiple harmonics approach which increased the number LED from earlier
work [13]. The advantage of this approach was number of LED supported scaled
logarithmically with the number of harmonics considered i.e five times the number specified
in[12]. Future work suggests close the gap between upper limits on the number of LED and
that achieved by successive estimator. Addition to it future work also suggested use of
multiple distributed photosensors.
Ashish Pandharipande and David caicedo [14] the work focused on energy efficient
illumination control design of LED based lighting systems in office space. The work was to
achieve optimum dimming level while rendering(i) uniform illumination at a given
illumination level in occupied workspace and (ii) minimum illumination level of lower value
in unoccupied regions. The daylight was considered. The method implemented estimated
and dissggregated illumination contributions of daylight and the different LED source at
workspace. Future work suggested involving extensive field tests to understand the impact of
specific occupancy patterns on the energy savings.

5. APPLICATIONS OF LED LIGHTINGS

Good colour renderance of LED lighting is used to illuminate monumental or sculpture.
Sculpture can be illuminated with different colours which exploits the colour dynamism and
intensity control of LEDs. Figure 5 shows a monument flooded with LED lighting.
LED LIGHTING TECHNOLOGY APPLICATIONS A REVIEW 471

Fig. 5 shows LED in architectural lighting

The dynamism of colour control of LEDs has effectively used in retail shop which enhances the
ovrerall appearance and ambience of the shop shown figure 6. The shopping mall’s flooded with LED
lighting attracts the customer which progresses the sales, marketing and economy.

Fig 6 shows LED in marketing.

Vividness and good colour renderence has made smart entry of LED in hospital lighting shown in
figure 7. The hospital lighting requires proper skin tone examination and tissue identification during
surgery. The hospital lighting requires a CRI 80-90. The LED lighting renders CRI 80-90. LEDs are
efficiently used in surgical lighting, lighting the ward rooms, illuminating care taker area, reception
area. The most of the daylight do not penetrate deep into the room this deficit can be overcome by
integrating the LEDs with daylight enhancing wellbeing of long stay patient and clinical staff. The
LED lighting in CT rooms can allay fear in patients such as children.
472 ANITHA MARINA COLACO AND SAVITHA G KINI

Fig. 7. Hospital lightings using LEDs

Figure 8 shows the LED lightings deployed for illuminating ground. The street lighting using the LEDs
is a nitche area, figure 9 shows LEDs in office and figure 10 LEDs in hotel lightings.

Fig.8 ground illuminated by LEDs

LED LIGHTING TECHNOLOGY APPLICATIONS A REVIEW 473

Fig.9 LED lighting in office Fig.10 LED lightings in hotels

6. LIMITATIONS OF LEDS
Even though the LEDs are most efficient than conventional light sources it requires heat sink for
thermal management, a dimming control and feedback unit to maintain stability. The retrofitting cost is
high. Prolonged utilization of incandescent lamp and fluorescent lamp makes LED lighting neglected.
The cost of installation is high compared other sources. The proper design of LED driver is challenging
as technology modernises.

CONCLUSION
This paper gives comprehensive review of present technology in lighting solutions. The LEDs are
efficient, reliable and zero maintenance once installed. The intelligent control of intensity and
chromaticity makes LED lucrative option for architectural, hospital, office, outdoor and indoor
lightings. The thermal management and chromaticity shift are issues which hinder retrofitting
existing conventional sources by LEDs. The paper also showed the LEDs deployed in various areas.

REFERENCES

[1] Shur Michael S., Fellow, Ieee, And Artu¯ Ras ZˇUkauskas “Solid-State Lighting: Toward
Superior Illumination” Proceedings of The IEEE, VOL. 93, NO. 10, OCTOBER 2005.pp1691-
1703

[2] Wen Yao- Jung, Grandson Jessica and Agogino Alice M “towards embedded wireless
networked intelligent daylighting systems for commercial buildings” IEEE conf on sensor
networks, ubiquitous and

[3] S.M Zanoli, D Barchiesi and Barboni “coupled controller for energy saving I building
automation based on a thermodynamics and illumination model” proc 19 th Mediterrean
conference on control and automation, June 20-23,2011.
474 ANITHA MARINA COLACO AND SAVITHA G KINI

[4] Hu Qingcong and Zane Regan “A 0.9 p.f led driver with small LED current ripple based on
series – input digitally controlled converter modules”. proc. conference on Applied power
electronics and exposition”,2010 twenty fifth annual IEEE, pp2314-2320/trans. Industrial
electronics IEEE trans vol 58, issue 2, 589-599

[5] Ching T. W. “Modular dimmable light emitting diode driver for general illumination
applications” in Proc of Electrical and computer Engineering (CCECE 2010) 23 rd Conference on
02-05 May 2010.

[6] Gutierez Jose A “On the use of IEEE 802.15.14 to enable wireless sensor networks in building
automation” In Proc. Conf. personal, indoor and Mobile Radio Communications,2004, 15 th
IEEE international symposium05-08 Sept. 2004 PP 1865-1869 Vol.3.

[7] Onagil and [Link], “Determination of the energy saving by daylight responsive lighting control
systems with an example from istanbul” building and environment, vol. 38, no.7, 2003, pp 973-
977.

[8] Fernandez-montes Alejandra, Luis Gonzalez_Abril Juan a. ortega and Francisco Velasco
Morente “ A study on saving energy in artificial lighting by making smart use of wireless sensor
network and actuators” ieee network 2009.

[9] Huynh T.P, Tan Y. K. and Tseng K. J. “energy aware wireless sensor network with ambient
intelligence for smart led lighting system control” proc iecon 2011-37th annual conference on
IEEE industrial electronics society.

[10] Pan Meng-Shiuan, Yeh Lun-Wu, Chen Yen-Ann, Lin Yu Hsuan and Tseng Yu –Chee “ A
WSN-BASED intelligent Light control system Considering user Activities and profiles” IEEE
Sensor journal, VOL.8. No 10, October2008, PP 1710-1721.

[11] Yeh Lun-Wu, Che-Yen, Chi-Wai Kao, Yu-Chee Tseng and Chih –wei Yi” Autonomous Light
Control by wireless sensor and Actuators” IEEE sensor Journal,Vol 10, No.6, June 2010, PP
1029-1041.

[12] Yang Hong, Bergmans Jan Q.M and Schenk Tim C.W., “Illumination sensing in LED lighting
systems based on frequency division multiplexing”. IEEE Trans on signal processing,vol.57,
No.11Nov,2009.

[13] Jean-Paul M.G. Linnartz, Lorenzo Feri, Hongming Yang Sel B Colak and Tim C.W
Schenk,“Code division based sensing of illumination contributions in solid state lighting”, IEEE
Trans, signal processing,vol.57, No.10 Oct 2009.

[14] Pandharipande Asish and Caicedo David “Daylight integrated illumination control of LED
systems based on enhanced presence sensing” trans. Energy and Building, April 2011 pp 944-
[Link] computing (SUTC’06)
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 475-485

COMPARISON OF MULTILEVEL TOPOLOGIES WITH AND

WITHOUT SELECTIVE HARMONIC REDUCTION

PRADEEP R1 AND JASEERALI E Y2

1
PG Scholar, Department of EEE,
Amrita Vishwa Vidyapeetham, Bangalore, India
2
PG Scholar, Department of EEE,
MEA Engineering College, Perinthalmanna, Kerala, India

Abstract
Multilevel inverter is an inevitable part of power electronics, as it is widely used in many industrial
applications. Multilevel inverter is more advisable when compared to the common inverter because of
its reduced THD and hence multilevel inverter is always used to supply AC machines. Since the
multilevel inverter is of that importance in industries, wide research had been doing for years to
propose various topologies of multilevel inverter. These topologies were aiming at reducing number
of switches, number of DC sources used and voltage across the switches which can ultimately reduce
the THD. Out of various topologies that has been proposed by researchers, multilevel inverter in
normal cascaded connection and multilevel inverter using series connected sub-multilevel topology
with and without incorporating reduced harmonic reduction are taken up for comparison in this paper.
These topologies are simulated in the paper and results are taken for comparison.

© http: //[Link]
476 PRADEEP R AND JASEERALI E Y

1. INTRODUCTION
Multilevel inverter is an arrangement of power semiconductors and DC sources in a proper
manner so as to get a stepped output voltage waveform in its output. Using of high level
multilevel inverter is advisable as the level increases the distortion in voltage waveform
decreases. Complexity of the control algorithm increases with increase in number of
levels.[1],[3],[4],[6].
Various types of multilevel inverter is shown in the form of flowchart

Figure 1. Flowchart showing basic topologies of multilevel inverter

Harmonic reduction method aims at eliminating selected harmonics and it also helps in
reducing harmonics by switching the main switch once in a cycle. All the topologies that are
simulated in this paper is at modulation index=1 and are of seven level.

2. SELECTIVE HARMONIC REDUCTION

Positive half cycle of seven level stepped output voltage waveform of a multilevel inverter is
as shown in the figure 2
COMPARISON OF MULTILEVEL TOPOLOGIES WITH... 477

Figure 2. Positive half cycle of a seven level stepped output of multilevel inverter
Fourier series is used to find the odd harmonics using the waveform shown in figure 2.[2]
4
ℎ𝑛 = [𝑉1 cos(𝑛𝛼1) + 𝑉2 cos(𝑛𝛼2) + 𝑉3cos⁡(𝑛𝛼3)]⁡⁡⁡⁡⁡⁡⁡⁡⁡(1)
𝑛𝜋

where n is the order of odd harmonics and α1, α2, α3 are the switching angles. V1, V2 and
V3 are the voltage at different levels. α1<α2< α3<π/2
Modulation index can be found out by the equation
ℎ1
𝑀= ⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡(2)
𝑚𝑉𝑑𝑐

Where M is the modulation index, h1 is the fundamental component, m is the number of

switching angle and Vdc is the DC voltage source used.
For removing the fifth and seventh odd harmonics these equations are written for fifth and
seventh order and is equated to zero as shown in (3) to (5)
3(1)𝜋
cos(𝛼1) + cos(𝛼2) + cos(𝛼3) = ⁡⁡⁡⁡⁡⁡⁡⁡(3)
4

cos(5𝛼1) + cos(5𝛼2) + cos(5𝛼3) = 0⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡(4)

cos(7𝛼1) + cos(7𝛼2) + cos(7𝛼3) = 0⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡(5)

When these equations are solved using newton Raphson computational method the value of
α1, α2, α3 are obtained as 11.68˚, 31.18˚ and 58.58˚. [2],[7].

3. SIMULATION AND RESULTS

3.1 Cascaded topology with and without selective harmonic reduction
The simulink model of cascaded multilevel topology is shown in figure 3 and the pulse
generator used in the simulink produces the pulse which is as shown in figure 4
478 PRADEEP R AND JASEERALI E Y

Figure 3. Simulink model of cascaded multilevel topology

Figure 4. Pulse generated using pulse generator

COMPARISON OF MULTILEVEL TOPOLOGIES WITH... 479

Output voltage waveform is as shown in figure 5

(a)

(b)
Figure 5(a) Voltage waveform (b) Current waveform

Figure 6. THD in voltage waveform in cascaded multilevel topology

480 PRADEEP R AND JASEERALI E Y

Cascaded topology incorporating selective harmonic reduction can be implemented in the

above simulink model by replacing pulse generators by a MATLAB function block. The
switching angles found for modulation index=1 is 11.68˚, 31.18˚ and 58.58˚. These angles
are used in MATLAB function block so as to switch the switches in such a way that one
switch switches only once in a cycle [2].
THD of the voltage waveform when incorporating selective harmonics reduction is as shown
in figure 7

Figure 7. THD of voltage waveform for cascaded multilevel inverter incorporating selective
harmonic reduction

When observing the voltage waveform of cascaded multilevel topology with and without
selective harmonic reduction, waveform appears to be same but THD has been reduced to
14.02% in case of topology with selective harmonic reduction. This may be because of
reducing the number of times the main switches switch in a cycle.
3.2 Multilevel inverter using series connected sub-multilevel topology with and
without selective harmonic reduction
Simulink model of multilevel inverter using series sub-multilevel topology incorporating
selective harmonic reduction is as shown in figure 8. Load used here is 1000ohm and 1mH
in series.
COMPARISON OF MULTILEVEL TOPOLOGIES WITH... 481

The voltage and current waveform of multilevel inverter using series sub-multilevel
topology incorporating selective harmonic reduction is as shown in figure 9. The switching
sequence is as shown in the table I

Figure 8. Simulink model of multilevel inverter using series sub-multilevel topology with
selective harmonic reduction

Table 1. Voltage Level for State of Switches

Voltage level G1 G2 G1’ G2’ Ga Gb

0 0 0 1 1 1 0
10 1 0 0 1 1 0
20 0 1 1 0 1 0
30 1 1 0 0 1 0
482 PRADEEP R AND JASEERALI E Y

(a)

(b)
Figure 9 (a) voltage waveform (b) current waveform multilevel inverter using series sub-
multilevel topology incorporating selective harmonic reduction THD of the voltage waveform
for multilevel inverter using series sub-multilevel topology incorporating selective harmonic
reduction is as shown in figure 10

Figure 10. THD of the voltage waveform for multilevel inverter using topology incorporating
selective harmonic reduction
COMPARISON OF MULTILEVEL TOPOLOGIES WITH... 483

Figure 11. THD of the voltage waveform for multilevel inverter series sub-multilevel using series
sub-multilevel topology without selective harmonic reduction

Multilevel inverter using series sub-multilevel topology without selective harmonic

reduction can be done using the figure 8 simulink model where the MATLAB function block
is replaced by the pulse generator and providing pulses to the switches [1].
THD of the voltage waveform for multilevel inverter using series sub-multilevel topology
without selective harmonic reduction is as shown in the figure 11
Even though the voltage and current waveform appears to be same for multilevel inverter
using series sub-multilevel topology with and without using selective harmonic reduction,
THD has been reduced to 12.03% in case of topology with selective harmonic reduction.
This may be because of reducing the number of times the main switches switch in a cycle.
THD is reduced in case of multilevel inverter using series sub-multilevel topology when
compared to cascaded topology as the number of switches required to produce seven level is
reduced in case of series sub-multilevel topology when compared to cascaded topology. As
the number of switches gets reduced switching losses decreases, harmonics reduce, control
complexity decreases.
Comparison is done between sub-multilevel topology and cascaded topology with and
without selective harmonic reduction at modulation index=1 and it is tabulated as below.
484 PRADEEP R AND JASEERALI E Y

Table 2 Comparison between Sub-Multilevel Topology and Cascaded Topology With and
Without Selective Harmonic Reduction at Modulation Index=1
Cascaded Cascaded Sub-multilevel Sub-multilevel
multilevel multilevel topology topology with
inverter inverter with without selective
without selective selective harmonic
selective harmonic harmonic reduction
harmonic reduction reduction
reduction

Number of 12 12 8 8
switches
Switching High Low Low Very low
losses
THD 17.05 14.02 14.85 12.03

Voltage Same Same Same Same

waveform
Current Same Same Same Same
waveform
Control High High Low Low
complexity
Efficiency Very Low High High Very high

Filter size Very Large Small Small Very small

Odd harmonic Nil 5th and 7th Nil 5th and 7th
removed

CONCLUSION
This paper presents the comparison of multilevel inverter in cascaded and series sub-
multilevel topology with and without selective harmonic reduction. Series sub-multilevel
topology is good when compared to cascaded topology as the number of switches is less.
Topology incorporating selective harmonic reduction is good when compared to topology
without selective harmonic reduction as odd harmonics causing the considerable harmonics
can be eliminated and switches switch only once in a cycle which again reduces the
harmonics. It was found that the multilevel inverter with series sub-multilevel topology with
selective harmonic reduction is having the minimum THD and hence this combination would
be the best for all applications where multilevel inverter is to be used
COMPARISON OF MULTILEVEL TOPOLOGIES WITH... 485

REFERENCES

[1] Kangarlu Mohammad Farhadi,”A Generalized Cascaded Multilevel Inverter Using Series
Connection of Submultilevel Inverters,” IEEE Trans. On power electronics, VOL.28, NO.2,
February 2013, pp.625-635
[2] Sirisukprasert Siriroj, Lai Jih-Sheng, Liu Tian-Hua, “Optimum Harmonic Reduction with a
Wide Range of modulation Indexes for Multilevel Converters,” 0-7803-6401-5/00/$10.00 ©
2000 IEEE, pp.2094-2098
[3] Rodriguez J., Lai J. S., and Peng F. Z., “Multilevel inverters: A survey of topologies, controls,
and applications,” IEEE Trans. Ind. Electron., vol. 49, no. 4, pp. 724–738, Aug. 2002.
[4] Kim J. H., Sul S. K., and Enjeti P. N., “A carrier-based PWM method with optimal switching
sequence for a multilevel four-leg voltage-source inverter,” IEEE Trans. Ind. Appl., vol. 44, no.
4, pp. 1239–1248, Jul./Aug. 2008.
[5] Lopez O., Alvarez J., Doval-Gandoy J., Freijedo F. D., Nogueiras A., Lago A., and Penalver C.
M., “Comparison of the FPGA implementation of two multilevel space vector PWM
algorithms,” IEEE Trans. [Link]., vol. 55, no. 4, pp. 1537–1547, Apr. 2008.
[6] Boora A. A., Nami A., Zare F., Ghosh A., and Blaabjerg F., “Voltagesharing converter to supply
single-phase asymmetrical four-level diode clamped inverter with high power factor loads,”
IEEE Trans. PowerElectron., vol. 25, no. 10, pp. 2507–2520, Oct. 2010.
[7] Babaei E., “A cascade multilevel converter topology with reduced number of switches,” IEEE
Trans. Power Electron., vol. 23, no. 6, pp. 2657–2664, Nov. 2008
[8] Babaei E., Hosseini S. H., Gharehpetian G. B., Haque M. Tarafdar, and Sabahi M., “Reduction
of dc voltage sources and switches in asymmetrical multilevel converters using a novel
topology,” Elsevier J. Electric Power Syst. Res., vol. 77, no. 8, pp. 1073–1085, Jun. 2007.
[9] Nami A., Zare F., Ghosh A., and Blaabjerg F., “A hybrid cascade converter topology with
series-connected symmetrical and asymmetrical diodeclamped H-bridge cells,” IEEE Trans.
Power Electron., vol. 26, no. 1, pp. 51–64, Jan. 2011.
[10] Tolbert L. M., Peng F. Z., and Habetler T. G., “Multilevel converters for large electric drives,”
IEEE Trans. Ind. App., vol. 35, no. 1, pp. 36–44, Jan./Feb. 1999.
[11] Babaei E., “Optimal topologies for cascaded sub-multilevel inverters,” J. Power Electron., vol.
10, no. 3, pp. 251–261, May 2010.
[12] Peng F. Z., Lai J-S, “Multilevel Converters - A New Breed of Power Converters,” IEEE Trans.
on Ind. Appl., Vo1.32, No.3, MaylJune, 1996, pp.509-517.
[13] Nabae A., Takahashi I., Agaki H., “A New Neutral-Point- Clamped PWM, Inverter,” IEEE
Trans. on Ind. Appl. [Link]- 17, No.5, Sep./Oct., 1981, pp.518-523.
[14] Bhagwat P. M. and Stefanovic V. R., “Generalized Structure of a Multilevel PWM Inverter,”
IEEE Trans. on Ind. Appl., [Link]-19, No.6 Nov./Dec., 1983, pp.1057-1069
[15] Patel H. S., Hoft R. G., “Generalized Techniques of Harmonic Elimination and Voltage Control
in Thyristor Inverter: Part IHarmonic Elimination,” IEEE Trans. on Ind. Appl., [Link].9,No.3,
May/Jun., 1973, pp.310-317.
[16] Sun J., Grotstollen H., “Solving Nonlinear Equations for Selective Harmonic Eliminated PWM
Using Predicted Initial Values,” in Proc. IECON 1992, pp.259-264.
[17] Rashid M. H., Power Electronics: Circuits, Devices and Applications, 3rd ed. Upper Saddle
River, NJ: Pearson Education, 2003.
International [Link] [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 487-500

COMPARATIVE ANALYSIS OF PID AND FUZZY

PI CONTROLLER FOR SPEED CONTROL
OF DC MOTOR

MANJUNATHA H M1, KUMARASWAMY L2

AND PURUSHOTHAMA G K3
1
Department of Electrical & Electronics Engineering, RIT, Hassan.
2, 3
Department of Electrical & Electronics Engineering, MCE, Hassan.

Abstract
Direct current motor drives are an important drive configuration for many applications across a wide
range of powers and speeds. The ease of control and excellent performance of the DC motors will
ensure the continuous increase in number of applications using them. The adaptation of standard
microcontrollers, are the common approaches for real-time fuzzy logic implementation to ensure
flexible and intelligent control. This paper discusses the design of a DC motor speed control system
by Fuzzy Logic using the ATMEGA 8535 microcontroller. A comparison of the results with
conventional PID controller illustrates the effectiveness and elegance of the approach.

---------------------------------
Keywords: PID controller, Fuzzy PI controller, Fuzzy Associative Memory, PMDC, Microcontroller.

© http: //[Link]
488 MANJUNATHA H M, KUMARASWAMY L AND PURUSHOTHAMA G K

1. INTRODUCTION
DC motors are widely used in drive systems because of their decoupled and easy to apply
control characteristics. The speed control is one of the important aspects of any drive
system. This requires, speed, torque and even direction of rotation to be changed to meet
new conditions. In controlling the speed of a DC motor, speed regulation during transient
loading conditions is to be improved.
This work uses the fuzzy logic approach to the speed control of DC motor. The Fuzzy logic
based drive system potentially offers distinct advantages because of the flexibility and
versatility. The simple fuzzy controller will provide a real-time man-machine interface.
The fuzzy logic control approaches are useful when the system model is inaccurate,
incomplete or unknown. All fuzzy logic systems have modules to perform three essential
processes: the fuzzification of input data, an inference engine, and the defuzzification
process which produces a crisp output [1].
Due to their linguistic nature, fuzzy logic systems are typically realized by coding in a high
level language. This approach is commonly used in industrial and process control situations,
especially in embedded control applications. It uses standard microcontroller hardware
programmed with a set of subroutines to perform real-time computations that implement the
fuzzy logic reasoning algorithms. While the microcontroller platform is better suited for
direct interface in an embedded system, the real-time computations typically require
simplified data set representations in an effort to maintain a computationally efficient
algorithm [2].
This work attempts to design a DC motor speed control system by Fuzzy Logic using the
microcontroller-ATMEGA 8535. The simulated performance results are compared with
conventional PID controller. The speed control is performed in closed-loop form with a
PID/Fuzzy PI controller. The controller is expected to maintain the speed at desired level,
when there is a variation of load. Required PWM pulses are generated by PID/Fuzzy PI
controller based on the error signal. These pulses are used to drive DC motor through driver
circuit to maintain the desired speed.

2. FUZZY CONTROLLER DESIGN

The conventional controller is used to improve the speed response of the DC motor. In some
applications transient behavior of the DC motor must meet time domain parameters such as
COMPARATIVE ANALYSIS OF PID AND FUZZY PI… 489

rise time, settling time, overshoot and oscillations. In critical applications, these parameters
should not exceed the specified limits. In conventional controller good speed response can be
achieved to some extent by tuning the proportional gain factor, integral gain factor and
derivative gain factor. However, in certain applications the constraints on time domain
parameters are too high that can not be met with conventional controllers. In this situation
fuzzy logic based modeling provides best solution. Fig.1 shows the typical speed responses
using conventional controller and Fuzzy PI controller [3].
Fuzzy logic is a rule based approach that has flexibility in the design of rules, shape and
selection of membership functions. Thus, it is possible to achieve better results using fuzzy
based speed control of DC motor.

Fig.1: Typical Responses of Conventional PID and Fuzzy PI controller.

A conventional PI controller uses an analytical expression of the following form to compute

the control action:
u(t) = Kpe(t) + Ki ∫e(t)dt (1)
where, Kp is the proportional gain factor,
Ki is the integral gain factor,
e(t) is the error signal
490 MANJUNATHA H M, KUMARASWAMY L AND PURUSHOTHAMA G K

Differentiating (1) with respect to t, derivative of control action is:

(t) = Kp (t) + Ki e(t) (2)
Discrete time version of equation (2) can be written as:
Δu(k) = Kp Δe(k) + Ki e(k) (3)
where, Δu(k) is change in control output,
k is the kth sampling time,
e(k) is the error.
Change in control action in terms of current control output and previous control output is
given by:
Δu(k) = u(k) - u(k-1) (4)
where,
u(k) is control output at kth sampling instant,
u(k-1) is the control output prior to u(k).
e(k) = yd - y(k) (5)
where, yd is the desired output,
y(k) is the actual system output and
Δe(k) is the change of error.
Δe(k) = e(k) - e(k-1 (6)
where, e(k-1) is the error prior to e(k).

3. STRUCTURE OF FUZZY PI CONTROLLER

The fuzzy PI control is used to achieve the desired response. The input to the fuzzy
controller is error (e) and change in error (Δe), where as output variable is change in
control output as per the equation (3). Input variables error e(k), change in error Δe(k) and
output variable change in control output Δu(k) are normalized to same domain between[-
700, 700]. The crisp inputs are fuzzified and based on the Mamdani’s inference method
fuzzy outputs are obtained. These fuzzy outputs are defuzzified to get a crisp output from
the processor that generates the pulses required to achieve the desired speed response. The
block diagram of fuzzy PI controller is shown in the Fig.2.
COMPARATIVE ANALYSIS OF PID AND FUZZY PI… 491

Fig.2: A block diagram of a fuzzy PI control system.

4. DERIVATION OF THE RULES

Typical speed response of the motor is as shown in Fig.3. It has been partitioned into 6
different distinct regions to be used to form the fuzzy linguistic rules, which are then
represented by fuzzy sets and fuzzy logic operators. It is necessary to observe the behavior of
the motor speed over these different sections of the response, in order to determine whether
change, ∆u(k), in controlled input to the driver circuit is to be increased according to the
speed error and change in the speed error. The sign of ∆u(k) should be positive if ∆u(k)
is required to be increased, and it should be negative otherwise. This simple rule used to
determine the sign of ∆u(k) is applied as following manner.
Typical speed response of motor shown in Fig.3 is partitioned in to 6 different distinct
regions to be used to form the fuzzy linguistic rules, which are then represented by fuzzy sets
and fuzzy logic operators. It is necessary to observe the behavior of the motor speed over
these different sections of the response, in order to determine whether change, ∆u(k), in
controlled input to the driver circuit is to be increased according to the speed error and
change in the speed error. The sign of ∆u (k) should be positive if ∆u (k) is required to
492 MANJUNATHA H M, KUMARASWAMY L AND PURUSHOTHAMA G K

be increased, and it should be negative otherwise. This simple rule used to determine the
sign of ∆u(k) is applied as following manner.

Fig.3: Typical Speed Response of Motor.

The rules effectively reflect a typical scenario for operation of the system. For instant, at
time t=0, the system starts at output value y=0 and expected to reach some desired value yd.
e(k)=1(from Equation( 5)). i.e. Error is positive big(PB) and change in error ∆e(k) is also
large positive(LP)(from equation(6)). There fore change in control action ∆u(k) is Big
decrease(BD) (from equation (3)).Assume that output reached a mark(1), where the error is
still large positive(LP) but ∆e(k) is negative (N) and hence, change in control action is
increase(I). Repeat this procedure to determine the rest of the rules. The rules are listed in the
form of table based on the procedure mentioned above and the list is as shown Table 1.
Table 2 shows the controller fuzzy variables employed.
COMPARATIVE ANALYSIS OF PID AND FUZZY PI… 493

Table 1: Fuzzy Associative Memory for the fuzzy controller.

∆e/e NB N Z P PB

LN BI BI I D BD
L BI I H H BD
Z BI I H D BD
P BI H H D BD
LP BI I D BD BD

Table 2: Fuzzy Controller’s Variables.

Input variables Output variable

Error ‘e’ Error change ‘∆e’ Control Action
(Duty Cycle)
NB: Negative Big LN: Large BD: Big Decrease
N: Negative Negative D: Decrease
Z: Zero N: Negative H: Hold
P: Positive Z: Zero I: Increase
PB: Positive Big P: Positive BI: Big Increase
LP:Large Positive

5. THE PROOF OF DC MOTOR CONTROL SYSTEM

In the present work, conventional PID and fuzzy PI controllers are implemented to control
PMDC motor speed at desired speed and the results are compared to assess the performance
of both types of controller. The closed-loop DC motor speed control is as shown in the block
diagram of Fig. 4.
494 MANJUNATHA H M, KUMARASWAMY L AND PURUSHOTHAMA G K

PC

RS 232

LCD
KEY KEYPAD
AT MEGA
8535 DRIVER CIRCUIT

PROXIMITY SENSOR

DC MOTOR

MOTOR LOAD

Fig. 4: Block diagram of the fuzzy based DC motor.

The major units are computer, microcontroller, keypad, LCD, Driver circuit, DC motor and
Proximity sensor [4]. The desired speed yd of the motor is set using keypad. Start command
is applied to the microcontroller through start key. Microcontroller accepts this command
and generates proper pulses at the driver circuit. The proximity sensor placed near the disk
sense the speed and provides feedback for the microcontroller. On receiving the feedback
signal, error signal is obtained comparing feedback signal with desired speed. Using this
error signal change in error signal is generated and both error signal e(k) and change in error
signal Δe(k) are given as crisp inputs to fuzzy PI controller. This fuzzy PI controller initiate
change in control output ∆u(k) required to control the speed of the DC motor at desired
speed. However, during few seconds of starting the motor runs around desired speed and
after few seconds speed response of the motor reaches the desired speed. Fig. 5 shows
experimental setup of PMDC motor based on fuzzy PI system.

6. RESULTS AND DISCUSSIONS

Experiments have been carried for conventional PID controller and Fuzzy PI Controller [5-6]
for different speeds to study the performance of the system. For different reference speeds,
the system is tested under NO-load, and ON-load.
COMPARATIVE ANALYSIS OF PID AND FUZZY PI… 495

Fig.5: Shows experimental setup of the system.

PID controller and Fuzzy PI controllers are implemented using software in the
microcontroller to run the DC motor at desired speed. At different speeds and load
conditions, the results are recorded. The graph of speed versus time is plotted to study the
performance of the system using conventional and fuzzy controller.
The Figs. 6-9 shows the speed response of motor at different conditions using both PID and
fuzzy PI controllers. Fig.6 shows output response of motor using PID controller with speed
rising from zero to predetermined speed command of 450RPM. From the response it is
observed that, the steady rise is achieved with overshoot of 30.2%, rise time of 6 Sec. and
settling time of 13 Sec.
Fig.7 shows the output response of motor using Fuzzy PI controller with speed rising from
zero to predetermined speed command of 450RPM. From the response it is observed that,
the steady rise is achieved with overshoot of 25.7%, rise time of 3 Sec. and settling time 11
Sec.
496 MANJUNATHA H M, KUMARASWAMY L AND PURUSHOTHAMA G K

Fig.6: Speed response of PID controlled motor running at 450 RPM with no load.

By observing speed responses of PID and Fuzzy PI controller, it can be seen that steady rise
is achieved with almost small overshoot and the speed settles to desired speed in less time in
fuzzy PI controller compared to PID controller.
Fig.8 shows the speed response of the motor using PID controller to the load applied while
the motor is under running. From the response it is observed that, the motor has overshoot of
21.3% and has a settling time of 12 sec. When the motor is loaded the speed drops and it has
taken 14 sec time to reach the desired speed.
COMPARATIVE ANALYSIS OF PID AND FUZZY PI… 497

Fig. 7: Speed response of Fuzzy PI controlled motor running at 450RPM with no load.

Normal load

Fig.8: Speed response of PID controlled motor running at 450 RPM with load.

Fig. 9 shows the speed response of the motor using Fuzzy PI controller to the load applied
while the motor is under running. From the response it is observed that, the motor has
minimum overshoot of 15.55% and has a settling time of 11 sec. When the motor is loaded
the speed drops and it has taken 11 sec time to reach the desired speed.
498 MANJUNATHA H M, KUMARASWAMY L AND PURUSHOTHAMA G K

Normal load

Fig. 9: Speed response of Fuzzy PI controlled motor running at 450 RPM with load.

By observing speed responses of PID and Fuzzy PI controller, it can be seen that when load
is applied to the motor the time required to reach desired speed is less in Fuzzy PI controller
compared to PID controller.

Fig.10: Speed response of PID controlled motor running at 450 RPM started with load.
COMPARATIVE ANALYSIS OF PID AND FUZZY PI… 499

The Fig.10 shows the speed response of the motor using PID controller and with initial
loading for 450rpm speed. From the response it is observed that time took for steady rise is
more with frequent overshoot of 9.33% and has settling time of 12sec.

Fig. 11: Speed response of Fuzzy PI controlled motor running at 450 RPM started with load.

The Fig. 11 shows the speed response of the motor using Fuzzy PI controller with initial
loading for 450rpm speed. From the response it is observed that the time took for steady rise
is more with frequent overshoot of 0.4% and has settling time of 7 sec.
By observing speed responses of PID and Fuzzy PI controller, it can be seen that when
motor is started with load the time required to reach desired speed is less in Fuzzy PI
controller compared to PID controller.

CONCLUSIONS
The paper is designed and implemented to control the speed of PMDC motor using fuzzy PI
control also PID controller is used to control speed of the PMDC motor and results of PID
controlled closed loop system is compared with fuzzy PI controlled loop system. When
motor is running at no load, the steady rise is achieved with almost small overshoot and the
speed settles to required speed in less time in fuzzy PI controller compare to PID controller.
When load is applied to the motor the time required to reach desired speed is less in Fuzzy PI
500 MANJUNATHA H M, KUMARASWAMY L AND PURUSHOTHAMA G K

controller compared to PID controller. When motor is started with load the time required to
reach desired speed is less in Fuzzy PI controller compared to PID controller.
Thus it can be concluded that fuzzy based speed control of DC motor is proved better control
than the conventional controller. Both conventional and fuzzy controllers are implemented
using software, thus no additional cost is involved to implement fuzzy based DC motor
speed control system.

REFERENCES

[1] Timothy J. Ross, “Fuzzy Logic with Engineering Applications”, Second Edition 2009, Wiley
and Sons.
[2] Kumaraswamy L. and Manjunatha H.M. “Fuzzy Logic Based Speed Control of DC Motor With
Real Time Interface Using 8-Bit Microcontroller” International conference on emerging trends
in engineering (ICETE-12).
[3] Montiela Oscar and Sepúlveda Roberto "Performance of a Simple Tuned Fuzzy Controller and
a PID Controller on a DC Motor" Proceedings of the 2007 IEEE Symposium on Foundations of
Computational Intelligence.
[4] Sen P. C. and MacDonald M. L., "Thyristorized DC Drives with Regenerative Braking and
Speed Reversal". IEEE Transactions on Energy Conversion, 1978, Vol. IECI-25, No. 4: 347-
354.
[5] Dr. Meenakshi M., “Microprocessor Based Digital PID Controller for Speed Control of D.C.
Motor” First International Conference on Emerging Trends in Engineering and Technology.
[6] Mountain Jeffrey R., “Fuzzy Logic Motor Speed Control With Real-Time Interface Using an 8-
Bit Embedded Processor”, 42nd South Eastern Symposium on System Theory University of
Texas at Tyler, Tx, Usa, March 7-9, 2010.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 501-521

IMPLEMENTATION AND OPTIMIZATION OF CORDIC

DESIGN ON MODULATION AND DEMODULATION

ANANT PUKALE1 AND RAMYA H. R2

1
Digital Communication Engineering, VTU
2
Telecommunication Engineering, VTU
MSRIT, Bangalore, India

Abstract
Rotation of vectors through fixed and known angles has wide applications in robotics, digital signal
processing, graphics, games, and animation. But, we do not find any optimized coordinate rotation
digital computer (CORDIC) design for vector-rotation through specific angles. Therefore, in this
paper, we present optimization schemes and CORDIC circuits for fixed and known rotations with
different levels of accuracy. For reducing the area- and time-complexities, we have proposed a
hardwired pre-shifting scheme in barrel-shifters of the proposed circuits. Two dedicated CORDIC
cells are proposed for the fixed-angle rotations. In one of those cells, micro-rotations and scaling are
interleaved, and in the other they are implemented in two separate stages. Pipelined schemes are
suggested further for cascading dedicated single-rotation CORDIC units for high-throughput and
reduced latency implementations. We have the optimized set of micro-rotations for fixed and known
angles. The derived optimized scale-factors from a set of micro rotations and dedicated shift-add
circuits are used to implement the scaling. We have synthesized the proposed CORDIC cells using
Xilinx field programmable gate-array platform and shown that the proposed design offer higher
throughput and less area-delay product than the reference CORDIC design for fixed and known angles
of rotation.

-------------------------------------
Keywords : Coordinate rotation digital computer (CORDIC),numerically controlled oscillator(NCO),
look up table (LUT) © http: //[Link]
502 ANANT PUKALE AND RAMYA H. R

1. INTRODUCTION
The name CORDIC is nothing but iterative based mathematics for coordinated rotation of
digital computer. In 1959, Volder[1][2] was the scientist who developed CORDIC iterative
form of computation for trigonometry functions , multiplication and division. Over the
period of time many application of CORDIC has been formed with variety of algorithm for
high performance and low cost hardware solutions[3]-[12]. This paper stresses on design of
CORDIC in terms fixed angle rotation and implementation of same on modulation and
demodulation units.
For known and fixed angel of vector rotation has wide applications in graphics, game,
robotics and robotics [4],[13],[14]. Modulation and demodulation can be performed by
successive vector rotation through fixed angels and translation of the links. The translation
operation is realized by simple addition of old coordinated values to obtain new coordinates
for rotational steps to be accomplished by suitable successive rotation of CORDIC circuit for
fixed rotation through know small angels[4]. There are plenty of examples of uniform
rotation starting from electrons inside an atom to the planets and satellites. A simple example
of uniform rotations is the hands of an animated mechanical clock which perform one degree
rotation each time. There are several cases where high-speed constant rotation is required in
games, graphic, and animation. In modeling, simulation, animation and games are area
where objects with constant rotation are more often used. Dedicated CORDIC circuits are
used to implement efficiently through known small angle rotation. In some areas like signal
processing, engineering application, communication by multiplication of complex number
with a known complex constants. Previously, CORDIC circuits were developed for
implanting complex multiplication which is more often used in digital signal processing
(DSP) application[16]-[18], but there is no detailed information with respect to efficiency of
CORDIC realization of fixed and known angle rotations and constant complex
multiplication.
Latency of computation is the major issue with the implementation of CORDIC algorithm
due to its linear-rate convergence [19]. It requires (n+1) iterations to have -bit precision of
the output. Overall latency of computation increases linearly with the product of the word-
length and the CORDIC iteration period. The speed of CORDIC operations is, there- fore,
constrained either by the precision requirement (iteration count) or the duration of the clock
period. The angle recoding (AR) schemes [5]–[9] could be applied for reducing the iteration
IMPLEMENTATION AND OPTIMIZATION OF CORDIC… 503

count for CORDIC implementation of constant complex multiplications by encoding the

angle of rotation as a linear combination of a set of selected elementary angles of micro
rotations. In the conventional CORDIC, any given n rotation angle is expressed as a linear
combination of values of elementary angles that belong to the set {(σ. Arctan(2-r)):σ ϵ {-1,1}
to obtain an -bit value of
.However, in AR methods, this constraint is relaxed by adding zero into the linear
combination to obtain the desired angle using relatively fewer terms of for ϵ
{1,0,-1}. The elementary angle set (EAS) used by the AR scheme is given by S EAS=
Hu and Naganathan [5] have proposed an AR
method based on the greedy algorithm that tries to represent the remaining angle using the
closest elementary angle ±arctan2-i. .Using this recoding schemes the total number of
iterations could be reduced to less than half of the conventional CORDIC algorithm for the
same accuracy. Wu et al. [7] have suggested an AR scheme based on an extended
elementary-angle-set (EEAS), that provides a more flexible way of decomposing the target
rotation angle. . EEAS has better recoding efficiency in terms of the number of iterations and
can yield better error performance than the AR scheme based on EAS. But the iteration
period for EEAS is longer, and involves double the numbers of adders/subtractors in the
CORDIC cell compared with that of the other. Most of the advantages gained in the AR
schemes are cancelled out by the hardware and time involved in scaling the pseudo-rotated
vector.
It is desirable to have exhaustive search than greedy search for elementary angle set (EAS)
due to the fact that angle of rotation for fixed rotation case is known a priori. Moreover, the
barrel shifter complexity is reduced to almost half of that of a CORDIC. So thus we have
proposed some techniques to minimize complexity of barrel shifter. As CORDIC is an
sequential process it is desirable to use pipelined implementation which is yet to be exploited
but this makes it out of parallel process execution.
This paper tries to present optimization schemes for reducing the number of iterations of
micro rotation and even reducing the complexity of barrel shifter for vector rotation in case
of fixed angle. We also aiming for cascaded pipelined circuit for future use, which is
expected to be involving less area delay complexity and faster than the existing approaches.
The contribution of paper are listed as below
1. Implementation of fixed angle vector rotation with optimized set of micro-rotation.
504 ANANT PUKALE AND RAMYA H. R

2. Scaling circuits are derived from shift-add operation.

3. To reduce barrel shifter complexity, a hardware pre-shifting scheme is been introduced.
4. Single rotation CORDIC circuits are designed.
The remainder of this paper is organized as follows. Section II deals with the optimization of
elementary angle set for different accuracies of implementation. Efficient circuits for
implementation of micro-rotations for fixed rotations are presented in Section III.
Implementation of micro rotation. Scaling optimization and implementation is discussed in
Section IV .Application on modulation and demodulation is discussed. Section V application
are presented and in VI comparion of results VII conclusion is presented.

2. OPTIMIZATION OF ELEMENTARY ANGLE SET

The rotation-mode CORDIC algorithm to rotate a vector U=[UXUY]T through an angle Φ to
obtain a rotated vector V=[VXVY]T is given by [1], [2]

Such that when n is sufficiently large

Where if σi = -1 then Φi < 0 and σi=1 otherwise, and is the scale-factor of the CORDIC
algorithm, given by

For fixed angle rotation case, the pre-computed value Φi and σi indicating sign bit can be
stored in sign bit register (SBR) in CORDIC circuit. This during the CORDIC iteration,
CORDIC circuit need not compute rest of angle Φi.[3].
IMPLEMENTATION AND OPTIMIZATION OF CORDIC… 505

Fig. 1. Reference CORDIC circuit for fixed rotations

As shown in fig.1. and with respect to (1a) and (1b) a reference CORDIC circuit for fixed
rotation X0 and Y0 can b fed as set or reset input in to the input register and hence later
feedback values Xi and Yi are even fed parallel to input register at ith iteration. So with the
conventional concept of CORDIC we feed in initial values X 0 and Y0 as well as feedback
values Xi and Yi via a pair of MUX.
Here we can find a set of small number of pre-determined elementary angle {αi, for 0 ≤ i ≤
m-1}, where αi = arctan (2-k(i)) for known and fixed angle rotation of vector through rotation
mode CORDIC circuit which are used in ith micro rotation of CORDIC algorithm(1), where
m is minimum number of micro-rotation. The rotation through any angle in 0 ≤ θ ≤ 2π, can
be mapped on to positive rotation 0 ≤ Φ ≤ π/4 without much arithmetic operation[10]. Hence
to perform optimization rotation mapping is done so that rotation angle lies in the range of 0
≤ Φ ≤ π/4. Thus in next step of optimization, the elementary angles {αi} as according to the
requirement of accuracy. Thus (1) can be modified as given in (3a) and (3b)

Such that minimum number m

506 ANANT PUKALE AND RAMYA H. R

Where the scale factor K depends on the set {αi}. now thus the accuracy of CORDIC
algorithm depends on how closely the resultant rotation ΦA due to all micro-rotation in (1)
approximated and desired rotation Φ, which are in turn factor to determine the deviation
actual rotation angle from estimated value.
Algorithm 1: it gives the simple pseudo code to optimize a set of micro-rotation. ϵΦ is the
maximum accuracy which is defined as maximum tolerable error between approximated
angle and desired angle is given as input, the parameters k(i) and σi are searched in
optimization algorithm such that the objective function ΔΦ minimizes. The algorithm begins
with the first micro rotation i.e. m=1, if the micro rotation smaller angle of deviation than ϵΦ
cannot be found, m is increased by 1 and optimization algorithm is increased by 1. In
algorithm exhaustive search is employed in entire parameter space for all combination of k(i)
and σi . Based on obtained result on micro rotation, the scaling parameters are searched with
different objective function. If optimal set of micro rotations cannot satisfy design constraint
for scaling, then sub-optimal set of micro-rotations are used. As shown in Table 1. For angle
310 and 350 are obtained by sub-optimal solutions because scaling requires more terms in
these two cases if optimal way is used.

In the experiment with the maximum input angular deviation єΦ=0.040 , I found that a set of
four selected micro-rotations is enough. In Table I, it is shown that rotations through any
angle in the range 0<Φ<=π/4 (in odd integer degrees) could be achieved with maximum
angular deviation ΔΦ=0.0370, where . ΔΦ =|Φ-ΦA| Using a maximum of two selected micro-
rotations, the rotations could be achieved with maximum angular deviation with ΔΦ=1.8750
(0.033 radian). In case of six micro-rotations, angular deviation ΔΦ could be reduced to
~0.0005.
IMPLEMENTATION AND OPTIMIZATION OF CORDIC… 507

Table 1 : Optimization of Full Rotations with Four Micro-Rotations

In Table II, it is shown further that rotations through 0.10<|Φ|<20 in an interval of 0.1 could
be obtained by four micro-rotations with angular deviation, ~0.0030. Here we can make an
observation that we can always achieve higher accuracy with more number of micro-
rotations. From Table II, we find that higher accuracy could be achieved in case of small
rotation angles like 10 or 20, compared to the most of the larger angles when the same
number of micro-rotations is used.

3. IMPLEMENTATION OF MICRO-ROTATIONS
For a given angle of rotation the elementary angles and direction of micro rotation are
already predetermined, the angle estimation data-path is not required in the CORDIC circuit
for fixed and know rotations. The control bits can be stored in ROM as only few micro
rotation are involved corresponding to control bits. Fig.2. shows the circuit for complex
constant multiplication. The barrel shifter is used to obtain number of required bit shift
corresponding control bits and direction of micro rotation is stored
508 ANANT PUKALE AND RAMYA H. R

Table 2 : Optimization Of Small Rotations With Four Micro-Rotations

Fig. 2. CORDIC cell for constant complex multiplications

IMPLEMENTATION AND OPTIMIZATION OF CORDIC… 509

in sign-bit register (SBR). The major contributors to the hardware- complexity in the
implementation of a CORDIC circuit are the barrel-shifters and the adders. There are several
options for the implementation of adders [22], from which a designer can always choose
depending on the constraints and requirements of the application. But, we have some scope
to develop techniques for reducing the complexity of barrel-shifters over the conventional
designs as discussed in the followings

Fig. 3. Hardwired pre-shifting in basic CORDIC module.

3.1 Minimization of Barrel-Shifter Complexity by Hardwired Pre-Shifting
A barrel sifter having [log2(S+1)] stages demultiplexors , where S is maximum number of
shifts required on word length L, where each stage will have L number of 1:2 MUXes. Thus
hardware complexity increases logarithmically with number of shifts and linearly with word
length. We can reduce number of stages of MUXes and effective word length of MUXes of
barrel shifter by as shown in fig.3. Hardwired pre-shifting in basics CORDIC. So thus load
only (L - l) MSB’s of an input to barrel shifter, where l is minimum number of shifts
required, anyhow l LSB’s are truncated while shifting. If s is maximum number of shifts in
the set selected micro rotation then the barrel shifter require only (s-l) shifts. The add/sub
unit will receive (L- l) LSB’s from barrel shifter and the l MSB’s of corresponding operand
of add/sub unit are hardwired to [Link] thus reducing hardware complexity of barrel shifter by
hardwired pre-shifting method. In turn it reduces time involved in barrel shifter by hardwired
510 ANANT PUKALE AND RAMYA H. R

pre-shifting, since delay of barrel shifter is due number of stages of MUXes, and its also
possible to reduce number of stages by hardwired pre-shifting.
More than 75% cases in table-1 require more than one shifts for minimum number of shift l.
where as in table-2 we find that l is always greater than five except at angle 1.50. the
hardware complexity of barrel shifter can be substantially reduced by hardwired pre-shifting
method. If number shifts are known priori, one design barrel shifter for particular number of
shifts. For implementation of 4 shifts (see table1) irrespective of maximum number of shifts,
the barrel shifter would require only 3 stages of MUXes of 2:1 by hardwiring the shifts.
3.2 High-Throughput Implementation Using Cascaded Multi-Stage CORDIC
If we observe carefully table1 and table2, the implementation of small rotations has the
minimum of 5≤ l. thus with an advantage of hardwired pre-shifting, if separate module of
CORDIC cascaded in a pipelined fashion, it would provide high throughput. To implement
CORDIC with higher degree of accuracy without affecting the throughput of result, we can
cascade multistage CORDIC, each stage consisting of single rotation cells.

Fig. 4. (a) Multi-stage single-rotation cascaded CORDIC circuit. (b) Structure of ith rotation
module.  s(i) indicates right-shift by bit-locations.
IMPLEMENTATION AND OPTIMIZATION OF CORDIC… 511

3.3Cascaded CORDIC with Single-Rotation Cells

As depicted in fig.4. the multiple stages of cascaded pipelined CORDIC circuit with each
stage performing specific rotation in a dedicated CORDIC module. The functional block of
each stage is as shown in fig.4(b). Each stage of rotation module will have pair of add/sub
units. To each add/subtract unit one of pair of input is loaded directly where as other input is
pre-shifted form (L-s(i)) LSB location, where s(i) being number of right shifts required in
the implementation of ith micro-rotation. The s(i) MSB locations are set to 0. The rotation
module does not require any barrel shifter as only 1 fixed micro-rotation is required and the
SBR are also not required as each module does either add or subtract operation. As show in
fig. 4 the output of preceding stage is taken as input to succeeding stages. In this module
critical-path amounts only one add/sub computation in adder module. Thus latency of n-
stage single rotation cascaded is n(TA+TFF) , where TFF and TA are D-flip flop delay and
addition/subtraction time, respectively.
More than two-third of rotation angles in table1 require 3 micro rotations for maximum
deviation of Φ up to 0.040. Using only 3 stages of CORDIC circuit as shown in fig 4 even
the complex multiplications can be implemented involving 3 micro rotation (n=3). This can
also be implemented using non pipelined form of (n-1) carry propagate adders having total
latency of where TFA and TA are, respectively, full adder delay and the
time required for L-bit addition-time. k(i) is the number of shifts of the ith stages, , L being
word-length of implementation.

4. SCALING OPTIMIZATION AND IMPLEMENTATION

To match with the optimized set of elementary angles for the micro-rotations, the
optimization of scaling factor is discussed here.
4.1 Scaling Approximation for Fixed Rotations
The generalized form of expression for the scaling factor as given in (2) can be explicitly
expressed for the selected set of micro-rotations as

where k(i) for 0 ≤ i < m1 is the number of shifts in the ith micro rotation. Except for k(i)=0
(i.e., rotation by 450 ), by binomial expansion, any term in (4) can be written as
512 ANANT PUKALE AND RAMYA H. R

Where x=2-2i , i being the number of shifts in a micro rotation, and can be expressed
alternatively in terms of i as

Replacing each term in (4) by the expression of (6), we can obtain an approximate scale-
factor as a product of shift-add terms of form

Where m2 is maximum number of scaling iterations required for the approximation and s(i)
is the number of shifts performed for the ith iteration of scaling, δi = ±1.
The number of accountable terms of (6) depends on desired degree of output accuracy with
an approximated scale factor KA [see (7)] is limited by number of micro rotation used. For
six micro-rotation CORDIC implementation, where the error in Φ is ~0.00050 , the first two
terms in (6) contribute for (0≤i≤8) , while up to the third, fourth and fifth terms contribute
for (0≤i≤3), (0≤i≤2), and (0≤i≤1) , respectively and similarly it can be found that for four
micro-rotation CORDIC implementation, where the error in Φ is ~0.040 , only the first two
terms in (6) contribute for (0≤i≤4) , while up to the third and the fifth terms contribute for
(0≤ i ≤ 2) and (0≤ i ≤ 1) , respectively. Accordingly, we have obtained the recursive shift-
add expressions of scale-factor KA in the form of (7).
Here is algorithm2 gives optimization method to search the parameter k(i) and σi. on
knowing the set of micro rotation from algorithm1,from (4) the ideal scaling factor can be
computed. The objective function ΔK=[1- KA/K] gives the deviation of KA/K from 1. The
maximum tolerable deviation ϵK is set to be same that of algorithm1 ϵΦ. Here the algorithm
begins with the single term of scaling, then is increased by 1 until ΔK becomes smaller than
ϵK..
IMPLEMENTATION AND OPTIMIZATION OF CORDIC… 513

Table 3: Optimized Shifts to Implement Scaling for the Case of

Rotation with Four Micro-Rotations
514 ANANT PUKALE AND RAMYA H. R

We derive here the expression of scale factors separately for 43 and 45 rotations to get
scaling with desired accuracy with less number of iterations compared with the above
approach.

Fig. 5. Shift-add scaling circuit using hardwired pre-shifted loading.

To have an accuracy up to , the scale-factor for rotation through 43 and 45 can
be expressed as

Equation (8) in recursive shift add forms

4.2 Implementation of Scaling

Micro rotations and scaling can be implemented on 2 separate stages or on same circuit.
Desired level of accuracy affects the implementation of scaling and micro rotation, and even
the scaling implementation depends on implementation of micro rotation. Therefore the
realization of scaling circuit corresponding to different implantation of micro rotation are
discussed here.
4.2.1 Generalized Implementation of Scaling
According to (7) the shift add circuit is shown in fig.5. for minimizing the barrel shifter
complexity the scaling circuit of fig.5 can be hardwired and could placed after CORDIC
circuit of fig.2 to perform micro rotation and scaling in 2 separate stage. The generalized
CORDIC circuit for fixed rotation is as shown in fig.6 can perform scaling and micro
rotation in interleaved manner of alternate cycles. The cell of fig.6 is similar to that of fig.2.
IMPLEMENTATION AND OPTIMIZATION OF CORDIC… 515

an addition line changer circuit is introduced to change un-shifted input path. As shown in
fig.(6b) the structure and functional block of line changer.

Fig. 6. CORDIC circuit for interleaved implementation of micro-rotations as well as scaling

circuit. (a) The CORDIC circuit. (b) Structure and function of line-changer. For control-bit=1 it
performs micro-rotations and for control-bit=0 it performs the shift-add operations for scaling.

4.2.2 Implementation of Scaling for Cascaded Single-Rotation CORDIC

Fig. 7. Shift-add circuit for single-rotation-cascaded-scaling using hardwired pre-shifting.

516 ANANT PUKALE AND RAMYA H. R

The dedicated pair of adder sub-tractors as shown in fig.7. it does not require any sign bit
register or multiplexor. A pair of input are fed to add/sub unit with one of input fed directly
where as other input is shifted to right by s(i) locations. The shift-add term sign factor will
decide the computation of either addition or subtraction.
5. APPLICATION
5.1 Numerically controlled oscillator (NCO)
IQ modulation is widely used in wireless communication in order to convert the base band
signal to pass band. In a typical wireless transmitter system the base band signal is converted
to pass band signal of required carrier frequency and then applied to a DAC after which the
analog wave is transmitted in air via an antenna. This wave is received by the antenna in the
receiver system and is converted to a digital signal by using an ADC. Then this pass band
signal is converted back to base band via demodulation schemes for further processing. NCO
is basically a digital oscillator which is used to generate synchronous sine and cosine carrier
waves which are used in modulation and demodulation processes in wireless
communication. It consists of a phase accumulator and a waveform generation unit as shown
in the following fig 8.

Fig8: numerically controlled oscillator block diagram

IMPLEMENTATION AND OPTIMIZATION OF CORDIC… 517

Fig 9: modulation block diagram

Phase increment = (Fo * 2^m)/Fclk

Where Fo is the desired output frequency, m is the accumulator precision and Fclk is the
clock frequency.
The above diagrams depict modulation and demodulation schemes. Equation for modulation
and demodulation are as follows:
mod_data = (i_data * nco_cos) + ( q_data * nco_sin)
The phase increment is applied as the input is accumulated every clock cycle inside the
phase accumulator. The waveform generation unit can be a memory based, multiplier based
or CORDIC based. For our application, CORDIC is used as the waveform generation unit.
The CORDIC algorithm computes the sine and cosine of an input phase value by iteratively
shifting the phase angle to approximate the Cartesian coordinate values for the input angle.
At the end of the CORDIC iteration, the x and y coordinates for a given angle represent the
cosine and sine of that angle, respectively. These sinusoidal waves are used to modulate the
base band I and Q signals and demodulate to get back I and Q signals as well.
demod_i_data = mod_data * nco_cos
demod_q_data = mod_data * nco_sin
The same NCO which has been used for modulation should be used for demodulation as
well. Otherwise wrong base band signals appear after demodulation. The demodulation
518 ANANT PUKALE AND RAMYA H. R

process is followed by a low pass filter (LPF) to get back the original base band I and Q
signals.

Fig 10: demodulation block diagram

6. COMPARISON RESULTS

Fig 11: comparison graph

IMPLEMENTATION AND OPTIMIZATION OF CORDIC… 519

parameters Radix 4 cordic Proposed

processor work
Number of slice flip flop 936 318
Number of 4 input LUTs 899 299
Number of used logic 494 251
Number of used shift register 405 123
Number of input output 65 8
Number of bonded IOBs 63 7
Power (mW) 175.9 91

Fig 12: comparison table

The numerically controlled oscillator as show in fig 11, the chart manhattans graph depicts
that number of flip flop required are reduced almost 3 times in proposed system. Number of
look up table required are also reduced drastically almost 3 times in proposed system and
power requirement is also reduced almost half in proposed system

CONCLUSION
The proposed CORDIC as used in modulation and demodulation of incoming in-phase and
quadrature phase data are modulated using sine and cosine waves generated by CORDIC
waveform generator and a reverse process is performed to retrieve the original fed data. The
above fig12: table of comparison shows that number of flip flop , number of 4 input LUT ,
number of shift register and power(mW) are drastically reduced so thus improving
throughput Less latency, Less ADP (Area Delay Product = Number of LUTs x CLK Period
of design). As show in above comparison table it is clear that the number of resources
required is drastically less than the conventional radix 4 CORDIC process than the processed
system.

REFERENCES

[1] Volder J. E., “The CORDIC trigonometric computing technique,” IRE Trans. Electron.
Comput., vol. EC-8, pp. 330–334, Sep. 1959.
[2] Walther J. S., “A unified algorithm for elementary functions,” in Proc. 38th Spring Joint
Comput. Conf., 1971, pp. 379–385.
520 ANANT PUKALE AND RAMYA H. R

[3] Hu Y. H., “CORDIC-based VLSI architectures for digital signal processing,” IEEE Signal
Process. Mag., vol. 9, no. 3, pp. 16–35, Jul. 1992.
[4] Meher P. K., Valls J., Juang T.-B., Sridharan K., and Maharatna K., “50 years of CORDIC:
Algorithms, architectures and applications,” IEEE Trans. Circuits Syst. I, Reg. Papers, vol. 56,
no. 9, pp. 1893–1907, Sep. 2009.
[5] Hu Y. H. and Naganathan S., “An angle recoding method for CORDIC algorithm
implementation,” IEEE Trans. Comput., vol. 42, no. 1, pp.99–102, Jan. 1993.
[6] Hu Y. H. and Chern H. H.M., “A novel implementation of CORDIC algorithmusing backward
angle recoding (BAR),” IEEE [Link]., vol. 45, no. 12, pp. 1370–1378, Dec. 1996.
[7] Wu C.-S., Wu A.-Y., and Lin C.-H., “A high-performance/low-latency vector rotational
CORDIC architecture based on extended elementary angle set and trellis-based searching
schemes,” IEEE Trans. Circuits Syst. II, Analog Digit. Signal Process., vol. 50, no. 9, pp. 589–
601, Sep. 2003.
[8] Juang T.-B., Hsiao S.-F., and Tsai M.-Y., “Para-CORDIC: Parallel cordic rotation algorithm,”
IEEE Trans. Circuits Syst. I, Reg. Papers, vol. 51, no. 8, pp. 1515–1524, Aug. 2004.
[9] Rodrigues T. K. and Swartzlander E. E., “Adaptive CORDIC: Using parallel angle recoding to
accelerate CORDIC rotations,” in Proc. 40th Asilomar Conf. Signals, Syst. Comput. (ACSSC),
2006, pp. 323–327.
[10] Maharatna K., Banerjee S., Grass E., Krstic M., and Troya A., “Modified virtually scaling free
adaptive CORDIC rotator algorithm and architecture,” IEEE Trans. Circuits Syst. for Video
Technol., vol. 15, no. 11, pp. 1463–1474, Nov. 2005.
[11] Kang C. Y. and Swartzlander E. E., Jr., “Digit-pipelined direct digital frequency synthesis based
on differential CORDIC,” IEEE Trans. Circuits Syst. I, Reg. Papers, vol. 53, no. 5, pp. 1035–
1044, May 2006.
[12] Vachhani L., Sridharan K., and Meher P. K., “Efficient CORDIC algorithms and architectures
for low area and high throughput implementation,” IEEE Trans. Circuits Syst. II, Exp. Briefs,
vol. 56, no. 1, pp. 61–65, Jan. 2009.
[13] Blundell B. G., An Introduction to Computer Graphics and Creative 3-D Environments. London,
U.K.: Springer-Verlag, 2008.
[14] Lang T. and Antelo E., “High-throughput CORDIC-based geometry operations for 3D computer
graphics,” IEEE Trans. Comput., vol. 54, no. 3, pp. 347–361, Mar. 2005.
[15] Cani M.-P., Igarashi T., and Wyvill G., Interactive ShapeDesign. San Rafael, CA: Morgan &
Claypool, 2007.
[16] Jones K. J., “2D systolic solution to discrete Fourier transform,” IEEE Proc. Comput. Digit.
Techn., vol. 136, no. 3, pp. 211–216,May 1989.
[17] Meher P. K., Satapathy J. K., and Panda G., “Efficient systolic solution for a new prime factor
discrete Hartley transformalgorithm,” IEEE Proc. Circuits, Devices Syst., vol. 140, no. 2, pp.
135–139, Apr. 1993.
[18] Freeman S. and O’Donnell M., “A complex arithmetic digital signal processor using cordic
rotators,” in Proc. IEEE Int. Conf. Acoust., Speech, Signal Process. (ICASSP), 1995, pp. 3191–
3194.
[19] Cavallaro J. R. and Luk F. T., “CORDIC arithmetic for a SVD processor,” J. for Parallel Distrib.
Comput., vol. 5, pp. 271–290, 1988.
IMPLEMENTATION AND OPTIMIZATION OF CORDIC… 521

[20] Lin C. H. and Wu A. Y., “Mixed-scaling-rotation CORDIC (MSRCORDIC) algorithm and

architecture for high-performance vector rotational DSP applications,” IEEE Trans. Circuits
Syst. I, Reg. Papers, vol. 52, no. 11, pp. 2385–2396, Nov. 2005.
[21] Park S. Y. and Cho N. I., “Fixed-point error analysis of CORDIC processor based on the
variance propagation formula,” IEEE Trans. Circuits Syst. I, Reg. Papers, vol. 51, no. 3, pp.
573–584, Mar. 2004.
[22] Omondi A. R., Computer Arithmetic Systems: Algorithms, Architectures, and Implementations.
New York: Prentice-Hall, 1994.
[23] Jain A. K., Fundamentals of Digital Image Processing. Englewood Cliffs, NJ: Prentice-Hall,
1989.
[24] Signal Compression : Coding of Speech, Audio, Text, Image And Video,ser. Selected Topics in
Electronics and Systems, N. E. Jayant, Ed. Singapore: World Scientific, 1997, vol. 9.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 523-536

OPERATION AND CONTROL OF FULL-RATED

CONVERTER WIND TURBINES

TROAMBAKA DATTA
Instrumentation and Control System India,
Mumbai, India.

Abstract
The present day scenario in the world energy market is challenged by factors such as protecting the
environment by controlling emission of pollutants and taking steps to reduce carbon footprints by
reducing consumption of fossils fuels and opting for power generated by renewable sources such as
solar and wind energy. Renewable energy sources (wind farms and solar plants) along with small
scale generators like fuel cells are spread across the power system, in contrast, to large power plants
which are concentrated sources of large power production and hence, have been categorised as
sources of Distributed Generation (DG). [7] However, it must be noted that renewable energy
technologies differ greatly from the generating stations based on synchronous plants and issues such
as stability (for example, contribution towards network damping) and security of the power grid (such
as capability to provide ancillary services) must be taken into consideration. [8][15] This paper is
focused on wind energy technology with advanced power electronic interfaces to integrate them with
the power grid. The operational characteristics of full-rated converter (FRC) based wind turbines and
the various control strategies it employs are investigated. For the demonstration of FRC wind
turbine’s capability to respond to network faults, the detailed model of the “Synchronous Generator
and Full Converter (Type 4)” in Sim Power Systems was used.

© http: //[Link]
524 TROAMBAKA DATTA

1. BACKGROUND
India is a fast developing country in terms of population, urbanization and industrial
development which are resulting in higher demand for electricity each year. This has resulted
in a rapid growth of the energy sector. However, the energy sector has failed to meet the
demand and the country suffers from shortages. [1] In July 2012, the country faced blackouts
in 20 states (out of 28) when three of the five grids failed. [2] By 2030, it is estimated that
the electricity demand will triple. [3] India must exploit all its resources to generate enough
electricity to meet the growing demands. Coal remains the main source of electricity
generation and further exploitation of oil and gas resources is needed to contribute to the
supply side. Nuclear power plants (6 plants in operation and 5 under construction) [4] are
becoming popular as an efficient and reliable source of electricity. Hydro electricity is the
most widely used renewable source of electricity in India which can be further exploited to
meet the demand. [5] Among renewable energy sources, wind energy is becoming steadily
popular placing India as the third largest annual wind power market. [6]
Along with factors such as depleting fossil fuel reserves, the present day the world energy
market is challenged by factors such as protecting the environment by controlling emission
of pollutants and taking steps to reduce carbon footprints by reducing consumption of fossils
fuels and opting for power generated by renewable sources such as solar and wind energy.

2. INTRODUCTION
The increasing domestic and industrial loads demand greater production of power. The
present energy trends have shifted its focus towards the integration of renewable energy,
among which wind energy has emerged as the most potent resource of renewable energy.
Massive developments in the installation of offshore wind farms are now being undertaken
to achieve renewable targets by 2020 and 2030 throughout the world.
Integration of large amount of power from renewable sources is seen to cause complexity in
the management of the power grid. [8]Wind farms are expected to provide the same level of
services as the conventional synchronous generator based plants in terms of providing
security and stability to the grid. With the increase in the number of wind farms, system
operators are now demanding for greater Grid Code compliance such as having fault ride
through capability. [9] Thus, for the penetration of wind energy into the power system on a
large scale, it is important to conduct studies related to the contribution of wind farms
OPERATION AND CONTROL OF FULL-RATED CONVERTER WIND TURBINES 525

towards power network stability. Full rated converter type wind turbines have emerged as
the best wind turbine technology that has the potential to follow the Grid Codes and
contribute towards the support of the power grid.

3. FULL-RATED CONVERTER WIND TURBINES

This section explores the concepts used for the conversion of wind energy into electric
power through the use of wind turbines. Although there are different wind turbine
technologies existing today, the discussion in this paper is limited to full converter wind
turbine (WT) topology.
In this configuration, the generator and the mechanical drive train is decoupled from the grid
by the converters. The generated power flows through the converter to the grid. Both
synchronous (permanent) and asynchronous generators can be used in this type of
configuration. Generator torque and speed control are provided through the converters.
The main advantages of using FRC architecture is as follows:
 Operates over variable wind speeds
 Generator is decoupled from the grid
 Mechanical shocks are reduced on the turbine during grid faults
 Better grid code compliance capability
 Higher annual power yield
 Full control over active power and reactive power production
3.1 Application of Power Electronics in FRC WT architectures
Power Converters is the most important element that integrates the operation of the wind
turbine with the power grid. These devices are used to convert electrical power (from AC to
DC and vice versa) along with changing parameters like voltage and frequency levels. There
are two main kinds of converter arrangements used in FRC architecture. One of them uses an
uncontrolled diode rectifier with a DC booster in the generator side controller as shown in
fig 1. The DC booster stabilises the DC link voltage. The network side converter consists of
a pulse-width modulated voltage source converter (PWM-VSC). [10]
526 TROAMBAKA DATTA

Fig 1: Schematic diagram of PMSG wind turbine with diode-booster arrangement [10]

Fig 2 : VSC with 3 phase diode bridge [13]

The second configuration uses to two back-to-back PWM-VSCs in its converter controls as
shown in fig. 3.2.

Fig 3: Schematic diagram of PMSG wind turbine with two back-to-back voltage source
converters [10]
OPERATION AND CONTROL OF FULL-RATED CONVERTER WIND TURBINES 527

Fig 4: Back-to-back VSC [13]

Here, the generator is controlled by the generator-side converter and the DC link voltage is
controlled by the network-side converter.
It is important to mention that even though voltage source inverters produce a periodic
output, it is not a pure sine wave. It resembles a staircase-like signal as illustrated below in
fig 5.

Fig 5: Inverter voltage waveform from self-commuted devices [13]

528 TROAMBAKA DATTA

In order to generate a sine wave conforming to the grid parameters, the classical scheme of
pulse width modulation (PWM) technique is used. In this technique, a reference signal which
varies sinusoidally is compared with a fixed-frequency triangular carrier waveform to create
a switching pattern as shown in fig 6. The transistor switches on whenever the triangle wave
becomes less than the reference sine wave. The transistor gets switched off when the triangle
wave becomes greater than the reference sine wave. Similar, action takes place in the second
half of the cycle.

Fig 6: Pulse width modulation control waves [13]

The switching frequency ranges from 8 to 20 kHz and is limited to avoid excessive
switching losses. PWM inverters in wind turbines make use of devices such as IGBT and
GTO. Harmonics arise due to the switching nature of these devices and most often filters are
employed to reduce their adverse effects. [13]
3.2 Control strategies
The main control objectives of the FRC wind turbine are power control, grid synchronisation
and monitoring techniques and wind turbine logic and safety as represented in fig 7. There
are specific algorithms which drive the operation of both the generator side converter and
grid side converter. [11] The control modules, namely, power control, converter module and
grid synchronization are discussed below.
OPERATION AND CONTROL OF FULL-RATED CONVERTER WIND TURBINES 529

Fig 7: Control levels in FRC wind turbine [11]

3.2.1 Power Control

The wind turbine is controlled by an aerodynamic system based on a pitch control system.
The output power is controlled based on the power tracking curve set by the dispatch centre
to maximise power production. Below rated power, the speed of the wind turbine varies
according to the wind speed by maintaining the pitch angle. The pitch controller also limits
the output power when the wind speed is above rated speed. [11]
3.2.2 Converter controller
The converter control deals with the control of both the generator side converter and grid
side converter. There are several control strategies that are used for the power converters
such as synchronous Voltage Oriented Control (VOC) with PI controllers, stationary VOC
with Proportional Resonant controllers, Current control, Space vector modulation, Adaptive
band hysteresis etc. Out of these the synchronous Voltage Oriented Control (VOC) with PI
controllers is widely used for grid applications. In this control scheme, coordinate
transformation between the stationary αβ and the synchronous dq reference frames is
performed. This provides fast transient response due to the presence of internal current
loops. The decomposition of the AC currents in two axes provides a decoupled control for
the active and reactive power. The current in d-axis control the active power output and the
DC link voltage and the q-axis current controls the reactive power. Fig 8 illustrates this
control scheme. [11]
530 TROAMBAKA DATTA

Fig 8: Schematic diagram of the VOC in synchronous reference frame [11]

The coordinate transformation is done by using a Phase Locked Loop (PLL). The
performance of this control strategy depends on the accuracy of the PLL system for
estimating the voltage grid angle. [11]
3.2.3 Grid synchronization
The wind turbine control requires an algorithm that can detect the phase angle of grid
voltage in order to synchronize the delivered power. Phase angle is an important component
in the control scheme as it is used to transform the feedback variables to a suitable reference
frame. Hence, phase angle detection has a significant role in the control of grid side
converter. There are several algorithms for detecting the grid phase angle such as zero
crossing detection, use of arctan function or PLL technique. [11]
3.3.4 Grid monitoring
Grid code requirements dictate the operation conditions of the generating units connected to
it with respect to the voltage and frequency values. This condition is in reference to the fault
ride through capability of the wind turbine which defines the time interval within which the
wind turbine might disconnect itself. This requires a fast and reliable grid monitoring
system. PLL structures are used in grid monitoring system. The grid voltage information is
obtained by performing the Clarke transformation as show in fig 9.
OPERATION AND CONTROL OF FULL-RATED CONVERTER WIND TURBINES 531

Fig 9: Grid monitoring structure based on PLL [11]

4. MODEL OF WIND FARM AND DISTRIBUTION NETWORK

For the purpose of our analysis, the detailed model of the “Synchronous Generator and Full
Converter (Type 4)” which is modelled in SimPowerSystems was used. The model
comprises of a wind farm exporting power to the grid via a distribution system. The wind
farm is made up of five 2MW wind turbines. The 10 MW wind farm is then connected to a
25 kV distribution through a step-up transformer which converts the wind farm voltage of
525 kV to 25 kV. The distribution system is connected to a 120 kV grid through a 30 km 25
kV feeder. The circuit is shown in the figure below.

Fig 10: Detailed model of Wind farm with FRC wind turbine
532 TROAMBAKA DATTA

The FRC wind turbine incorporated in the circuit is made up of a synchronous generator.
The generator side converter consists of a diode rectifier with DC-DC IGBT based PWM
boost converter. The network side converter consists of a voltage source IGBT based PWM
inverter which performs the DC-AC conversion. A part of the internal structure of the FRC
wind model is shown below.

Fig 11: Internal structure of the FRC wind turbine

The generator side converter is responsible for controlling the excitation of the generator and
the speed and pitch control. Speed regulation is achieved by taking the power tracking curve
as the reference input signal and comparing it with the measured rotor speed and regulating
it through PI blocks. The duty cycle of the diode converter is also controlled through
appropriate current regulators. The duty cycle pulse is used to control the boost circuit. Pitch
control is performed by similar means. The network side converter controls the DC link
voltage and the active and reactive power output to the grid. Here, it uses vector control
strategy to meet its control objectives. Regulation of the DC link voltage and the reactive
power is performed to obtain the current components in the d-axis and q-axis. Through
further processing, the current components are used for active power and reactive power
regulation.

5. CIRCUIT MODEL AND SIMULATION

5.1 Steady state system levels
The circuit is modelled to operate at a fixed rated speed of 15m/s. The generator side
converter maintains the rotor speed at 1 pu. The reactive power production is maintained at 0
OPERATION AND CONTROL OF FULL-RATED CONVERTER WIND TURBINES 533

MVAR. At steady state with no faults, the wind farm produces 10 MW power and the DC
link voltage is maintained at 1100V.
5.2 Fault simulation
A fault was induced in the system in the form of voltage sag at the grid side (120 kV
system). A six-cycle 0.10 pu voltage drop is programmed at t= 0.03s by changing the
parameters of the 120kV voltage source block as shown in the figure below in fig. 6.3. The
level of voltage sag and the time it persists can be easily modified. The simulation is
performed and the change in system parameters in observed. The change in the system
behaviour is observed at t=0.03 s when the fault is made to occur. It is observed that there is
a drop in the active power output of the wind turbine and an increase in the DC link voltage.
The reactive power changes very lightly.

Fig 12: Change in system parameters after the fault

534 TROAMBAKA DATTA

From fig. 12 it can be seen that the control system of the wind turbine tries to regulate the
DC link voltage and the reactive power during the time the fault persists in the system. The
rotor speed of the generator is observed to be steady which demonstrates that the generator
operation is decoupled from the grid side fault. The system regains its initial state after the
fault is eliminated. Through these observations, the reliability of the FRC wind turbines is
further proved.

CONCLUSION
From the study and analysis of wind turbine technology that has been presented in this
report, it is clear that FRC wind turbines are most suited for installation in wind farms built
for large scale power production. The control schemes of FRC wind turbines decouple the
generator operation from the grid and this characteristic feature makes it suitable for many of
the grid requirements such as fault ride through, voltage support etc. As the generator is
decoupled from the effects of faults and other variations in the grid, it can function without
suffering torque oscillations due to grid disturbances.
The performance of the FRC wind turbine can be further increased by incorporating
additional devices and control loops. This can constitute addition of devices such as power
system stabiliser (PSS) in the control loop of the wind turbine can greatly enhance its
capabilities. The inclusion of PSS has shown to improve the damping capability of the FRC
wind turbine and makes the whole power system more stable. [16][17] The performance of
drive train torsional vibration damper is further improved by incorporating state based filter
models. The addition of drive train dampers in the control loop makes the whole drive train
more stable and avoids breakdowns. This greatly increases the lifetime and overall
performance of the wind turbine. [18][19][20] The simulations performed on the FRC wind
turbine model available in SIMULINK show how it supports the power network during
faults by changing its active power output and changing the level of its DC link voltage. It
also shows how the rotor speed of the generator is not affected during the fault, thus making
the decoupling between the generator and grid clearer.
OPERATION AND CONTROL OF FULL-RATED CONVERTER WIND TURBINES 535

REFERENCES

[1] “Energy statistics 2012- 19th Issue”, Section 6.4: Consumption of Electricity; Central Statistics
Office, Ministry Of Statistics And Programme Implementation, Government of India, New
Delhi. ([Link])
[2] “India blackouts leave 700 million without power”, The Guardian.
([Link]
[3] “India Wind Energy Outlook 2012”, Pg 7, Global Wind Energy Council. ([Link])
[4] “Energy Scenario and Vision 2020 in India”, Pg 11, P Garg, Journal of Sustainable Energy &
Environment 3 (2012) 7-17, Ministry of Environment and Forests, India.
([Link]
[5] “Energy Markets and Technologies in India”, R.V. Shahi, Secretary, Government of India,
Ministry of Power. ([Link]
[6] “India Wind Energy Outlook 2012”, Foreword, Global Wind Energy Council. ([Link])
[7] The Institution of Engineering and Technology, “Distributed Generation”,
[Link] Retrieved on 10/01/2014
[8] Tan Kelvin, Choy Valerie, Zhang Ming Hui, “Key Technical Issues for Renewable Energy Grid
Integration and Deployment in Asia”, [Link]
Kelvin_Tan_T1_S3_REWA_2012.pdf Retrieved on 23/08/2013
[9] Tsili M., Patsiouras Ch. and Papathanassiou S., “Grid Code Requirements for Large Wind
Farms: A Review of Technical Regulations and Available Wind Turbine Technologies”,
National Technical University of Athens (NTUA), School of Electrical and Computer
Engineering.
[10] Olimpo Anaya-lara, Nick Jenkins, Janaka Ekanayake, Phill Cartwright, Mike Hughes “Wind
Energy Generation: Modelling and Control”, Chapter 6.
[11] Iov F., Ciobotaru M., Blaabjerg F.,” Power Electronics Control of Wind Energy in Distributed
Power Systems”, Aalborg University, Institute of Energy Technology.
[12] Blaabjerg F., Chen Z., Teodorescu R., Iov F.,” Power Electronics in Wind Turbine Systems”,
Aalborg University, Institute of Energy Technology.
[13] Manwell J. F., McGowan J. G. and Rogers A. L., “Wind Energy Explained- Theory, Design and
Application, Second Edition, pg.237-246.
[14] Olimpo Anaya-lara, Nick Jenkins, Janaka Ekanayake, Phill Cartwright, Mike Hughes “Wind
Energy Generation: Modelling and Control”, Chapter 2.
[15] Zhang F., Leithead W.E and Anaya-lara O, “Wind Turbine Control Design to Enhance the Fault
Ride-Through Capability”, Department of Electronic and Electrical Engineering, University of
Strathclyde, Glasgow, UK.
[16] Jos´e Luis Dom´ınguez-Garc´ıa, Oriol Gomis-Bellmunt, Fernando Bianchi, Andreas Sumper,
“Power System Stabilizer Control for Wind Power to Enhance Power System Stability”,
PHYSCON 2011, Le´on, Spain, September, 5–September, 8 2011.
[17] Caliao N.D., Ramtharan G., Ekanayake J., Jenkins N., “Power oscillation damping controller for
fully rated converter wind turbines”, UPEC2010, 31st Aug - 3rd Sept 2010.
536 TROAMBAKA DATTA

[18] Wright A.D. and Fingersh L.J., “Advanced Control Design for Wind Turbines Part I: Control
Design, Implementation, and Initial Tests”, Technical Report, NREL/TP-500-42437, March
2008, Chapter 2.
[19] Licari John, Ugalde-Loo Carlos E., Ekanayake Janaka B. and Jenkins Nicholas, “Damping of
Torsional Vibrations in a Variable-Speed Wind Turbine”, IEEE Transactions on Energy
Conversion, Vol. 28, No. 1, March 2013.
[20] Peeters J., Vandepitte D., Sas P., Leuven K.U., “Structural analysis of a wind turbine and its
drive train using the flexible multibody simulation technique”.
International [Link] [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 537-545

DC VOLTAGE CONTROL OF WECS FEEDING

AN INDUCTION MOTOR LOAD

ARJUN M1 AND A B RAJU2

1
MIT Manipal, 2B V B Hubli

Abstract
This paper presents the DC voltage control of Self Excited Squirrel Cage Induction Generator(IG)
feeding an IM load. This Induction machine is driven from a prime mover which is usually a wind
turbine. The stator terminals of the induction machine is connected to a back to back converter. The
converter is further to an Induction motor load. The capacitor is initially charged to 12V and it
provides initial magnetisation for the IG. Control strategy has been developed to keep the DC voltage
constant, and also to connect the IM load when the DC voltage reaches steady state reference value.

-----------------------------------
Keywords : Induction generator (IG), Voltage source converter (VSC), Sine Pulse width modulation
(SPWM), Wind Energy Conversion systems(WECS).
© http: //[Link]
538 ARJUN M AND A B RAJU

1 INTRODUCTION
Every activity needs energy. It is also available in different forms and can be converted from
one form to another. It is this property of energy that makes humans utilize it for every
activity. However the conversion of energy forms is not straightforward. Research is in
acceleration from past till present days to find better means of energy conversion and
utilization. Due to increase in human activity, the urge for energy conversion has caused
energy crisis in the present days. Therefore the research is now inclined towards harnessing
energy from renewable energy sources. Of the available energy sources, wind energy and
solar energy are the major energy sources for the production of electricity.
According to the Ministry of New and Renewable Energy(MNRE) reports of March 2013,
the installed capacity of WECS in India is 19 GW and the ministry holds an aspirational
target of 25000 MW by 2017. This shows that wind energy holds a prominent role in power
generation.
Induction Machines and Synchronous machines are widely used electrical machines for wind
energy conversion systems. Induction machines are robust, low cost and maintenance free
machines. They have ability for self protection against overloads and faults. They are
available in lower ratings and do not require additional DC supply unlike synchronous
machines. Hence in this paper induction machine is used as source of electrical energy.
Induction machines can be either grid connected or stand alone. Stand alone systems are
used where the load has no access to grid(remote areas). Stand alone induction generators
are also called as self excited induction generators. In grid connected mode the voltage and
frequency is fixed and decided by grid. However in stand alone systems, the voltage and
frequency depend on load, speed and other factors and hence analysis is not
straightforward[1].
A major disadvantage of induction machine is that it needs a source of excitation. A vast
literature is available on the methods of excitation. Excitation can be as simple as connecting
capacitor banks across stator terminals to using a complex voltage source converters[2]-[4].
Simple capacitor banks can only be used when the load and the prime mover speed is
constant. Therefore usage of power electronic converter becomes inevitable. In [5] a current
controlled VSC is used as a source of excitation. In [6] the author proposes a control scheme
using a hysteresis control and the behaviour of the system is analysed for sudden application
and removal of load. [7] presents a PWM converter with vector control scheme to control
DC VOLTAGE CONTROL OF WECS FEEDING… 539

DC bus voltage scheme. However, with the usage of complex control strategies, the
reliability of the system gets affected[8]. Hence it is important to choose a new methodology
which provides smoother capacitance control using power electronic converters along with
simple control schemes.
Induction motors are the most commonly seen loads. It is also true that 60% of the total
loads are induction motor loads. Keeping this in view, the load for the WECS is chosen as
IM. In this paper, wind energy conversion system feeding a IM load is discussed. It is shown
through simulations that the control strategy developed is able to maintain a constant DC
voltage across capacitor. The induction motor load is connected to WECS only after the DC
link voltage reaches the reference value. This is to avoid the damage caused to the load
during the overshoots in the system.

2. DESCRIPTION OF THE SYSTEM

Fig 1: Block diagram of the system

The system block diagram is shown in Fig 01. Induction machine is driven from a prime
mover. The stator terminals of the induction machine is connected to a three phase two level
voltage source converter. A capacitor of proper rating is connected to the other side of the
converter. The battery is charged with an initial voltage of 12V. This battery provides initial
excitation for the IG. A three phase balanced load is applied across the stator terminals of the
machine. Control strategy for maintaining a constant DC link voltage is developed. The
actual value of the DC link voltage is continuously monitored with the reference value, the
error is passed though a PI controller. A limiter sets proper slip frequency value. This slip
540 ARJUN M AND A B RAJU

frequency is added to the rotor frequency to get stator frequency. The harmonic oscillator
generates sinusoidal signals whose frequency is that of the stator frequency and these are
multiplied with suitable modulation index and compared with two 1.05kHz sawtooth waves,
and the pulse width modulated signal are given to the inverter switches.
When the machine starts running as a generator, the DC link voltage starts building up. DC
link voltage reaches the steady state value as set by the reference command. If the IM load is
connected to the supply during the charging of capacitor, it is more likely that the stator
draws more current to feed both motor and the capacitor. Also, during the
transients/overshoots, the motor develops a high electromagnetic torque(and hence current).
If this torque exceeds rated torque, the chances for the motor damage is more. Hence it is
important to connect the motor to the converter only after the transients die and the the
steady state condition is reached. Therefore control strategy is developed to close the switch
after a steady DC voltage is achieved.

3. MODELLING INDUCTION MACHINE

Induction machine is modelled in a stationary reference frame. The set of differential
equations from (1)-(5) describe the induction machine.
 ̇ψds  vds  ids
 ̇ψ =  v −Rs  i  (1)
 qs  qs  qs
 ̇ψdr  idr  0 1  ψdr
 ̇ψ =−Rr  i −ωr  −10  ψ  (2)
 qr  qr  qr
ψ
 ψds  0s L 0mL   ids
L 0L 0 i

 qs=  s m  qs (3)

 ψdr  Lm 0 Lr 0   idr
 ψqr  0 Lm 0 Lr   iqr
3P
te= 2 2 (ψdsiqs−ψqsids) (4)

dωr P
dt = 2J(te−tm) (5)
DC VOLTAGE CONTROL OF WECS FEEDING… 541

4. VOLTAGE SOURCE CONVERTER

Using switching functions(Sa,Sb and Sc ), expressions for inverter pole voltages(Vao,Vbo

and Vco) and phase voltages(Van, Vbn and Vcn) are given by

 vao vdc  Sa

v S

 bo= 2  b (6)
 vco   Sc
 van 1  2 −1−1  vao
v v

 bn= 3  −1 2 −1  bo (7)

 vcn  −1−1 2  vco
Inverter current is given by
iinv=Saia+Sbib+Scic (8)

5. DC LINK CAPACITOR

 − C  (iinv) when vdc<vref;

1
vdc=  1 (9)
 - C  (iinv-im) when vdc=vref.
where im is the motor current

6. RESULTS

Fig 2 : IM connected to WECS during transient period

542 ARJUN M AND A B RAJU

Fig 3 : IM disconnected to WECS during transient period

Fig 4 : Induction motor load change

7 . DISCUSSION OF RESULTS
7.1 IM connected to WECS during transient period
Fig 2 shows the system response when the IM is connected during the process of capacitor
charging. It is seen from the graph of Vdc that there is an overshoot and it takes a

considerable time to reach the reference voltage of 848V. The size of loads are considerably
DC VOLTAGE CONTROL OF WECS FEEDING… 543

lesser to generating units. If such loads are connected to the systems during transients, the
loads experience the overshoots across them. It is clear from results that, the stator voltage
Vam also contains overshoots, and hence the electromagnetic torque(and hence currents,

because torque is a function of current) shoots up to a very high value which will damage the
load. This has to be avoided. A load of 2Nm is applied to IM and the torque overshoots to
50Nm. Vag is the stator voltage of IG and Nr is the rotor speed in rpm of the IM load.

7.2 IM connected to WECS during steady state

Fig 3 shows the system response of the proposed model. IM is connected to the system by
closing the switch after the transients die. Control system for continuous monitoring of DC
link voltage is developed. Once the voltage level reaches a steady state, the switch is
automatically closed. This is done to ensure safer operation of the loads. From the graph of
Vdc, it is clear that the steady state is reached at t=9.5s. (A large settling time is chosen just

to make the results clearly visible). It is seen that till t=9.5s, the value of Vam, Nr and Te are

all zero. It means IM is isolated from the system. Once the steady state is reached the IM is
connected to the system and the speed gets developed. The load is now safe from the
transients.
7.3 Induction motor load change
Fig 4 shows the response of the system for the IM load change. The load on the motor was
changed at different time instants. Initially the motor is run at 2Nm. At t=15s, it was
suddenly loaded to 7Nm. At t=17s then load was decreased to 5Nm. It can be seen from the
graphs that the system responded dynamically for the load changes. The load changing was
done to ensure that system developed works for dynamic load variations. In all the above
cases, the DC link voltage was maintained constant.

CONCLUSION
In this paper a voltage control scheme for stand alone system employing back to back
converter feeding a IM load is studied. Control scheme for maintaining constant DC link
voltage is studied. Control strategy for monitoring the DC link voltage and to appropriately
connect the load is also studied. The system is tested for its response for dynamically
variation on IM loads. The system gives satisfactory performance in all the cases. The
simulation results validate the studies done.
544 ARJUN M AND A B RAJU

IG machine specifications
IM Parameters Values
phase 3
poles 4
Rated Power 5 HP
Rated Speed 1460 rpm
Rated Voltage 240 V
frequency 50Hz
Stator Resistance 2.0 Ω
Rotor Resistance 2.4 Ω
Xls 4.12 Ω
Xlr 4.12 Ω
moment of inertia of rotor J 0.1384 kg-m2

Voltage Source Converter Specifications

modulation index ma = 0.8

carrier frequency fc = 1.05KHz

IM machine specifications
IM Parameters Values
phase 3
poles 4
Rated Power 2 HP
Rated Speed 1440 rpm
Rated Voltage 240 V
frequency 50Hz
Stator Resistance 5.01 Ω
Rotor Resistance 5.76 Ω
Xls 8.22 Ω
Xlr 8.22 Ω
LmM 0.84 H
moment of inertia of rotor J 0.0642 kg-m2
DC VOLTAGE CONTROL OF WECS FEEDING… 545

REFERENCES

[1] Haque M H, “A Novel Method of Evaluating Performance Characteristics Of Self-Excited

Induction Generators”, IEEE Transactions on energy Conversions, Vol 24.

[2] Rai H. C. and Tandon A. K., “Voltage Regulation of SEIG Using Passive Elements”, 6th
European Conference on Electrical Machines and Drives, vol. 1, no. 1, pp. 240–245, 1983.

[3] Brennan M.B. and Abbondanti A., “Static Excitation for Induction Generator”, IEEE
Transactions on Industry Applications, vol. 13, no. 5, pp. 422–428, Sept/Oct, 1977.

[4] Malik N. H. and Al-Bahrani A. H., “Influence of the Terminal Capacitor on the Performance
Characteristics of SEIG”, IEE Proceedings of Institution of Electrical Engineering,Part C, vol.
137, no. 2, pp. 168– 173, March, 1990.

[5] Kuo S.-[Link] Wang L., “Analysis of voltage control for a self-excited induction generator using
a current-controlled voltage source inverter (CC-VSl)”, IEE Proc.-Genes. Transni. Distrrb, Vol.
148, No. 5, September 2001

[6] Jayaramaiah G. V. and Fernandes B. G., “Dynamic Simulation of Three Phase Self-Excited
Induction Generator With PWM-VSI”, National Conference on Control,Communication and
Informations Systems, vol. 1, no. 1, pp. 132–137,Jan,GOA, 2004.

[7] Amin Mahmoud M., Mohammed O. A., Fellow “DC-Bus Voltage Control of Three-Phase
PWM Converters connected to Wind Powered Induction Generator", 978-1-4244-6551-4/10.

[8] Jayaramaiah G. V. and Fernandes B. G., “Novel Voltage Controller for Stand-alone Induction
Generator using PWM-VSI”, IEEE Industry Applications Conference, pp. 204-208, Oct. 2006.

[9] BandaraKapila, Sweet Tracy, Ekanayake Janaka, “Photovoltaic applications for off-grid
electrification using novel multi-level inverter technology with energy storage”,ELSEVIER-
Renewable Energy 37,2012, pp.82-88.

[10] Rahim N.A., Selvaraj J., Krismadinata C., “Five-level inverter with dual reference modulation
technique for grid-connected PV system”, ELSEVIER-Renewable Energy 35, 2010, pp.712-720.

[11] Ramirez Dionisio, Martinez Sergio, Carrero Carmelo, Platero Carlos A., “Improvements in the
grid connection of renewable generators with full power converters”, ELSEVIER-Renewable
Energy 43 (2012), pp.90-100.

[12] Bose B K, Modern Power Electronics and AC drives, Printice - Hall 1998

[13] Raju A B, “Digital Simulation of Vector Controlled Induction Motor Drives: Some Case
Studies”, Proceedings of National Symposium on Power System Instrumentation and Control,
Dec. 1997.

[14] Mohan Ned, Undeland Tore M. and Robbins William P., Power Electronics Converters,
Applications, and Design, Third Edition, Wiley-India Edition

[15] Rashid Muhammad, Power Electronics Handbook, Academic Press 2001

International [Link] [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 547-558

VOLTAGE SAG/SWELL COMPENSATION BY DYNAMIC

VOLTAGE RESTORER

MISTRY RAM HEMANTKUMAR1 AND HEMIN D. MOTIWALA2

1
Electrical Department, Sarvajanik College of Engineering & Technology, Surat, India
2
Prof. Electrical Department, Sarvajanik College of Engineering & Technology, Surat, India

Abstract
In 21th century, the expansion of power system and electronic devices has been grown at very fast rate.
The most noticeable topic for electrical engineer is Power quality in recent years. Power quality
problem is an occurrence manifested as a nonstandard voltage, current or frequency that results in a
failure or a misoperation of end user equipments. With Power quality problem utility distribution
networks, industrial loads, sensitive loads etc. are suffered. To overcome the problem related to power
quality, Custom power devices are introduced. A number of power quality solutions are provided by
Custom power devices. Power quality encompasses several aspects: voltage sags and swells,
harmonics, over voltage, flicker, interruptions etc., lasting only a few cycles can cause significant
damage for a manufacturing process and computer hardware installations. At present, a wide range of
very flexible controllers, which capitalize on newly available power electronics components, are
emerging for custom power applications. Voltage source converter based Dynamic Voltage Restorer
(DVR) can be used effectively for mitigation of voltage sag and swell. This paper presents the operating
principle, power circuit topologies, mathematical modeling, control philosophies and application of
DVR for power quality improvement along with systematic review of published literature. The
modeling of DVR is carried out component wise and their performances are analyzed using MATLAB
software. The simulation shows that the control technique is very effective and yields excellent
compensation for voltage sag/swell mitigation.

------------------------------------
Keywords : Dynamic Voltage Restorer, Power quality problem, Voltage sag and swell, Voltage Source
Converter (VSC), series compensation, control system, sensitive load, simulation.
© http: //[Link]
548 MISTRY RAM HEMANTKUMAR1 AND HEMIN D. MOTIWALA

1. INTRODUCTION
The various power quality problems are due to the increasing use of non linear and power
electronic loads. Harmonics and voltage distortion occur due to these loads. Power quality
problems can cause malfunctioning of sensitive equipments, protection and relay system.
These problems include a variety of electrical disturbances, which may originate in several
ways and have different effects on various kinds of sensitive loads. As a result of this
vulnerability, increasing numbers of industrial and commercial facilities are trying to protect
themselves by investing in more sophisticate equipment to improve power quality. Voltage
magnitude is one of the major factors that determine the quality of power supply. Loads at
distribution level are usually subject to frequent voltage sags due to various reasons. Voltage
sags are highly undesirable for some sensitive loads, especially in high-tech industries. It is a
challenging task to correct the voltage sag/swell so that the desired load voltage magnitude can
be maintained during the voltage disturbances. The effect of voltage sag/swell can be very
expensive for the customer because it may lead to production downtime and damage. Voltage
sag/swell can be mitigated by voltage and power injections, absorption into the distribution
system using power electronics based devices, which are also known as custom power device.
Different approaches have been proposed to limit the cost causes by voltage sag/swell. One
approach to address the voltage sag/swell problem is dynamic voltage restorer (DVR). It can
be used to correct the voltage sag and swell at distribution level. Distribution system is mainly
affected by voltage sag and swell power quality issue. The simulation model is developed
using MATLAB SIMULINK in section V and discusses simulation results with different
sag/swell conditions. The main objective of this paper is to investigate and proposes a new
configuration of DVR in order to develop such device for voltage sag/ swells mitigation in the
network. This paper explores design and analysis of a novel Dynamic Voltage Restorer along
with controller for efficient stabilization and utilization.

2. VOLTAGE SAGS AND SWELLS

Voltage sag is a decrease of the normal voltage level between 10 and 90% of the nominal rms
voltage at the power frequency, for durations of 0.5cycle to 1 minute. Voltage swell, on the
other hand, is defined as an increase in rms voltage or current at the power frequency for
durations from 0.5 cycles to 1 minute. Typical magnitudes are between 1.1 and 1.8 up. Swell
magnitude is also described by its remaining voltage, in this case, always greater than 1.0.
VOLTAGE SAG/SWELL COMPENSATION BY DYNAMIC VOLTAGE RESTORER 549

IEEE 519-1992 and IEEE 1159-1995 describe the voltage sags/swells.

Figure 1: Definitions of events by IEEE Std.l159-1995

Voltage sags/swells caused by unsymmetrical line-to-line, single line-to-ground (SLG),
double-line-to-ground and symmetrical three phase faults effects on sensitive loads, short
circuits, lightning strokes, faults and inrush currents are the causes of voltage sags whereas
start/stop of heavy loads, badly dimensioned power sources, Switching off a large inductive
load or Energizing a large capacitor bank is a typical system, badly regulated transformers,
single line to ground fault on the system lead to voltage swells. Due to the fact that voltage
swells are less common in distribution systems, they are not as important as voltage sags.
Voltage sag and swell can cause sensitive equipment (such as found in semiconductor or
chemical plants) to fail, or shutdown, as well as create a large current unbalance that could
blow fuses or trip breakers. These effects can be very expensive for customers, ranging from
minor quality variations to produce downtime and equipment damage.

3. DYNAMIC VOLTAGE RESTORER

The Dynamic Voltage Restorer (DVR) is a Custom Power Device used to eliminate supply
side voltage disturbances. DVR also known as Static Series Compensator maintains the load
voltage at a desired magnitude and phase by compensating the voltage sags/swells and voltage
unbalances presented at the point of common coupling by connects a voltage source in series
with the supply voltage in such a way that the load voltage is kept inside the established
tolerance limits. Other than voltage sags and swells compensation, DVR also has added other
features like: line voltage harmonics compensation, reduction of transients in voltage and fault
550 MISTRY RAM HEMANTKUMAR1 AND HEMIN D. MOTIWALA

current limitations. The main function of the DVR is the protection of sensitive loads from
voltage sags/swells coming from the network. The DVR injects the independent voltages to
restore and maintained sensitive to its nominal value. The injection power of the DVR with
zero or minimum power for compensation purposes can be achieved by choosing an
appropriate amplitude and phase angle. The basic structure of a DVR is shown in Fig.1. It is
divided into five categories: (i) Voltage Injection Transformers: In a three-phase system,
either three single- phase transformer units or one three phase transformer unit can be used for
voltage injection purpose.

VDVR
Supply
Network Load
Vs VL

Filter

Control Energy
System VSC
Storage

Figure 2 : Basic structure of DVR

Three single phase injection transformers are connected in delta/open winding to the
distribution line. These transformers can be also connected in star/open winding. The
star/open winding allows injection of positive, negative and zero sequence voltages whereas
delta/open winding only allows positive and negative sequence voltage injection. (ii) Passive
Filters: Filters are used to convert the inverted PWM waveform into a sinusoidal waveform.
This is achieved by eliminating the unwanted harmonic components generated VSI action.
Higher orders harmonic components distort the compensated output voltage. (iii) Inverter: An
Inverter system is used to convert dc storage into ac form. These inverters have low voltage
ratings and high current ratings as step up transformers are used to boost up the injected
voltage. (iv) Energy Storage Unit: It is responsible for energy storage in DC form. Flywheels,
batteries, superconducting magnetic energy storage (SMES) and super capacitors can be used
as energy storage devices. It is supplies the real power requirements of the system when DVR
is used for compensation. (v) Capacitor: DVR has a large DC capacitor to ensure stiff DC
VOLTAGE SAG/SWELL COMPENSATION BY DYNAMIC VOLTAGE RESTORER 551

voltage input to inverter. In DVR configuration sometimes By-Pass Switch is used. If the over
current on the load side exceeds a permissible limit due to short circuit on the load or large
inrush current, the DVR will be isolated from the system by using the bypass switches and
supplying another path for current. Basic principal of DVR is to transfer the voltage sag and
swell compensation value from DC side of the inverter to the injected transformer after filter.
The compensation capacity of a particular DVR depends on the maximum voltage injection
capability and the active power that can be supplied by the DVR. When DVR’s voltage
disturbance occurs, active power or energy should be injected from DVR to the distribution
system. A DC system, which is connected to the inverter input, contains a large capacitor for
storage energy. It provides reactive power to the load during faulty conditions. Capacitor
terminal voltage reduces when energy is drawn from energy storage capacitors. Therefore,
there is a minimum voltage required below which the inverter of the DVR cannot generate the
require voltage thus, size and rating of capacitor is very important for DVR power circuit. The
most important advantage of these capacitors is the capability to supply high current pulses
repeatedly for hundreds of thousands of cycles.

4. DVR CONTROL TECHNIQUES

The compensation for voltage sag and swell using a DVR can be performed by
injecting/absorbing reactive power or real power. When the injected voltage is in quadrature
with the current at the fundamental frequency, compensation is achieved by injecting reactive
power and the DVR is self-supported with dc bus. But, if the injected voltage is in phase with
the current, DVR injects real power and hence a battery is required at the dc side of VSI. The
control technique adopted should consider the limitations such as the voltage injection
capability (inverter and transformer rating) and optimization of the size of energy storage. The
synchronous reference frame (SRF) theory is used for the control of self-supported DVR. The
dqo transformation or Park’s transformation is used to control of DVR. The dqo method
expresses the voltage error and phase shift information as instantaneous space vectors with
start and end times. The voltage is converted from abc reference frame to d-q-o reference. For
simplicity, zero phase sequence components is ignored. The detection of error in voltage is
carried out in each of the three phases. The control scheme for the proposed system is based on
the comparison of a reference voltage and the measured terminal voltage (Va, Vb, Vc). The
error signal allows generation of a commutation pattern by means of the sinusoidal pulse
552 MISTRY RAM HEMANTKUMAR1 AND HEMIN D. MOTIWALA

width modulation technique (SPWM). The injection voltage is generated by difference

between the reference load voltage and supply voltage and is applied to the VSC to produce
the preferred voltage, with the help of pulse width modulation (PWM). Fig 3 shows the basic
control scheme and parameters that are measured for control purposes. The control algorithm
produces a three phase reference voltage to the series converter that tries to maintain the load
voltage at its reference value. The d-reference component is set to a rated voltage and the
q-reference components are set to zero.

Vsa V0 Via*
V0
abc 0 dq0
Vib* Sinusoidal

-
To Vd Vd* To
Vsb PWM
dq0 LPF 1 Vq abc
transform V + 0 transformation Vic* Pulses to
Vsc q inverter
ation

PLL ᶿ ᶿ

Figure 3: Control strategy of DVR

5. SIMULATION RESULTS AND ANALYSIS

Table 1: System Parameters And Constant
Main Supply Voltage per Phase 415 V
Line Impedance Ls = 0.01 mH
Rs= 1.0 Ohm
Line Frequency 50 Hz
Series transformer turns ratio 1:1
DC Bus Voltage 700 V
Filter Inductance 2 mH
Filter capacitance 35 μF
Load P = 1 KW
QL = 5 KVAR
VOLTAGE SAG/SWELL COMPENSATION BY DYNAMIC VOLTAGE RESTORER 553

A detailed system as shown in Figure 3 has been modelled by MATLAB/SIMULINK to

study the efficiency of suggested control strategy. The system parameters and constant
value are listed in Table I. Figure 4 below shows the complete model of DVR constructed in
simulink environment (MATLAB). The model consists of source, inverter, control block,
EMI filter, Injection transformer, and load. The different supply voltage disturbances are
generated by using source. The inverter is used to convert DC supply to AC supply. The
output of the inverter contains fundamental voltage and the voltages of switching
frequencies and multiples of switching frequencies. The voltages of switching frequencies
and multiples of switching frequencies are eliminated by using EMI filter. The pulses are
generated by the comparator by comparing sinusoidal signal with triangular signal. The AC
voltage of the inverter is injected in each phase of the line by using injection transformers.
The reference signals to the PWM inverter are generated by using control block.

Continuous

powergui
A Vabc Vs A Vabc
A A A Iabc RL LOAD
Iabc
N B B B B a Goto6
B a A
C C C b b B
C c C c
THREE PHASE C
VOLTAGE SOURCE
2

INJECTING TRANSFORMER
1

1
a

FILTER
C
A

B
A

VSC

In1 Out1 In1

SWITCHING
CONTROL SYSTEM

Figure 4: DVR MATLAB/SIMULINK MODEL

554 MISTRY RAM HEMANTKUMAR1 AND HEMIN D. MOTIWALA

5.1 Voltage Sags

The first simulation show of three phase voltage sag is simulated. The simulation started with
the supply voltage 90% sagging as shown in Figure 5 (a).In Figure 5 (a) also shows a 90%
voltage sag initiated at 0.10s and it is kept until 0.20s, with total voltage sag duration of 0.1s.
Figures 5 (b) and (c) show the voltage injected by the DVR and the corresponding load voltage
with compensation. As a result of DVR, the load voltage is kept at 1 p.u. The effectiveness of
the DVR under unbalanced conditions is shown in figure 6, also shows the occurrence of 30%
single phase voltage sag on a utility grid. Through simulation the supply voltage with one
phase voltage dropped down to 30% as shown in Figure 6 (a). The DVR injected voltage and
the load voltage are shown in Figures 6 (b) and (c) respectively. Its corresponding load
voltages are shown in Figure 6(c) where it is possible to see that the compensation method is
keeping the load voltages constant at 1p.u.

(a) Supply Voltage

(b) Injected Voltage

VOLTAGE SAG/SWELL COMPENSATION BY DYNAMIC VOLTAGE RESTORER 555

(c) Load Voltage

Fig 5: Balance voltage sag

(a) Supply Voltage

(b) Injected Voltage

(c) Load Voltage

Fig 6: Unbalance voltage sag
556 MISTRY RAM HEMANTKUMAR1 AND HEMIN D. MOTIWALA

5.2 Voltage Swells

The second simulation shows the DVR performance during a voltage swell condition. The
simulation started with the supply voltage swell is generated as shown in Figure 7 (a). As
observed from this figure the amplitude of supply voltage is increased about 25% from its
nominal voltage . Figures 7 (b) and (c) show the injected and the load voltage respectively. As
can be seen from the results, the load voltage is kept at the nominal value with the help of the
DVR. Similar to the case of voltage sag, the DVR reacts quickly to inject the appropriate
voltage component (negative voltage magnitude) to correct the supply voltage.

(a) Supply Voltage

(b) Injected Voltage

(c) Load Voltage

Fig 7: Balance voltage swell
VOLTAGE SAG/SWELL COMPENSATION BY DYNAMIC VOLTAGE RESTORER 557

CONCLUSION
The modelling and simulation of a DVR using MATLAB/SIMULINK has been presented. A
control system based on dqo technique which is a scaled error of the between source side of
the DVR and its reference for sags/swell correction has been presented. The simulation shows
that the DVR performance is satisfactory in mitigating voltage sags/swells. From simulation
results also show that the DVR compensates the sags/swells quickly and provides excellent
voltage regulation. The DVR handles both balanced and unbalanced situations without any
difficulties and injects the appropriate voltage component to correct rapidly any anomaly in
the supply voltage to keep the load voltage balanced and constant at the nominal value.

REFRANCES

[1] Hingorani N.G., Flexible AC Transmission", IEEE Spectrum, vol. 30, pp. 40-44, 1993.
[2] Hingorani N.G. and Gyugyi L, Understanding FACTS – Concepts and Technology of Flexible
AC Transmission Systems, IEEE Press, New York, 2000. K.R. Padiyar “Facts controllers in
power transmission and distribution” new age international (P) Ltd publishers, 2007.
[3] Sankaran C. “Power Quality”, CRC Press 2002.
[4] Alvarez C., Alamar J., Domijan A. Jr., Montenegro A., and Song, “An investigation toward
new technologies and issues in power quality,” in Proc. 9th Int. Conf. Harmon. Qual.
[5] Torabi S.F., Nazarpour D., Shayestehfar Y. “compensation of sags and swells voltage using
dynamic voltage restorer (DVR) during single line to ground and three-phase faults ” in
International Journal “Technical and Physical Problems of Engineering” (IJTPE) , p.p 126-132,
September 2012.
[6] Golkar Masoud Aliakbar, “Power Quality in Electric Networks: Monitoring and Standards” the
second world engineering conference, pp. 137-141 July 2002.
[7] Jowder Fawzi AL, “Modeling and Simulation of Different System Topologies for Dynamic
Voltage Restorer” Electric Power and Energy Conversion Systems, EPECS '09. International
Conference, IEEE, pp. 1-6, 2009.
[8] Hingorani N.G., “Introducing Custom Power", IEEE Spectrum, vol. 32, pp. 41-48, 1995.
[9] Barakati S. Masoud, Sadigh Arash Khoshkbar, Mokhtarpour Ehsan “Voltage Sag and Swell
Compensation with DVR based on Asymmetrical Cascade Multicell
Converter”(NAPS),pp.1-7,2011.
[10] Anil Kumar R., Siva Kumar G., Kalyan Kumar B. “Compensation of Voltage Sags and
Harmonics with Phase-Jumps through DVR” Member, IEEE, pp. 1361-1366, 2009.
[11] Jowder Fawzi AL “Modeling and Simulation of Different System Topologies for Dynamic
Voltage Restorer” Electric Power and Energy Conversion Systems, EPECS '09. International
Conference, IEEE, pp. 1-6, 2009.
[12] Nielsen John Godsk and Blaabjerg Frede “Control Strategies for Dynamic Voltage Restorer
Compensating Voltage Sags with Phase Jump” Applied Power Electronics Conference and
Exposition, IEEE, vol. 2, pp. 1267-1273, 2001.
558 MISTRY RAM HEMANTKUMAR1 AND HEMIN D. MOTIWALA

[13] Katole D.N. Research scholar: Department of Electrical Engg. [Link] College of
engineering Nagpur, Maharashtra, India. “Analysis and Mitigation of Balanced Voltage Sag with
the Help of Energy Storage System” ICETET pp. 317-321, 2010.
[14] Tiwari H.P. and Gupta Sunil Kumar “Dynamic Voltage Restorer against Voltage Sag”
International Journal of Innovation, Management and Technology vol. 1, no. 3, pp. 232-237,
2010.
[15] Boonchiaml Paisan, Mithulananthan Nadarajah, Rajamangala University of Technology
Thanyaburi Thailand, “Detailed Analysis of Load Voltage Compensation for Dynamic Voltage
Restorers” TENCON, IEEE region 10 conference, pp. 1-4, 2006.
[16] Ferdi B., Dib S., [Link] University of Bechar, Algeria. “Adaptive PI Control of Dynamic
Voltage Restorer Using Fuzzy Logic” Journal of Electrical Engineering: Theory and Application
Vol.1, pp. 165-173, 2010.
[17] [Link]-D-Vimal Raj, [Link], [Link] Kumar, Sapta Rishi Roy, [Link]“multi
level inverter based dynamic voltage restorer with pi and fuzzy logic controller” NSC, pp. 65-70,
December 2008.
[18] Wang Bingsen, Giri Venkataramanan and Illindala Mahesh, “Operation and Control of a Dynamic
Voltage Restorer Using Transformer Coupled H-Bridge Converters”, IEEE transactions on power
electronics, vol. 21, pp. 1053-1061, July 2006.
International [Link] [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 559-567

WAVELET TRANSFORM FOR IDENTIFYING

POWER QUALITY PROBLEMS

SWETA BELLIWALI1, VIJAYALAXMI DHARWAD2

AND A. B. RAJU3
1
SDMCET, 2SDMCET , 3BVBCET, Dharwad, India

Abstract
With the modernization of industries and commercial centers the use of nonlinear loads, typically of
electronic nature has increased, which usually injects harmonics and inter-harmonics into the power
system and thus creates Power Quality problems. Any deviation in voltage, current, phase or
frequency that results in failure of customer equipment lead to power quality problems. Generally
faced power quality problems are transients, long duration voltage variations like overvoltage,
undervoltage, sustained interruptions, short duration voltage variations like interruption, sags, swells
etc. To mitigate these problems it is required to identify and characterize these problems. Some of the
common techniques used for identification of PQ problems are RMS technique, DFT, FFT, STFT.
The main drawback of these techniques is that these transforms can not give time and frequency
domain information at a time. In this paper wavelet transform is used to identify some of the power
quality problems. Wavelet transform is based on decomposing the original signal into two subsignals,
which further can be decomposed. These subsignals are called as coefficients which localize the
disturbance in time domain. Here the power quality disturbances are simulated in Scilab and the
coefficients of these signals are obtained using the codes built in Scilab.

© http: //[Link]
560 SWETA BELLIWALI, VIJAYALAXMI DHARWAD AND A. B. RAJU

1 INTRODUCTION
Power quality is an issue that has attracted wide interest in recent years for power utilities,
electric energy consumers and also for the manufactures of electric and electronic
equipment. The end users demand for clean electricity is constantly rising due to the
widespread use of sensitive electrical devices. Power quality disturbance can be defined as
any deviation in voltage, current, phase or frequency that results in failure or misoperation of
customer equipment.
While power disturbances occur on all electrical systems, the sensitivity of today’s
electronics makes them more susceptible to them. For some sensitive devices, a momentary
disturbance can cause scrambled data, interrupted communications, a frozen mouse, system
crashes and equipment failure. A power voltage spike can damage valuable components.
Thus it is necessary to mitigate such problems. The first step in power quality evaluation is
the identification of power quality problems. Some of the techniques used are RMS
technique, DFT, FFT, STFT[1]-[3]. RMS technique does not distinguish fundamental
frequency, harmonics or noise components. FFT performs well for estimation of periodic
signals in stationary state; however it doesn’t perform well for detection of suddenly or fast
changes in waveform e.g. transients or voltages dips. FFT also suffers from spectral leakage
effects and picketfence effect.
Wavelet transform has good time-frequency localization properties, which has become a
research hotspot in recent years. The detection method based on wavelet transform is
transforming the signal into different time-frequency scales, and extracting signal feature
from the wavelet coefficients.

2 POWER QUALITY
Power quality (PQ) is a compatibility issue between the supply systems and loads. As long
as both can coexist without causing any ill effects on each other, the quality can be regarded
as good or adequate. When power is generated, it has very predictable characteristics. It
energizes all electrical equipment equally and satisfactorily. However, as the power travels
through the wires and energizes the equipment, the various pieces of equipment it energizes
can change the quality of the power, making it less suitable for the next application. These
changes in power quality are especially common in large industrial and commercial
complexes and include increases and decreases in voltage, momentary power outages, and
WAVELET TRANSFORM FOR IDENTIFYING POWER QUALITY PROBLEMS 561

“noise” on the electrical system. At its most extreme, poor power quality can even cause
equipment to malfunction. The quality of electric power has become an important issue
because with the introduction and wide spread use of sensitive electronic equipment,
customers have become much more aware and sensitive to transient and other power
anomalies.
There has been some ambiguity throughout the electrical industry and businesses community
in the use of terminology to describe various power disturbances. For example, the term
“surge” is seen by one sector of the industry to mean a momentary increase in voltage as
would be typically caused by a large load being switched off. On the other hand, usage of the
term “surge” can also be seen as a transient voltage lasting from microseconds to only a few
milliseconds with very high peak values. These latter are usually associated with lightning
strikes and switching events creating sparks or arcing between contacts.
The IEEE Standard has addressed the problem of ambiguity in terminology. This IEEE
defined power quality disturbances are as follows [4].
• Transients: It can also be defined as that part of the change in a variable that
disappears during transition from one steady state operating condition to another.
• Sag: It is a decrease to between 0.1 and 0.9 pu in rms voltage or current at the power
frequency for durations from 0.5 cycle to 1min.
• Swell: It is an increase to between 1.1 and 1.8pu in rms voltage or current at the
power frequency for durations from 0.5 cycle to 1min.
• Interruption: It occurs when the supply voltage or load current decreases to less than
0.1 pu for a period of time not exceeding 1min.
• Harmonics: Sinusoidal voltages or currents having frequencies that are integer
multiples of frequency at which the supply system is designed to operate.
• Voltage Fluctuation: The systematic variations of the voltage envelope or series of
random voltage changes, the magnitude of which does not normally exceed the
voltage ranges of 0.9 to 1.1pu.
• Power Frequency Variations: The deviation of the power system fundamental
frequency from it specified nominal value (e.g., 50 or 60 Hz).
• DC offset: The presence of a dc voltage or current in an ac power system. This can
occur as the result of a geomagnetic disturbance or asymmetry of electronic power
converters.
562 SWETA BELLIWALI, VIJAYALAXMI DHARWAD AND A. B. RAJU

Figure 1: Power quality disturbances

3. WAVELET TRANSFORM
Wavelet means ‘small wave’. So wavelet analysis is about analyzing signal with short
duration finite energy functions. They transform the signal under investigation into another
representation which presents the signal in a more useful form. This transformation of signal
is called wavelet transform. Unlike Fourier transform, we have a variety of wavelets that are
used for signal analysis. Choice of a particular wavelet depends on the type of application in
hand [5]. Fig. 2 shows some of the examples of wavelets.
Mathematically, a wavelet is denoted as shown in eqn.(1):
1 t−b
ψa,b(t)= √|a| ψ( a ) (1)

where b is location parameter and a is scaling parameter. For the function to be a wavelet, it
should be time limited. For a given parameter a, we translate the wavelet by varying the
parameter b.
Wavelet transform is defined as shown in eqn.(2):
1 t−b
W(a,b)= 
 f(t) ψ( ) dt (2)
√|a| a
According to eqn.(2) for every (a,b), we have a wavelet transform coefficient
representing how much the scaled wavelet is similar to the function at location t=(b/a)
WAVELET TRANSFORM FOR IDENTIFYING POWER QUALITY PROBLEMS 563

The wavelet used for the study is Daubechies4 wavelet. It is shown in fig. 2. If a signal f has
an even number N of values, then the 1-level Daub4 transform is the mapping f→(a1|d1)

from the signal f to its first trend subsignal a1 and first fluctuation subsignal d1. The 2-level

Daub4 transform is the mapping a1→(a2|d2) from the first trend subsignal a1 to a second

trend subsignal a2 and second fluctuation subsignal d2. Similarly 3-level Daub4 transform is

the mapping a2→(a3|d3) from the second trend subsignal a2 to a third trend subsignal a3

and third fluctuation subsignal d3. This process is shown in fig. 3

Figure 2: Examples of wavelets

Figure 3: Wavelet tree

564 SWETA BELLIWALI, VIJAYALAXMI DHARWAD AND A. B. RAJU

4 RESULTS
The codes to analyse the power quality disturbances using wavelets was developed in Scilab.
The results were compared with that obtained using ‘dwt’ an inbuilt command for wavelet
analysis. Figs. 4 to 11 shows the wavelet analysis for various disturbances done using inbuilt
command and developed code in Scilab. The figures on the left are obtained using inbuilt
command ‘dwt’. The figures on the right are obtained using the developed code in Scilab.
On comparing the results it was found that the coefficients obtained in both the methods are
in close agreement.

Figure 4: Wavelet analysis for transient using Figure 5: Wavelet analysis for sag using
inbuilt command and developed code inbuilt command and developed code
WAVELET TRANSFORM FOR IDENTIFYING POWER QUALITY PROBLEMS 565

Figure 6: Wavelet analysis for swell using Figure 7: Wavelet analysis for interruption
inbuilt command and developed code using inbuilt command and developed code
566 SWETA BELLIWALI, VIJAYALAXMI DHARWAD AND A. B. RAJU

Figure 8: Wavelet analysis for harmonics Figure 9: Wavelet analysis for notching using
using inbuilt command and developed code inbuilt command and developed code

Figure 10: Wavelet analysis for power Figure 11: Wavelet analysis for dc offset using
frequency variation using inbuilt command and inbuilt command and developed code
developed code
WAVELET TRANSFORM FOR IDENTIFYING POWER QUALITY PROBLEMS 567

CONCLUSION
Power quality problems are increasing with the use of power electronics devices, which
cause malfunctioning of sensitive devices. As a first step of power quality evaluation
identification of problem plays an important role. Several methods are used for the
identification of the disturbances such as FFT, STFT. Fourier transforms suffers from
Leakage and picketfence effects and thus makes identification and localization of the
disturbance difficult. So, an alternative mathematical tool- wavelet transform which gives
localisation in time domain is best suited for identifying the transient disturbances.
The power quality disturbances are generated through Scilab codes and these signals are
decomposed using wavelet transforms to identify the disturbances.

REFERENCES

[1] Valenzuela Javier and Pontt Jorge “Real-Time Interharmonics Detection and Measurement
Based on FFT Algorithm”.

[2] Chang G. W., Chen C. I and Chin Y. C. “Modified High-Resolution Singular Value
Decomposition Method for Power Signal Analysis by Using Down-Sampling Technique”.

[3] Heydt G. T., Fjeld P. S., Liu C. C., Pierce D., Tu L. and Hensley G., “Applications of the
Windowed FFT to Electric Power Quality Assessment”, IEEE Transactions on Power Delivery,
Vol. 14, No. 4, Octobcr 1999.

[4] Durgan Roger C., McGranaghan Mark F., Santoso Surya and Beaty H. Wayne, Electrical Power
Systems Quality, “Tata McGraw-Hill Edition 2008.”

[5] Soman K. P., Ramachandran K. I. and Resmi N. G. Insight Into Wavelets, “Tata McGraw-Hill
Edition 2008.”

[6] Santoso Surya and Hofmann Peter, “Power Quality Assessment Via Wavelet Transform
Analysis”, IEEE Transactions on Power Delivery, Vol. 11, No. 2, April 1996.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 569-578

IMPLEMENTATION OF ZVZCS CONVERTER FOR

VOLTAGE SMOOTHING IN A PMDC DRIVE

SARATH. S1 AND T. GOVINDARAJ2

1
[Link] Scholar, Department of Electrical and Electronics Engineering
Muthayammal Engineering College Rasipuram, Tamil Nadu, India
2
Professor, Department of Electrical and Electronics Engineering
Muthayammal Engineering College, Rasipuram, Tamil Nadu, India

Abstract
This project proposes a novel method for controlling a PMDC motor using a resonant zvzcs converter.
The converter provides a full bridge output voltage without any spikes. The converter employs an
auxiliary circuit to provide the reactive current for the full-bridge semiconductor switches, which
guarantees zero voltage switching at turn-ON times for all load conditions.

----------------------------------
Keywords: ZVZCS converter, PMDC motor, PI controller.
© http: //[Link]
570 SARATH. S AND T. GOVINDARAJ

1. INTRODUCTION
The power conversion systems with wide load variations and hybrid electric vehicles are the
areas where large fluctuations in the load occur. Efficient transfer of energy with the least of
losses are highly desired. For such systems, this project proposes a converter topology using
ZVZCS switching. AC/DC converters, which are used for heavy load applications, usually
consist of two stages: input power factor correction (PFC) for AC/DC conversion and
DC/DC conversion for battery charging .The conventional full-bridge converter with the
series inductor (zvs switching) loses its ZVS capability at light loads, and the converter with
the parallel inductor loses its ZVS under heavy loads. Loss of ZVS implies extremely high
switching losses at high switching frequencies and very high EMI. Hence, the proposed
converter has the advantage that can overcome all these shortcomings and has proved better
efficiency over varying load conditions. The applications of this converter also extends to
switching power supplies, battery chargers, uninterruptible power systems, renewable energy
generation systems, and telecom power supplies. Full-bridge topology is the most popular
topology used in the power range of a few kilowatts for DC/DC converters. Since the switch
ratings are optimized for the full-bridge topology, this topology is extensively used in
industrial applications. High efficiency, high power density, and high reliability are the
prominent features of this topology.
For applications in the range of a few kilowatts, MOSFETs are mostly used to implement the
full-bridge converters. In order to have robust and reliable operation, MOSFETs should be
switched under zero voltage. Operation with zero voltage switching (ZVS) has numerous
advantages including, for example, reduction of the converter switching losses and a noise-
free environment for the control circuit. Zero voltage switching is usually achieved by
providing an inductive current flowing out of the full-bridge legs during the switch turn-ON
and by placing a snubber capacitor across each switch during the switch turn-OFF. The
inductive current can be produced by inserting an inductor in series with the power
transformer or by inserting an inductor in parallel with the power transformer.
In a practical full-bridge configuration, the internal drain-to source capacitor of the
MOSFET is usually utilized as the snubber capacitor, the series inductor is usually the
leakage inductance of the power transformer, and the parallel inductor is implemented by
using the magnetizing inductance of the power transformer. Thus, external passive
components are not required, which makes the power circuit very simple and efficient.
IMPLEMENTATION OF ZVZCS CONVERTER FOR… 571

However, the full-bridge converter with the series inductor loses its ZVS capability at light
loads, and the converter with the parallel inductor loses its ZVS under heavy loads. Loss of
ZVS implies extremely high switching losses at high switching frequencies and very high
EMI due to the high di/dt of the snubber discharge current. Loss of ZVS can also cause a
very noisy control circuit, which leads to shoot-through and loss of the semiconductor
switches. The ZVS range can be extended by increasing the series inductance. However,
having a large series inductance limits the power transfer capability of the converter and
reduces the effective duty ratio of the converter. In battery charger applications, ZVS is
vitally important since the converter might be operating at absolutely no-load for a long
period of time. In this application, when the battery is charged, the load is absolutely zero
and the converter should be able to safely operate under the zero load condition. Since ZVS
in conventional full-bridge PWM converters is achieved by utilizing the energy stored in the
leakage inductance to discharge the output capacitance of the MOSFETs, the range of the
ZVS operation is highly dependent on the load and the transformer leakage inductance.
Thus, this converter is not able to ensure ZVS operation for a wide range of load variations.
A novel approach has been adopted to extend the ZVS range in the full-bridge converter. In
this approach, an auxiliary inductor is put at the leading leg by deriving an auxiliary winding
on the main transformer and confirms ZVS for the leading leg. Although the proposed
scheme can effectively extend ZVS of the leading leg MOSFETs, it is not able to guarantee
ZVS for the lagging leg of the converter. Thus, when the battery is fully charged, the lagging
leg switches may not be switched under ZVS. Moreover, the voltage at the secondary side
still suffers from the spikes due to the leakage inductance and voltage-fed rectifier,
commonly seen in full-bridge converters.
A fundamental problem related to the conventional full bridge phase-shift DC/DC converter
is the voltage spikes across the output diodes. It shows the schematic of the conventional
full-bridge converter. Basically, the leakage inductance of the transformer causes the voltage
spikes across the output diodes. Theses spikes are intensified by increasing the switching
frequency of the converter. Thus, the diodes should be designed overrated to be able to
withstand the voltage spikes, which leads to higher losses due to the higher forward voltage
drop of the diodes and poorer reverse recovery characteristics. In addition, the spikes
significantly increase the EMI noise of the converter. This fact makes the topology not very
practical for high frequency, high voltage applications. There are quite a few references that
572 SARATH. S AND T. GOVINDARAJ

proposed solutions for the voltage spikes across the output diodes. Some references tried to
decrease the leakage inductance as much as possible though the transformer winding
structures, which effectively decreases the peak of the voltage spikes across the output
diodes. However, reducing the leakage inductance decreases the ZVS operating range of the
full-bridge converter, which results in a very narrow range of ZVS operation. An R-C-D
snubber circuit is used to mitigate the voltage spikes across the diodes. The main problem
with the snubber circuit is the amount of losses in the snubber resistor, which considerably
degrades the efficiency of the converter especially at higher power and it can only reduce the
peak value of the voltage spikes. An active clamp circuit has been added to the converter to
clamp the voltage across the output diodes. This method can effectively clamp the voltage
spikes of the output diodes. However, the active clamp circuit increases the complexity of
the converter and causes small losses in the clamp circuit. Several energy recovery clamp
circuits (ERCCs) have been proposed. An improved ERCC method has been proposed to
accommodate the effects of voltage spikes for a wide range of input voltage. Although the
ERCC techniques are able to reduce the voltage stress of the output diodes, the amount of
the voltage stress depends on the duty ratio and input voltage of the converter in most of the
ERCCs techniques. In addition, using extra semiconductors is inevitable in all these
aforementioned methods. The problem of voltage spikes is essentially related to the voltage-
driven output rectifiers. This is due to the fact that the full-bridge inverter produces high
frequency voltage pulses across the output diode rectifier, which is connected to the output
inductor as shown. The voltage-driven rectifier works perfectly if there is no leakage
inductance in between the output of the full-bridge inverter and the diode rectifier. However,
the existence of the leakage inductance makes the rectifier connect two current sources, i.e.,
leakage inductance and output inductance, together. This connection creates high voltage
spikes across the output diodes. In this project, a new topology is proposed based on a
current driven rectifier, which effectively rectifies the voltage stress problems related to the
full-bridge DC/DC converter

2. BLOCK DIAGRAM DESCRIPTION

The block diagram consist of a DC voltage source, filter circuit, a resonant converter acting
as an inverter, transformer stage, diode rectifier, PMDC motor and controller.
IMPLEMENTATION OF ZVZCS CONVERTER FOR… 573

Fig. 1. Block diagram

Input source: The input voltage is provided as a regulated dc voltage from a steady source.
A rectified dc voltage from an ac source can also be considered accordingly
Filter circuit: As per part of reducing the voltage stress across the switches one method is to
reduce voltage across each set of switch of the zvzcs converter. So we provide a capacitive
filter circuit after the voltage source.
The input supply will be divided and fed to the zvzcs converter. So the effective voltage
experienced by the switch will be half than one without divider circuit.
Full bridge converter circuit: The converter is designed to satisfy the switching of a zvzcs
topology. The switches are IGBT’s and hence losses are reduced drastically in addition to
the advantage of the zvzcs. The switching takes place at zero voltage crossing and zero
current crossing.
Transformer stage: To transmit power from primary side to secondary side and to step up
the voltage we use transformer. The transformer loss incurred will be very less and it is
evident from very high efficiency of transformers.
Diode rectifier: The diode rectifier rectifies the ac voltage received at the secondary of the
transformer, producing a regulated dc voltage at the output with the help of a capacitive
filter.
Controller: The controller helps to turn the switches on and off at instants of zero voltage.
The controller is a PI (proportional integral) controller. The output of the controller will be
given to the switches of the zvzcs converter. The controller helps to turn the switches on and
off at instants of zero voltage. The controller is a PI (proportional integral) controller. The
output of the controller will be given to the switches of the zvzcs converter.
574 SARATH. S AND T. GOVINDARAJ

PMDC motor: Permanent Magnet DC motors are useful in a range of applications, from
battery powered de- vices like wheelchairs, welding equipments etc. They are frequently the
best solution to motion control and power transmission applications where compact size,
wide operating speed range, ability to adapt to a range of power sources are major
considerations .Because of their linear speed-torque curve, they particularly suit adjustable
speed and servo control applications

3. PI CONTROLLER

Fig 2. Block of PI controller.

The block diagram for the PI controller is shown above. In the new topology , an auxiliary
inductor is placed on the main transformer , and hence confirms zvs operation for leading leg
thyristors of the converter .It also provides zcs switching for the output rectifiers which
eliminates the reverse recovery losses of the output diodes . The direct supply from the
rectifier is given to the motor load.

4. CIRCUIT DESCRIPTION
Global energy crisis and increasing fuel costs demands the power electronic energy
conversion equipments, which offering good conversion efficiency, high performance.
Recently soft switching converters are providing good DC-DC energy conversion with
reduced switching loses and EMI
IMPLEMENTATION OF ZVZCS CONVERTER FOR… 575

Fig 3. Circuit diagram of the proposed converter

Here Fig 2 shows the proposed converter circuit. Auxiliary inductors are designed based on
the amount of reactive power required to guarantee ZVS for the MOSFETs. In other words,
the reactive current should be enough to completely charge and discharge the MOSFET
output capacitors. However, since there are substantial differences in operating modes for the
leading leg, the design procedure is given accordingly. According to the operating modes of
the proposed converter, the worst case for ZVS operation is no-load. Thus, ZVS should be
guaranteed for no-load in order to make sure ZVS operation for all operating conditions. At
absolutely no-load, the primary current is zero during the switching transitions of the leading
leg. Therefore, the current through auxiliary circuit of the leading leg should only charge and
discharge the output capacitors of the MOSFETs. The series inductor should be designed so
that the converter full-load condition corresponds to the critical conduction mode of the
series inductor. This series inductance Lseq plays a major role in the energy transfer from the
primary to the secondary side of the transformer. This inductance by nature is an AC
inductor. So regular design methods of an AC inductor with air-gap has to be followed if an
external inductor is to be designed for the series inductor, Ls. This inductance can be
integrated as the leakage inductor of the transformer too. Although in that case, a transformer
with a precisely designed leakage inductance will be necessary but the advantage of having
this integrated series inductance is immense including elimination of an actual physical
576 SARATH. S AND T. GOVINDARAJ

inductor along with its core and copper losses and the adverse effects of its fringing flux on
the EMI and the operation of the surrounding devices on the PCB of the converter

Fig 4 Main Simulation Circuit.

5. EXPERIMENTAL RESULTS
The proposed circuit is developed and analyzed with MATLAB software.
MATLAB simulation circuit diagram for proposed full bridge ZVZCS converter is shown in
fig 11
Resultant output waveforms after simulation is as given below
5.1Resultant Waveforms

Fig 5 .Rectifier Output voltage

IMPLEMENTATION OF ZVZCS CONVERTER FOR… 577

Fig 6 .Converter Output voltage

Fig. 14 Motor torque

Fig. 15 Motor speed

CONCLUSION
In this project, a novel yet simple full-bridge topology has been implemented on a PMDC
motor. The proposed converter eliminates the adverse effects of the voltage spikes at the
secondary side of the transformer, as well as the freewheeling mode of operation, which are
578 SARATH. S AND T. GOVINDARAJ

intrinsic to the conventional full-bridge converters. Moreover, the proposed converter

assures reliable operation at no load by applying the symmetric auxiliary circuits on both
legs of the full-bridge converter. Better efficiency of the proposed converter over full range
of operation not only validate the operation of the converter but also confirm the superiority
of the proposed topology over the conventional full-bridge converter.

REFERENCES

[1] Chu Enhui, Hou Xutong, Zhang Huaguang, Senior Member, IEEE, Mengyang Wu, and
Xiuchong Liu “Novel Zero-Voltage and Zero-Current Switching (ZVZCS) PWM Three-Level
DC/DC Converter Using Output Coupled Inductor”, IEEE Transactions on Power Electronics,
vol. 29, no. 3, March 2014
[2] Cha H., Chen L., Ding R., Tang Q., and Peng F. Z., “An alternative energy recovery clamp
circuit for full-bridge PWM converters with wide ranges of input voltage,” IEEE Trans. Power
Electron., vol. 23, no. 6, pp. 2828– 2837, Nov. 2008.
[3] Raggle K., Nussbaumer T., and Kolar J. W., “Guideline for a simplified differential-mode EMI
filter design,” IEEE Trans. Ind. Electron., vol. 57, no. 3, pp. 1031–1040, Mar. 2010.
[4] Kim Hyungjoon, Yoon Changwoo, and Choi Sewan, IEEE Senior Member Seoul National
University of Technology Dept. of Control and Instrumentation Eng. “A three-Phase ZVZCS
DC-DC converter for fuel cell applications”
[5] Cho J. G., Baek J.W., Yoo D.W., Lee H. S., and Rim G. H., “Novel zerovoltage and zero-
current-switching (ZVZCS) full bridge PWM converter using transformer auxiliary winding,” in
Proc. 28th Annu. IEEE Power Electron. Spec. Conf., (PESC), vol. 1, St. Louis, MO, Jun. 22–27,
1997, pp. 227–232
[6] Mweene L. H., Wright C. A., and Schlecht M. F., “A 1 kW 500 kHz front-end converter for a
distributed power supply system,” IEEE Trans. Power Electron., vol. 6, no. 3, pp. 398–407, Jul.
1991.
[7] Ayyanar R. and Mohan N., “Novel soft-switching DC–DC converter with full ZVS-range and
reduced filter requirement. I. Regulated-output applications, IEEE Trans. Power Electron., vol.
16, no. 2, pp. 184–192, Mar.2001.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 579-597

CONTROLLING OF SWITCHED RELUCTANCE MOTOR

DRIVE SYSTEM USING DSP TMS320LF2407

SOUMYA I CHANDAKAWATHE AND POOJA CHINNIVAR

Department of Electrical and Electronics Engineering.
The Oxford College of Engineering Bangalore India

Abstract
The switched reluctance motors (SRM) drive is increasing attention from various researchers as well
as industry for adjustable speed drives applications. Combining the unique features of an SRM with
simple and efficient power converter that uses a superior motor drive system emerges which may be
preferable for many applications compared to other ac or dc motor drive systems. The attractions of
SRM drives compared to conventional drives will be significantly enhanced if the rotor position
sensor can be eliminated. Researchers prefer switched reluctance motor drive system due to their
insensitivity of high temperature, easy to cool and also the development of high performance control
strategies. This paper presents the principles of switched reluctance motor drive system andits
controlling using DSP TMS320LF2407.

-----------------------------------
Keywords : SRM;DSP TMS320LF2407; PI controller; Power Inverter; PWM control;Hysteresis
Control
© http: //[Link]
580 SOUMYA I CHANDAKAWATHE AND POOJA CHINNIVAR

1. INTRODUCTION
Mainly on principle of Reluctance, Switched Reluctance Motors generate electromagnetic
torque and it is a singly excited synchronous machine. In Switched Reluctance Motor there
is only one source of torque, that is the tendency of polarized rotor pole to align with an
excited stator pole. Normally in electric machines, between the magnetic fields there is an
attractive force and repulsive force that is caused by armature and field windings and
ultimately it results as a main cause for torque production. It is possible to attain the optimal
performance by appropriately positioning of current pulse according to magnetic status of
machine, hence sensing of position of the rotor is very important in controlling of the drive.

In SRM, sensor will be the essential part of the control in drive system. Unipolar power
inverter feeds the the [Link] or PWM current controller is used to get the required
current profile. To excite SRM, square shaped current pulse is use. Other current profiles are
also used to minimize effects of high torque and audible sound. Nowadays low cost , better
performance DSP controllers are put in use to establish complex control algorithm.
Advantages [1] of SRM over other drive systems:
 It can withstand high temperature.
 Rugged and Modular Structure, hence easy to handle.
 Simple geometryand hence easy maintenance.

2. BASIC OPERATION OF SRM DRIVE

Electromechanical energy conversion process is illustrated here to elaborate the process of
torque generation. To generate reluctance torque, it is required to excite stator phase at
unaligned position shown by ‘a’ in [Link] stator and rotor pole pair undergo maximum
CONTROLLING OF SWITCHED RELUCTANCE MOTOR… 581

length of [Link] fig ‘u’ shows the approaching of rotor to aligned position and there will
be maximum attraction force of the radial component and zero tangential component. So it is
necessary to make current to zero before aligned position. Magnetic energy converted to
mechanical energy is shown by shaded area in fig.2. The area shown as ‘R’ indicates
magnetic energy which is not converted.
mechanical work
Power quality in SRM drives =
magnetic form of energy

Here ψ is flux linkage and θ is rotor position.

.

Excitation of stator phase is done during movement of the rotor to the aligned position from
unaligned for motoring action. To attain generating action, stator phase is excited during the
motion of rotor to unaligned position from aligned position. Stator phases are excited in
sequence to attain continuous rotation. In 8/6 SRM drive, field distribution at the time of
commutation is illustrated in Fig.3.
Rotation direction is reverse as that excitation of stator. In every electrical cycle, there is
path of short flux in motor back iron. This leads to asymmetry in process of torque
production.
582 SOUMYA I CHANDAKAWATHE AND POOJA CHINNIVAR

Excitation of stator exactly synchronized with position of rotor is an important stage to attain
optimal control strategy in SRM [Link] is having magnetic characteristics like phase
inductance or phase flux linkage which depict resemblance with the position of the rotor[2].
For this reason, they can be utilized for the control purpose.
SRM drives uses power electronic inverter which is an asymmetric bridge and shown in
[Link] of both the switches makes the winding to get the dc link voltage. When the
switches are open winding gets negative Dc link voltage and continuous winding current is
guaranteed by freewheeling diodes. Surely, closing of one switch and opening of other, the
path for current [7] which is short circuited is given by the corresponding freewheeling
diode.
The control system expects current regulation based on PWM or hysteresis controller which
can be implemented using present topology. When the speed is high, EMF induced in the
winding is powerful and it won’t provide way for current waveform control. So it is clear
that current regulation problem belongs to low speed operation.
CONTROLLING OF SWITCHED RELUCTANCE MOTOR… 583

3. CONTROL IN SRM DRIVES

Different types of adjustable speed motor drives are mainly distinguished by the factor-
electromagnetic torque control. Fundamentals of commutation in SRM drives is illustrated in
[Link] is possible to get positive motoring or negative generating modes of operation by the
appropriate position of the current pulse.
The induced electromotive force and electromagnetic torque generated by SRM drive
expressed in terms co-energy as -

where Wc, L,θ,i , and ω represents co-energy, phase inductance, rotor position, phase
current, and angular speed, respectively. Here non linear effects of magnetic saturation are
not considered. It is possible to attain positive torque only placing of current pulse in an area
which is having increasing inductance profile. In the same way generating mode of operation
is accomplished when there is positioning of excitation in a region which is having
decreasing inductance profile. It is necessary to tune commutation instants (θ on, θ off) as a
function of angular speed and phase current so as to increase productivity of this drive.
Optimization of torque per ampere is necessary to attain it.
584 SOUMYA I CHANDAKAWATHE AND POOJA CHINNIVAR

4. METHOLOGY OF OPEN LOOP CONTROLFOR TORQUE

It is possible to design open loop control strategy for SRM drive by selecting commutation
instants, control variable and reference current appropriately. The open loop control strategy
consists of following stages:
 Initial rotor position is detected.
 Commutation thresholds accordance with the polarity of current level, torque and speed
are calculated.
 Rotor position is monitored [3] and active phases are selected.
 A control strategy for regulation of phase current at low speed.
 Each stage is discussed in detail in following sections:
4.1 Detection of the Initial Rotor Position
The most important work when the rotor is stationary is to find out the appropriate phase for
excitation at the beginning. Ones this is done as per direction of rotation, excitation of stator
phase sequentially is put in place. To start the motor, the suitable way to detect the rotor
position is to align one of stator phases with the rotor [3]. This is attained by stator phase
CONTROLLING OF SWITCHED RELUCTANCE MOTOR… 585

excitation with sufficient current for a short time interval. It is possible to set up a reference
initial position when the rotor is in an aligned position. This strategy needs rotor to move at
beginning, which is unacceptable in some cases. Here sensor less scheme at standstill is
incorporated. At standstill rotor position detection is discussed here. All phases are fed with
series of voltage pulses with fixed and sufficient short duration. Selection of the proper
phase for conduction is done by comparing the peak current magnitude. Fig [Link] 12/8
SRM drive with set of normalized inductance profile.

As per magnitude of inductance six different regions are obtained by dividing a full electrical
period. The following equation is true for measured current magnitude as the induced
voltage at small amplitude currents are not there

where, VBus, and LABC represent duration of pulses, dc link voltage, and phase
Inductances, respectively. Detection process for 12/8 SRM drive is summarized in table I.
For starting, the appropriate phase can be found out when the range of position is detected.
586 SOUMYA I CHANDAKAWATHE AND POOJA CHINNIVAR

Detection process at standstill is summarized in flowchart of fig.7.

4.2 Commutation Thresholds Calculation

Next step includes calculation of each phase commutation angle and storing it in memory.
Calculation of threshold is very if commutation angles are fixed. It is important here that
excitation of every phase is only ones within each electrical cycle. Phase shift in symmetric
machine is given by
CONTROLLING OF SWITCHED RELUCTANCE MOTOR… 587

Where Ns and Nr number of rotor and stator poles.

Suppose N pulses per mechanical revolution is generated by a particular encoder, then every
mechanical degree corresponds to 4N/360 pulses received by the QEP(Quadrature Encoder
Pulse) counter of [Link] quadrature input of LF2407 is designed for commercial
available encoders with quadrature output. By this component, resolution of the sensed
position is increased by a factor of 4.
SRM drive with a typical current pulse is shown in fig [Link] neglecting the effects of
rotational back EMF in the neighborhood of commutation which is a valid assumption as
turn on and turn off instants occur close to unaligned and aligned position respectively. The
delay angles are computed as

where ,La ,ω ,V , and r are unaligned inductance, aligned inductance, angular speed, bus
voltage, and stator phase resistance, respectively.
588 SOUMYA I CHANDAKAWATHE AND POOJA CHINNIVAR

4.3 Monitoring of Rotor Position and Selection of Active Phases

After the completing the steps, conduction band is enforced and current is regulated. Fig 9
shows block diagram which illustrates structure used in particular algorithm which is the
main control strategy for SRM drive[4]. Using LF2407, rotor position is easily controlled.
When second timer in event manager is correctly initialized it is possible to program it to act
as counter for QEP encoder input. At any
Step of the program, the content of this counter can be accessed. Hence rotor position can be
easily monitored. Current value of the rotor position[8] is easily compared with commutation
thresholds and identification of ON phases is done .Then current in active phases , where
active is refer to as a phase that is turned ON will be regulated.

4.4 A Control Strategy for Regulation of Phase Current At Low Speeds

When the induced EMF is small at low speeds, phase current control method is essential.
When such routines are not there, there is increase of phase current exponentially which can
cause damage to semiconductor or motor winding. At low speeds hysteresis and PWM
CONTROLLING OF SWITCHED RELUCTANCE MOTOR… 589

control methods are used for phase current regulation where as at high speed phase current
does not increase because of large back EMF and hence regulations schemes are not
required. Fig10 depicts recording of regulated current waveform along with gate pulse at low
speed region.

It is required to sense active phase currents for hysteresis control. On-chip analog to digital
converter of LF2407 is used to convert sampled current into digital form. Fig 4 shows a two
switch per phase inverter for which control rules are given by:
 When Imin≥I, then both switches are ON. Hence bus voltage is applied across the
terminals of coil.
 When Imax≤I, both switches are turned OFF. Hencenegative bus voltages is applied
across coil terminals.
 When Imin≤I≤Imax, it is not necessary to do any changes in switch position.
Interrupt service time should be very small since only two phase conduct at the same time
and given the speed of 30MIPS (Millions of Instructions per second) in [Link]
current is controlled in SRM drive by PWM technique. Phase current controlled[5] by
typical PWM regulator in SRM drive is shown in Fig [Link] of variable time constant
of stator winding a fixed set of gains in the controller is not sufficient and gain scheduling
technique need to be included. Gains corresponding to aligned and unaligned rotor time
590 SOUMYA I CHANDAKAWATHE AND POOJA CHINNIVAR

constants are found out and method of linear interpolation is used to compute gains at
intermediate positions.

Fig 12 depicts the block diagram where there are different steps with LF2407 peripherals.
Commutation angle and current profiles are inputs to program. Rotor position and angular
speed of the drive are found out by using quadrature output of an encoder. For control of
current, it is required to sample phase current and convert it into digital form. The output
gates are selected from
General Purpose Input Output (GPIO) pins. The interface, conditioning circuit and buffers
are not shown in this Fig. To get the final gating signal, the control routine used for detection
of active phases and hysteresis/PWM control of phase current is combined in software.
CONTROLLING OF SWITCHED RELUCTANCE MOTOR… 591

5. CLOSED LOOP CONTROL OF SRM DRIVE

Fig 13. Shows SRM drive with typical cascaded control configuration. The power switches
get the gate signals from main control block. Functions like current regulations and phase
commutations are also carried out by it. Reference current, commutation instant and a
sequence of excitation are necessary to carry out these tasks. Adjustable speed motor drive
can operate in different quadrants of torque speed plane as per its application

Fig 14 depicts minimum needs an adjustable speed motor drive to carry out torque, speed,
and position control task. Positive (motoring) or negative (generating) torque commands are
given by speed controller for speed regulation. Similarly positive (clockwise) and negative
(counterclockwise) speed commands are given by position controller [2]. It is necessary to
change the rotation direction in air gap field, so as to achieve four quadrant operation in
SRM drives.
Closed loop torque control system is shown in the block diagram of [Link] modules
contained in this fig are:
 Average/instantaneous electromagnetic torque estimator.
 A feed forward function for fast and convergent tracking of the commanded torque.
 A computational block to find out commutation instants as per sign of demanded torque
and magnitude of phase current.
592 SOUMYA I CHANDAKAWATHE AND POOJA CHINNIVAR
CONTROLLING OF SWITCHED RELUCTANCE MOTOR… 593

Depending on equation 1, estimator for average /instantaneous electromagnetic torque is

developed. Phase inductance/flux linkage [6] analytical model is also incorporated in the
design as depicted below:

whereL0, L1, and L2 denote polynomials.

At steady state, in a 12/8 SRM drive, Fig 16 shows comparison between estimated and
measured torque during response periodic ramp function in closed loop control. Good
accuracy is shown by average torque estimator. There is an error of 0.4 Nm because iron and
stray losses are not considered in torque estimator. To carry out this test permanent magnet
drives behaving as active load is set in a speed control loop running in same direction at 800
rpm

.
594 SOUMYA I CHANDAKAWATHE AND POOJA CHINNIVAR

[Link] LOOP SPEED CONTROL OF SRM DRIVE

The speed control is explained since the next stage is designing SRM drive with high
performance. It is possible to carry out closed loop speed control by a cascaded type of
control as illustrated in [Link] of the speed is done by position information which is
before given by encoder. Since SRM is synchronous machine, electrical frequency of
excitation is selected for control purpose. The relationship between mechanical and electrical
speed is given by

Where Nris number of rotor poles. Finally tightly regulated speed control[4] is performed
successfully which is based on inner torque control system performance as shown in fig.17.
CONTROLLING OF SWITCHED RELUCTANCE MOTOR… 595

[Link] FOR RUNNING THE SRM DRIVE USING AN OPTICAL ENCODER

To run a 4-phase SRM drive using LF2407 DSP controller, an algorithm is depicted here.
Commutation angles are adjusted by varying t-on and t-off. It is required to tune HLIMIT
and LLIMIT to adjust current magnitude. In an encoder based architecture, in order to
control conduction band and phase current a three level interrupt is used. A reference for
rotor position is provided by aligning one phase. Next, commutation instants for each phase
are calculated. A current limiting algorithm is necessary as there is no control over current at
the time of aligning process. In order to understand the algorithm properly, the following
flowchart is shown.
596 SOUMYA I CHANDAKAWATHE AND POOJA CHINNIVAR

Fig.18. Overall SRM drive flowchart

CONTROLLING OF SWITCHED RELUCTANCE MOTOR… 597

CONCLUSION
Now-a-days SRM drives are giving entry into ASMD system. Many control methods are
discussed in SRM drives. It includes the fundamental concept of designing a closed loop
control strategies. SRM having advantages of its large performance control strategies,
capacity, easy maintenance, and also it can withstand high temperature. These
methodologies are implemented successfully and it will give torque generation process.
Many researches are investigating about the advanced technologies such as Position Sensor
less, Adaptive control. By carrying out the discussed control strategies of SRM drive it is
possible to have good efficiency, ability to sustain fault and compactness. Finally in this
paper a flowchart is developed to run 4-phase SRM drive using DSP controller LF2407 .

REFERENCES

[1] Toliyat Hamid. A., Campbell Steven G,”DSP based electromechanical motion control “, Texas
A&M university Department of Electrical engineering College station Texas.
[2] Acarnley P. P., Hill R. J., and Hooper C. W., “Detection of rotor position in stepping and
switched reluctance motors by monitoring of current waveforms,” IEEE Trans. Ind. Electron.,
vol. 32, pp. 215–222, June 1985.
[3] Kjær P. C., Blaabjerg F., Pedersen J. K., Nielsen P., and Andersen L., “A new indirect rotor
position detection method for switched reluctance drives,” in Proc. ICEM’94, Paris, France,
1994, vol. 2, pp. 555–560.
[4] Miller T. J. E., “Switched reluctance motors and their control,” in Monographs in Electrical and
Electronic Engineering. Oxford, U.K.: Clarendon, 1993.
[5] Kjaer P. C., Gribble J. J., and Miller T. J. E., “High grade control of switched reluctance
machines,” IEEE Trans. Ind. Applicat., vol. 33, pp. 1585–1593, Nov./Dec. 1997.
[6] Miller T. J. E., Ed., Electronic Control of Switched Reluctance Motors. ser. Newnes Power
Engineering Series. Oxford, U.K.: Newnes, 2001.
[7] Lyons J. P., MacMinn S. R., and Preston M. A., “Flux/current methods for SRM rotor position
estimation,” in Proc. IEEE-IAS Annu. Meeting, 1991, pp. 482–487.
[8] Hedlund G. and Lundberg H., “Motor energizing circuit,” U.S. Patent 4 868 478, Sept. 19, 1989.
[9] Husain I. and Ehsani M., “Rotor position sensing in switched reluctance motor drives by
measuring mutually induced voltages,” IEEE Trans. [Link]., vol. 30, pp. 665–672,
May/June 1994.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 599-611

MATHEMATICAL MODELLING AND SIMULATION

OF SPACE VECTOR MODULATION FOR THREE
LEVEL DIODE CLAMPED INVERTER

PRASHANTH PURAD, GOPALRADDI, AKSHAYA KATKAR

AND VEERESH BYAHATTI
Department of Electrical and Electronics Engineering,
B.V.B. College of Engineering and Technology,
Hubli 580031, Karnataka, India

Abstract
This paper describes implementation of Space Vector Pulse Width Modulation technique for a three
phase three level VSI using Matlab/Simulink and also a comparison is made with SPWM to show that
Space Vector Modulation (SVM) technique provides sinusoidal output waveform with reduced
harmonic contents. Multilevel inverters are increasingly being used in high-power, medium voltage
applications due to their superior performance compared to two-level inverters. The paper also gives
an overview of three level Diode Clamped inverter.

-----------------------------------
Keyword : Sine Pulse Width Modulation (SPWM), Space Vector Modulation (SVM), harmonic
spectrum, Diode Clamped Inverter (DCI), Nearest Three Vector (NTV)
© http: //[Link]
600 PRASHANTH PURAD, GOPALRADDI, AKSHAYA KATKAR AND VEERESH BYAHATTI

1. INTRODUCTION
Multilevel inverters are increasingly being used in high-power medium voltage applications
due to their superior performance compared to two-level inverters, such as lower common-
mode voltage, lower dv/dt stress, lower harmonics in output voltage and current and reduced
voltage on the power switches. Among various modulation techniques for a multilevel
inverter, Space Vector Pulse Width Modulation (SVPWM) is an attractive candidate due to
the following merits. It directly uses the control variable given by the control system and
identifies each switching vector as a point in complex (α, β) space. It is suitable for digital
signal processor (DSP) implementation. It can optimize switching sequences. The space
vector diagram of a three-phase three level inverter shown in Fig. 2 consisting of six sectors.
Each sector consists of (n+1) triangles. The tip of the reference vector can be located within
any triangle. Each vertex of any triangle represents a switching vector. A switching vector
represents one or more switching states depending on its location. For a three level inverter
there are 27 switching states in the space vector diagram of a 3-level inverter. The SVPWM
is performed by suitably selecting and executing the switching states of the triangle for the
respective on-times. It is known as Nearest Three Vector (NTV) approach. The performance
of the inverter significantly depends on the selection of these switching states.

2. THREE LEVEL INVERTER

Fig.1 shows a schematic diagram of a three level Diode Clamped Inverter. The Diode-
Clamped multilevel inverter consists of clamping diodes and cascaded dc capacitors to
produce ac voltage waveforms with multiple levels. There are three arms of the inverter for a
three phase inverter. Each arm is made up of four active switches T a1 to Ta4,Tb1 to Tb4 and Tc1
to Tc4 respectively on each leg with four anti parallel diodes. On the dc side of the inverter,
the dc bus capacitor is split into two capacitors, providing a neutral point N. The diodes
connected to the neutral point, Da1 and Da2, are the clamping diodes.
When switches Ta2 and Ta3 are turned ON, the inverter output terminal A is connected to the
neutral point through one of the clamping diodes. The voltage across each of the dc
capacitors is Vdc/2, which is normally equal to half of the total input dc voltage V dc. The
switch voltage is limited to half the level of the dc-bus voltage Vdc. Thus, the voltage stress
of switching device is greatly reduced. The output voltage Vo has three different states: +Vdc,
0 and -Vdc. [2] Taking phase A as an example, For voltage level +Vdc, Ta1 and Ta2 need to be
MATHEMATICAL MODELLING AND SIMULATION… 601

turned ON; for 0 level, Ta2 and Ta3 need to be turned ON; and for -Vdc, Ta3 and Ta4 need to be
turned ON. We can define these states as P, O, and N, respectively.

Fig. 1 Basic configuration of three-level Neutral Point Clamped Inverter

Fig. 2 Space vector diagram for three level VSI

602 PRASHANTH PURAD, GOPALRADDI, AKSHAYA KATKAR AND VEERESH BYAHATTI

3. DIFFERENT MODULATION TECHNIQUES

3.1. Sinusoidal Carrier Pulse width Modulation
This highly popular method is based on the comparison of a sinusoidal reference Vref with
two carriers Vcr1 and Vcr2. As shown in Fig. 3 and pulse are generated as shown below.
This method is simple and can be easily implemented.
 If Vref > Vcr1Ta1=ON ; Ta2=ON Vrn = Vdc;
 If Vref < Vcr2 Ta1=OFF; Ta2=OFF Vrn = -Vdc; else Vrn = 0;

Fig. 3 SPWM Technique

3.2. Space Vector Modulation Algorithm
A three-level inverter is characterized by 33= 27 switching states as indicated in Fig.2 where
the space vector diagram for the three-level inverter which is divided into the six sectors (S1-
-S6) is shown. There are 24 active states, and three zero states that lie at the centre of the
hexagon. Each sector has four regions. Switching state ’P’ denotes that the upper two
switches in leg A are ON and the inverter terminal voltage Ua, which is the voltage at
terminal A with respect to the neutral point NP, is +Vdc/2, whereas ’N’ indicates that the
lower two switches conduct, leading to Ua =Vdc/2. Switching state ’O’ signifies that the
inner two switches Ta2 and Ta3 are ON and --------- is clamped to zero through the clamping
diodes. [6]The principle of SVPWM method is that the command voltage vector is
MATHEMATICAL MODELLING AND SIMULATION… 603

approximately calculated by using three adjacent vectors. The duration of each voltage
vectors obtained by vector calculations;
T1V1+T2V2+T3V3 =TSV*
T1 + T2 + T3 = TS
Where V1, V2, and V3 are vectors that define the triangle region in which V* is located. T1,
T2 and T3 are the corresponding vector durations and Ts is the sampling time. In a three-level
inverter similar to a two-level inverter, each space vector diagram is divided into 6 sectors.
For simplicity here only the switching patterns for Sector 1 will be defined so that
calculation technique for the other sectors will be similar. Sector 1 is divided into 4 regions
as shown in Fig.4 where all the possible switching states for each region are given as well.
SVPWM for three level inverters can be implemented by using the steps of sector
determination, determination of the region in the sector, calculating the switching times,
T1,T2, T3 and finding the switching states. The complexity is due to the difficulty in
determining the location of the reference vector, the calculation of ON times, and the
determination and selection of switching states..
3.2.1 Determining the sector
θ (Theta) is calculated using the Equation 3 and then the sector, in which the command
vector Vref is located, is determined as;
If θ is between 0<= θ < 60, then V*is in Sector 1,
If θ is between 60 <= θ < 120, then V* is in Sector 2,
If θ is between 120 <= θ < 180, then V*is in Sector 3,
If θ is between 180 <= θ < 240, then V* is in Sector 4,
If θ is between 240 <= θ < 300, then V*is in Sector 5,
If θ is between 300 <= θ < 360, then V* is in Sector 6.
3.2.2 Determination of region
The triangle that bounds the reference vector is found based on coordinates of the tip of the
reference vector. The origin (0, 0) is then moved to the origin-vertex of the corresponding
triangle. The triangles are determined by calculating auxiliary parameters k1 and k2 which are
defined by Equation 4 and Equation 5. Vα and Vβ are coordinates of the tip of the space
vector,k1 determines whether the small triangle is in the right hand side of the sector (k 1=1)
or in the left-hand side (k1=0). k2 determines if it is in the upper half (k2=1) or in the lower
half (k2=0). The reference vector is shifted to the new set of axes that intersect at the main
604 PRASHANTH PURAD, GOPALRADDI, AKSHAYA KATKAR AND VEERESH BYAHATTI

vertex (or origin) of the triangle. Assuming a type one triangle, the coordinates of the tip of
the shifted reference vector Ai P ,where P= (V α,Vβ) is the tip off the original space vector
and Ai is the origin of the triangle, are found as given in Equation 6 and 7. The triangle
index is then found by using Equation 8 Once the shifted reference is found, time shares are
calculated using Equations 9,10 and 11. [6]
3.2.3 Finding the switching states
By considering the switching transition of only one device at any time, the switching orders
given below are obtained for each region located in Sector 1. Therefore, switching signals
for Sector 1 are
Region 1:-1-1-1, 0-1-1, 00-1, 000, 100, 110, 111
Region 2: 0-1-1, 1-1-1, 10-1, 100
Region 3: 0-1-1, 11-1, 10-1, 100, 110
Region 4: 00-1, 10-1, 11-1, 110

Fig.4 Sector 1

4. EQUATIONS USED TO BUILD SIMULINK MODEL FOR SVM

3 reference signal at required output frequency are generated
Va = sin(2πft)
MATHEMATICAL MODELLING AND SIMULATION… 605

Va = sin(2πft- )

Va = sin(2πft- )

Calculation of direct axis and quadrature axis component

Vα =Va (1)

Vβ= (Vb-Vc) (2)

θ =tan-1 ( ) (3)

k1 = [ Vα + ] (4)

k1 = [ ] (5)

Time Calculation

Vαi = Vα - k1+ * k2 (6)

Vβi = Vβ - (7)

Delta = k21 + 2k2 (8)

T1 = Ts[ Vαi+ ] (9)

T2 = Ts[ ] (10)

T3 = Ts – T2 – T3 (11)

Fig. 5 shows the complete flow chart for implementing a three level Space Vector
Modulation technique
606 PRASHANTH PURAD, GOPALRADDI, AKSHAYA KATKAR AND VEERESH BYAHATTI

Fig. 5 Flowchart for the proposed scheme.

5. SIMULATION RESULTS
Parameters
Vdc =200 V
Load= (25+j3.7)
Output frequency = 50 Hz
MATHEMATICAL MODELLING AND SIMULATION… 607

Switching frequency fs =1950 Hz.

5.1. Simulink model of Sine Pulse Width Modulated (SPWM) inverter
The simulated model of SPWM three level inverter is shown in Fig.5. The three blocks in the
model consists of subsystem for generation of gating signal. Fig. 6 shows the line voltage of
SPWM inverter which consists of five voltage level . The current drawn
by the load is shown in Fig.7 it can be seen from the figure the current waveform in not a
pure sine wave, it consists of harmonics as shown in Fig.8. from the harmonic spectrum of
the line voltage.

Fig.5 Simulink model of SPWM three level inverter

Fig.6 Line voltage waveform of SPWM three level inverter

608 PRASHANTH PURAD, GOPALRADDI, AKSHAYA KATKAR AND VEERESH BYAHATTI

Fig.7 Current waveform for SPWM inverter

Fig.8 Harmonic Spectrum of Line voltage for SPWM

5.2. Simulink model of Space Vector Modulated Inverter
The Fig.9 shows the complete Simulink model of Space Vector Modulation. It consists of
various subsystems, such as
 Sector identification
 Determination of the region in the sector
 Subsystem for calculating the switching time T1,T2,T3
MATHEMATICAL MODELLING AND SIMULATION… 609

 Finding switching states

Fig.9 Simulink model of SVM three level inverter

The various equations used to build the mathematical model is already been discussed above.
Fig.10 shows the line voltage of the three level inverter. It can been seen that the output
voltage switches from +Vdc to +Vdc/2 then to 0. In the negative cycle from 0 to -Vdc/2 then -
Vdc.

Fig. 10 Line voltage waveform of SVM three level inverter

Fig. 11 shows the three phase current waveform. It can be seen that the output current
waveform is more sinusoidal when compared to SPWM technique. It is this feature of the
SVM technique that makes it much popular and widely used when compared to other
modulation technique.
610 PRASHANTH PURAD, GOPALRADDI, AKSHAYA KATKAR AND VEERESH BYAHATTI

Fig. 12 shows the Harmonic Spectrum SVM inverter. It can be seen from the graph that the
magnitude of harmonic content is very small at lower frequencies. The harmonics at higher
frequencies can be easily eliminated using harmonic filters.

Fig. 11 Phase current waveform

Fig. 12 Harmonic Spectrum of Line voltage for SVM inverter

CONCLUSIONS
In this paper mathematical modelling and simulation of space vector modulation is presented
and it is shown that SVM technique has more advantages over other modulation technique as
MATHEMATICAL MODELLING AND SIMULATION… 611

the harmonic contents is greatly reduced and the output voltage and current is more
sinusoidal then other modulation technique. Hence SVM is considered as a most efficient
candidate due to the above stated advantages.. The use of Multilevel inverter ensures lower
dv/dt stress on the switches, lower harmonics in output voltage and current, and reduced
voltage on the power switches i.e. switches of lower power rating can be used.

REFERENCES

[1] Ned Mohan, Tore M undeland, William P RobbinsPower ElectronicsConverters, Application

and Design -2nd Edition, Jhon Wiley and sons.
[2] Obulesu P., Kumar M. V., Design and simulation of direct torque controlof Induction Motor
drive using Matlab/Simulink,International Journal ofPower and Energy Systems, Vol. 27, No.2,
2007.
[3] Multilevel Inverters: A Survey of Topologies,Controls, and Applications,Jos Rodrguez, Senior
Member, IEEE, Jih-Sheng Lai, Senior Member,IEEE, and Fang Zheng Peng, Senior Member,
IEEE, IEEE transactionson industrial electronics, vol. 49, no. 4, august 2002.
[4] Gupta A., Khambadkone A. M. A space vector pwm scheme for multilevelinverters based on
two-level space vector pwm,IEEE Transactionon Industrial Electronics, Vol. 53, October 2006.
[5] Celanovic, N.; Boroyevich, D, comprehensive study of three-levelneutral-clamped voltage
source PWM inverters,”IEEE Trans. IndustrialElectronics,
[6] Space Vector Modulation for Voltage-Source Inverters: A Unified Approach,H. Pinheiro, F.
Bottern, C. Rech, L. Schuch, R. F. Camargo, H.L. Hey, H. A. Grndling, J. R. Pinheiro Power
Electronics and Control Research Group GEPOC
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 613-624

EMBEDDED SYSTEM FOR ON-LINE FAULT

DETECTION OF INDUCTION MOTORS

S. M. SHASHIDHARA1 AND P. SANGAMESWARA RAJU2

1
Professor, Dept of Electronics and Communication Engineering,
Proudhadevaraya Institute of Technology, Hospet, India.
2
Professor and Head, Dept of Electrical and Electronics Engineering,
Sri Venkateswara University, Turupati, India

Abstract
With the proliferation of advanced technologies in signal processing, spectrum estimation of AC
machines has lead to the accurate and precise identification of different faults. The popular machine
in most of the industrial applications is a squirrel cage induction motor and failures of such significant
motors may have severe consequences such as product quality, ageing, safety and costs. Most of the
condition monitoring techniques in induction motors generally involve with single and specific fault
identification. This paper proposes a novel investigation technique for optimized identification of two
or more combined faults of an induction motor. The contribution of this work is a new methodology
that suits for hardware development, which communicates induction motor data with Compact RIO
FPGA based controller system to identify the faults such as broken rotor bars, vibration effects
(eccentricities), leakage flux condition and stator current status. To ensure the performance response
of proposed methodology, tests are conducted on a 2 kW induction motor in a laboratory, which show
highly satisfactory results that prove its suitability for on-line detection of single faults and multiple
combined faults in a flexible way through a reconfigurable embedded implementation in a field
programmable gate array (FPGA) environment.

--------------------------------
Keywords : Fault detection, Induction Motor, Squirrel Cage, FPGA, CompactRIOTM.

© http: //[Link]
614 S. M. SHASHIDHARA AND P. SANGAMESWARA RAJU

1. INTRODUCTION
Electromechanical energy conversion systems (motors and generators) are effectively
employed for various industrial applications including renewable energy conversions,
electric drive systems, mining, aircraft, and petroleum products, etc. The induction motors
dominate other electro-mechanical rotating machines with respect to rotor inertia, maximum
efficiency, highest speed capability, size, volume and cost [1]. Hence, the assessment of
running conditions and reliability of induction machines are critical to avoid random and
catastrophic failures in most of the applications. Consequently, the issue of continuous
monitoring and noninvasive diagnosis of these induction machines is of high concern, and is
becoming increasingly significance [2].
In the last two decades, a large amount of research work has been reported on using the
stator current spectrum to sense the rotor faults associated with mechanical unbalance and
rotor broken bars [3-8]. These techniques require highly skilled operator/user in order to
distinguish a normal operating condition from a potential failure state. This is so, as the
monitored spectral components (either current or vibration) are based on number of factors,
including normal operating conditions [9].
In [10], power supply and motor faults are detected from electrical parameter measurements
through a membership functions method based on a pattern recognition solution has been
reported.
The wavelet transform techniques are other known techniques which have proven their
efficacy on induction motor fault detection, reported in [11-13]. These techniques are
associated with time-frequency information from non-stationary signals. Artificial
intelligent techniques have also found significant role in fault detection and diagnosis of
induction motors. For instance, neural networks and Weibull hazard rates techniques [14]
are employed for the determination of remaining active life bearings. In [15], the authors
have described about the severity of the broken rotor bars and load level of the system
identification by fuzzy logic and generic algorithms. In [16] and [17], the authors have made
effort in employing information entropy technique for the detection of rotor broken bars.
Unfortunately, this technique is limited to single isolated fault identification because of its
inherent statistical nature [17].
Now a day, the employment of CompactRIO is an attractive technique which exhibits
flexible and improved performance along with its compact size. This technique has been
EMBEDDED SYSTEM FOR ON-LINE FAULT… 615

employed for various applications such as [18] medical equipment monitoring, automatic
control in magnetic system and engine control in motor cycle vehicles.
Hence, the research effort has been made in the effective utilization of Laboratory Virtual
Instrument Engineering Workbench (LabVIEW) CompactRIO with Field Programmable
Gate Array (FPGA) environment for the continuous monitoring of induction motor. The
flexibility of proposed technique is that, it accepts multiple sensor signals and capable of
accommodating custom signal analysis tools. Moreover, this system provides a on-line and
continuous monitoring platform with standard custom built indicator module. In addition,
this proposed system has a special feature of visual inspection opportunity for staff about in
condition monitoring.

2. EXISTING FAULT FREQUENCIES IN INDUCTION MOTORS

It is well known that, a current spectrum of a faulty system is composed of potential fault
information. However, the frequency components of each fault can be determined through
various techniques. It is also important to know that, just as in vibration analysis, as the fault
progresses, its characteristic spectral components continue to increase over a time.
2.1 Bearing faults
The bearing fault current harmonics can be described [1], [9], and [19], by following valid
equations,
𝑓𝑏𝑛𝑔 = 𝑓𝑠 ± 𝑁𝑏 . 𝑓𝑖,𝑜 (1)
where fs is supply frequency of the system. fo and fi are related to outer and inner race,
respectively, and Nb is the number of balls, and f0 and fi are given by,
𝑓𝑟 BD 𝑓𝑟 BD
𝑓0 = 2
[1 − PD cos 𝛽] and 𝑓𝑖 = 2
[1 + PD cos 𝛽]

where, ‘BD’ and ‘PD’ are, respectively, the ball diameter and the pitch diameter, and β is the
contact angle between the balls and the ball bearing rings.
2.2 Rotor faults
End rings and broken bars of the rotor induce the same harmonics in the stator currents.
Hence, the amplitude of stator current will be modulated by the slip frequency factor, which
is given by [Link]. Here,’s’ is per unit slip. This increase in the modulation is based on the
severity of the fault. In the current spectrum, the harmonics [18-20] due to the rotor fault
can be found from the following characteristic equation:
616 S. M. SHASHIDHARA AND P. SANGAMESWARA RAJU

𝑓𝑟𝑏 = 𝑓𝑠 . (1 ± 2. 𝑛. 𝑠) (2)
where ‘n’ is an integer.
2.3 Stator Faults
The stator current related faults can be identified by using the fault frequency measurement,
which can be observed by the FFT analysis of the voltage, current and axial flux signals [18-
19] and [21].
𝑛(1−𝑠)
𝑓𝑠𝑓 = 𝑓𝑠 [ 𝑝
± 𝑘] (3)

where ‘k’ is an integer, and ‘p’ is the number of pole pairs of the induction motor. It has
been reported that, [Link] or a combination of rotor frequency (fr), [Link], and [Link] can be a
symptom of stator faults in vibration monitoring.
2.4 External Fault Frequencies
In the literature, other possible random faults of Induction Motor have been reported in [1],
[9], [18], and [19]. The summary of these fault frequencies are listed in the Table I. The
term “soft foot” is the term used for the inappropriate contact between a machine casing, and
the baseplate used to support it. Softfoot is a common issue when aligning rotating
equipment.
Table 1 : Physical Set-Up related Fault Frequencies

Faults Frequency Levels

Mechanical Unbalance 1 (fr)

Misalignment 1 (fr), 2 (fr), and 3 (fr)

0.5 (fr), 1(fr), 1.5(fr),2(fr), 3

Soft Foot (fr), 4 (fr), 5 (fr), 6 (fr),
7 (fr),and 8 (fr),

3. COMPACTRIO BASED CONDITION MONITORING SYSTEM

The proposed On-line condition monitoring of Induction Motor is implemented in FPGA
environment for obtaining on-line operation with early detection of faults. It is
recommended that, the CompactRIO is most suitable system for real time processing units
[18], [22], which can be easily reconfigurable, and provides high speed parallel operation of
EMBEDDED SYSTEM FOR ON-LINE FAULT… 617

tasks. It supports LabVIEW programs for the real time processing of systems in FPGA
environment. The optimized program has been developed for condition monitoring of
induction motor in the LabVIEW environment which significantly accelerates the execution
time and provides an open architecture for further modifications.

Fig. 1 Block diagram of CompactRIO based Induction Motor Fault detection System

Fig. 1 shows the complete block diagram of CompactRIO based condition monitoring
system of Induction motor. In this proposed system, CompactRIO mainly composed of RT
Controller 9004, Artificial Intelligence (AI) module and FPGA 9104 unit. Suitable sensors
are employed for data acquisition from the machine and for further analysis in CompactRIO
system.
CompactRIO system: This is heart of the system which essentially reduces the complexity,
running time as well as provides tools for monitoring and identification of faults in the
system. In the proposed system Fig. 1, CompactRIO has real time controller module (cRIO-
9004), an FPGA framework and an analogue input (AI) module. It performs averaging, Fast
Fourier Transform (FFT) analysis, windowing, peak and fault detection analysis. In this,
data transfer processes has been executed effectively through the network port. However,
read/write instruction, execution time critical loop tasks, reading data from machine status
has been carried out by FPGA Chassis. The analog input module is employed for capturing
vibrations, current, voltage and flux signals using suitable signal conditioning devices and
sensors. Fig. 2 shows the flow chart of the developed program for condition monitoring.
A. Indicator Module: A custom design indicator module is aimed to provide an on-line
fault diagnosis to the operator/user. It mainly deals with various groups of machine faults
and its associated levels. It also indicates the fault status and condition of the machine. The
618 S. M. SHASHIDHARA AND P. SANGAMESWARA RAJU

module is controlled by the PFGA Chassis via the Serial Peripheral Interface (SPI) protocol.
The CompactRIO based condition monitoring system is programmed using LabVIEWTM
along with necessary constraints based on motor condition. Fig. 2 shows flow-chart of the
developed software for fault identification and detection. As it is depicted in flow-chart
Fig.2, FPGA performs multiple operations simultaneously such as, acquiring the 8-channel
data & performing low-pass anti-aliasing filtering, and transferring the data via direct
memory access (DMA). It also drives the custom-made indicator module via SPI.
B. Execution of Software: The Induction machine data analysis is mainly performed in
the real time controller. In this execution, the process sequence is as follows: initialization
of data, machine data calibration from relevant sensors, followed by FFT analysis, fault
frequency peak detection, and frequency averaging. Once FFT analysis of induction motor
is monitored, different subroutines are implemented to detect faults based on the suitable
fault frequency equations given in section II. Algorithms are implemented to estimate the
specific fault levels based on the amplitudes of the fault frequencies.

Fig. 2 Flowchart in FPGA system

EMBEDDED SYSTEM FOR ON-LINE FAULT… 619

4. EXPERIMENTAL RESULTS
To ensure the expected performance and to analyze the steady-state response of the proposed system,
various tests have been conducted. The steady-state current signal obtained from the motor is used to
find the multiple combined faults and its corresponding classification, this is because, the motor
current signal describes the dynamic characteristics of the induction motor. The specification of the
induction motor under test considered for various faults is 4-pole, 2 kW, 420 V, and 5 A. The
sampling frequency is 2000 samples and the anti aliasing frequency is 2 kHz.

Fig. 3 Experimental Set-Up of CompactRIOTM based Induction Motor Fault detection System

Figure 4 NI CompactRIOTM Module

620 S. M. SHASHIDHARA AND P. SANGAMESWARA RAJU

In Compact RIO system, an execution Trace Toolkit from National Instruments (NI) Inc TM. has been
used for measurement of execution time. In the time analysis, it has been noticed that, 8-channel
acquisition and cut-off filtering processes have been take less than 1.1 ms. It was computed that
bearing, stator, rotor, supply faults frequencies model calculations takes 2ms time and Limit testing of
bearing, stator, rotor, set-up, eccentricities, and supply faults takes 25 ms This is the time duration for
the execution of the fault detection and the diagnosis algorithms. The execution for the various
sections such as signal processing algorithms in the FPGA environment is depicted in Fig 8.
As it can be seen in the Table II, only the Limit testing of bearing, rotor, set-up,
eccentricities, and supply faults algorithm have taken the longest time among the other
algorithms. Furthermore, for 8-channel monitoring, total processing time, fault analysis and
Compact RIO standard operating system, and opening and closing applications took
approximately 200 ms. Through the experimental investigation, it has been noticed that, the
execution time durations of the algorithms for the proposed work are in acceptable range for
the condition monitoring system of induction motor applications.
Fig. 4 shows the zoomed section of spectrum lines at(1 ± 𝑠)𝑓𝑠 for the possible diagnostic
evaluations of the broken rotor bars in which, the Discrete Fourier Transforms (DFT)
amplitude in range 42 Hz-66 Hz broken bar signals are shown. However, the other spectrum

lines can be detected in addition to the fundamental signal, like lines of eccentricity (1 ±
(1−𝑠)
𝑝
) 𝑓𝑠 , rotor bar breakage lines (1 ± 2𝑠)𝑓𝑠 .

Fig. 5 demonstrates the performance of the proposed system for multiple analog signal
inputs, which are considered simultaneously. Fig. 7 shows the axial leakage flux of the
motor observed on the LabVIEW front panel screen, and Fig. 8 illustrates the three
vibrations signals for the multiple faults of broken rotor bars and soft-foot. As it can be
noticed from the frequency spectrum of Fig. 7, several test set-up related fault frequency
amplitudes can be identified by the vibration sensors, and the rotor and stator fault frequency
amplitudes can be identified using the axial leakage flux sensor.
Stator-current spectrum of the induction motor around the rotor-slotting frequency is
indicated in Fig 6. As anticipated, the test is capable of indicating with accuracy the rotor-
slotting frequencies. Sideband lines appear in the frequency spectrum around the supply
frequency fs (Fig. 6), caused by broken bars. Fig 8 indicates the real time three axes
vibration signals. Fig 9 gives the details of the time duration limits taken for various
components of the software execution.
EMBEDDED SYSTEM FOR ON-LINE FAULT… 621

Figure 5 Three phase voltage waveforms in Compact RIO based Induction Motor Diagnosis

Figure 6 Identification of broken rotor bars of Induction Motor

Figure 7 Real time axial flux leakage signal

622 S. M. SHASHIDHARA AND P. SANGAMESWARA RAJU

Figure 8 Real time three axes vibration signals

Fig 9 time duration limits for various components of the software execution.

The repeatability tests also have been performed for the above mentioned faults. During
these tests analysis, the motor with the fault run continuously and the corresponding sensor
signals have been tracked and analyzed in the CompactRIO based system.

CONCLUSIONS
In this paper, a Compact RIO based condition monitoring system for identification of faulty
section of an Induction Motor has been described and presented. The existing various fault
EMBEDDED SYSTEM FOR ON-LINE FAULT… 623

frequency analysis of Induction Motors are described and corresponding mathematical

analysis has been discussed. Effort is made on the optimized utilization of Compact RIO
system in FPGA environment for fault identification and detection of Induction Motor. The
paper illustrates the development of software and hardware units, which are established in
the reconfigurable Compact RIO environment. Broken rotor bar faults and three axes
vibration effects have been tracked and monitored from the proposed method of
identification. The estimated cost of the proposed system is comparatively less than that is
priced in the commercial rate, but still the cost can be even reduced considerably for the
multiple units of such systems.

REFERENCES

[1] Hachemi, B. M. EI., and Kliman, G. B., “What stator current processing-based technique to use
for induction motor rotor faults diagnosis?,” IEEE Trans. on Energy Conversion, Vol. 18, No. 2,
pp. 238-244, June 2003.
[2] Benbouzid, M. E. H., “Bibliography of Induction motors faults detection and diagnosis,” IEEE
Trans. Energy Conversion, Vol. 14, No. 4, pp. 1065-1074, December 1999.
[3] Chow, T. W. S., and Fei, G., “Three phase induction machines asymmetrical faults identification
using bi spectrum,” IEEE Trans. Energy Conversion, Vol. 10, pp. 688-693, December 1995.
[4] Kilman, G. B., and Stein, J., “Methods of motor current signature analysis,” Elec. Mach. Power
Syst., Vol. 20, pp. 463-474, September 1992.
[5] Schoen, R. R. T., Habetler, T. G, Kamran, F., and Bartheld, R. G., “Motor bearing damage
detection using stator current monitoring,” IEEE Trans. Ind. Applications., Vol. 31, pp. 1274-
1279, Nov. 1995.
[6] Kilman, G. B., and Stein, J., “Induction motor fault detection via passive current monitoring,” in
Proceedings Intl. Conf. Electrical Machines, Vol. 1, pp. 13-17, Cambridge, 1990.
[7] Schoen, R. R., Lin, B. K., Habetler, T. G., Schlag, J. H., and Farag, S., “An unsupervised, on-
line system for induction motor fault detection using stator current monitoring,” IEEE Trans.
Indl. Applications., Vol. 31, pp. 1280—1286, November 1995.
[8] Filippetti, F., Franceschini, G., Tassoni, C., and Vas, P., “A fuzzy logic approach to on-line
induction motor diagnostics based on stator current monitoring,” in Proceedings, IEEE-Intrl.
Conf., Power tech., Vol. EMD, Sweden, pp. 156-161.
[9] Benbouzid, M. E. H., Vieira, M., and Theys, C., “Induction motors faults detection and
localization using stator current advanced signal processing techniques,” IEEE Trans. Power
Electron., Vol. 14, No. 1, pp. 14-22, January 1999.
[10] Ondel, O., Boutlexu, C. E., and Blanco, E., “Fault detection and diagnosis in a set inverter-
induction machine through multidimensional membership function and pattern recognition,”
IEEE Trans. Energy Conversion, Vol. 24, No. 2, pp. 431-441, June 2009.
624 S. M. SHASHIDHARA AND P. SANGAMESWARA RAJU

[11] Cusido, J., Romeral, L., Ortega, A., Rosero, J. A., and Garcia-Espinosa, A., “Fault detection in
induction machines using power spectral density in wavelet decomposition,” IEEE Trans. Indl.
Electron., Vol. 55, No. 2, pp. 633-643, February 2008.
[12] Ordaz-Moreno, A., Romero-Troncoso, R. J., Vite-Frias, J. A., Rivera-Guillen, J. R., and Garcia-
Perez, A., Automatic online diagnosis algorithm for broken-bar detection on induction motors
based on discrete wavelet transform for FPG implementation,” IEEE Trans. Indl. Electron., Vol.
55, No. 5, pp. 2193-2202, May 2008.
[13] Kia, S. H., Henao, H., and Capolino, G. A., “Diagnosis of broken-bar fault in induction
machines using discrete wavelet transform without slip estimation,” IEEE Trans. Indl. Appln.,
Vol. 45, No. 4, pp. 1395-1404, July 2009.
[14] Mohammad, A. K., Saon, S., and Hiyama, T., “Predicting remaining useful life of rotating
machinery based artificial neural network,” Comput. Math. Appl., Vol. 60, No. 4, pp. 1078-
1087, August 2010.
[15] Razik, H., De Rossister-Correa, M. B., and Da Silva, E. R. C., “A novel monitoring of load level
and broken bar fault severity applied to squirrel-cage induction motors using a genetic
algorithm,” IEEE Trans. Indl. Electron., Vol. 56, No. 1, pp. 4615-4626, November 2009.
[16] Bouzida, A., Touhami, O., Ibtiouen, R., Fadel, M., Rezzoug, A., and Beloucherani, A., “Fault
diagnosis in industrial induction machines through discrete wavelet transform,” IEEE Trans.
Indl. Electron., Vol. 58, No.9, pp. 4385-4395, September 2011.
[17] Romero-Troncoso, R. J., Saucedo-Gallaga, R., Cabal-Yepez, E., Garcia-Perez, A., Osornio-Rios,
R. A., Alvarez-Salas, R., Miranda-Vidales, H., and Huber, N., “FPGA-Based online detection of
multiple combined Faults in induction motors through information entropy and fuzzy inference,”
IEEE Trans. Indl. Electron. Vol. 58, No. 11, pp. 5263-5269, November 2011.
[18] Bakhri, S., Ertugrul, N., Soong, W. L., and AI-Sarawi, S., “Investigation and development of a
real-time on-site condition monitoring system for induction motors,” in Proceedings
Australasian universities Power Engg., Conf. pp. 1-6, Perth, Australia, 9-12, December 2007.
[19] Rangel-Magdaleno, J. J., Romero-Troncoso, R> J., Osornio-Rios, R. A., Cabal-Yepez, E., and
Contreras-Medina, L. M., “Novel methodology for online half-broken-bar detection on
induction motors,” IEEE Trans. Instrumentation and Measurement, Vol. 58, No. 5, pp. 1690-
1698, May 2009.
[20] Thomson, W. T., and Fenger, M., “Current signature analysis to detect induction motor faults,”
IEEE Trans. Indl. Appl., Mag., vol. 7, pp. 26-34, 2001.
[21] Mathur, N., “NI CompactRIO – Reconfigurable control and acquisition system”, Application
Note, National Instrument, 2006.
[22] Website: [Link]
[22] Website: [Link]
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 625-637

SIMULATION OF STATOR ORIENTED VECTOR

CONTROLLED DOUBLY-FED INDUCTION
GENERATOR FOR HARNESSEING
WIND ENERGY EFFECTIVELY

CHIRAG VADALIYA1, AMIT N PATEL1 AND VINOD PATEL2

1
Department of Electrical Engineering, Institute of Technology,
Nirma University, Ahmedabad, Gujarat, India.
2
Amtech Electronics (India) limited, Road Number 6,
Sector 26, Gandhinagar, Gujarat, India.

Abstract
Wind energy plays an increasingly important role in the world because it is friendly to the
environment and limitless. During the last decades, the concept of a variable-speed wind turbine (WT)
has been receiving increasing attention due to the fact that it is more controllable, efficient, and has
good power quality. In order to most effectively utilize the wind energy and improve the efficiency of
wind generation system an optimum control strategy of doubly-fed induction generators (DFIG) is
proposed. This paper presents a stator flux oriented vector control strategy for a 2MW/690V doubly
fed induction generator (DFIG)-based wind energy generation system to control the rotor side
converter to control the active and reactive power and grid side converter control to maintain dc link
voltage constant.

---------------------------------
Keywords : Doubly-fed induction generator (DFIG), Vector control, active and reactive power
control, Wind power. © http: //[Link]
626 CHIRAG VADALIYA, AMIT N PATEL AND VINOD PATEL

1. INTRODUCTION
Due to the increasing concern about CO2 emissions, renewable energy systems and
especially wind energy generation have attracted great interests in recent years. Large wind
farms have been installed or planned across the world and the power ratings of the wind
turbines and wind farm are increasing. Many studies [1-3] are oriented toward this type of
energy production in the aim to make it more efficient. Wind turbine must be adjusted
accordingly to wind speed, hence the variable speed generator based wind turbines are
mostly used in wind power industry to capture more energy from wind [4]. There are many
type of variable speed wind generator like permanent magnet synchronous generator
(PMSG), squirrel cage induction generator (SCIG) and doubly fed induction generator
(DFIG). Out of these control and performance of DFIG based wind energy conversion
system is analyzed in this paper.
The DFIG based wind turbine offers several advantages over to other wind turbines
including variable speed operation (± 30% around the synchronous speed) and four
quadrants active and reactive power capabilities. Such a system also results in lower
converter costs (typically 30% of total system power) and lower power losses compared to a
system based on a fully fed synchronous generator with full rated converter. It is also
capable of generating active power at constant frequency and the generated active and
reactive power is controlled independently for better grid integration.
Figure 1 shows the DFIG system. The stator of the DFIG is directly connected to the grid
and rotor is connected to grid through four quadrant power converters comprises of two back
to back PWM-VSC (rotor side converter and grid side converter). The rotor side converter
(RSC) controlling active and reactive power and the grid side converter (GSC) maintain DC-
link constant and controlling reactive power to control power factor. The converters are
controlled using vector control technique. The stator voltage oriented vector control strategy
is used to de-couple control of active and reactive power. This paper explains the model of
DFIG in the “d-q reference frame” and stator flux oriented vector control strategy of DFIG.
SIMULATION OF STATOR ORIENTED VECTOR… 627

Fig.1 DFIG based wind energy conversion system

2. MODELLING OF DFIG
The induction machine d-q or dynamic equivalent circuit is shown in Figure 2. Based on the
equivalent circuit, the main equations of doubly fed induction generator stator and rotor
voltage can be written as follows in matrix form:
𝑉𝑠𝑎𝑏𝑐 = 𝑟𝑠 𝑖𝑎𝑏𝑐𝑠 + 𝑝𝜆𝑎𝑏𝑐𝑠 (1)
𝑉𝑟𝑎𝑏𝑐 = 𝑟𝑟 𝑖𝑎𝑏𝑐𝑟 + 𝑝𝜆𝑎𝑏𝑐𝑟 (2)
Applying synchronously rotating reference frame transformation [9] to equation (1) and (2),
the voltage equations become
𝑉𝑑𝑠 = 𝑟𝑠 𝑖𝑑𝑠 − 𝜔𝑠 𝜆𝑞𝑠 + 𝑝𝜆𝑑𝑠 (3)
𝑉𝑞𝑠 = 𝑟𝑠 𝑖𝑞𝑠 + 𝜔𝑠 𝜆𝑑𝑠 + 𝑝𝜆𝑞𝑠 (4)
𝑉𝑑𝑟 = 𝑟𝑟 𝑖𝑑𝑟 − (𝜔𝑠 − 𝜔𝑟 )𝜆𝑞𝑟 + 𝑝𝜆𝑑𝑟 (5)
𝑉𝑞𝑟 = 𝑟𝑟 𝑖𝑞𝑟 + (𝜔𝑠 − 𝜔𝑟 )𝜆𝑑𝑟 + 𝑝𝜆𝑞𝑟 (6)
Where 𝜔𝑠 is the rotational speed of the synchronous reference frame, 𝜔𝑟 is the rotor speed,
and the flux linkages are given by
𝜆𝑑𝑠 = 𝐿𝑙𝑠 𝑖𝑑𝑠 + 𝐿𝑚 (𝑖𝑑𝑠 + 𝑖𝑑𝑟 ) = 𝐿𝑠 𝑖𝑑𝑠 + 𝐿𝑚 𝐼𝑑𝑟 (7)
𝜆𝑞𝑠 = 𝐿𝑙𝑠 𝑖𝑞𝑠 + 𝐿𝑚 (𝑖𝑞𝑠 + 𝑖𝑞𝑟 ) = 𝐿𝑠 𝑖𝑞𝑠 + 𝐿𝑚 𝐼𝑞𝑟 (8)
𝜆𝑑𝑟 = 𝐿𝑙𝑟 𝑖𝑑𝑟 + 𝐿𝑚 (𝑖𝑑𝑟 + 𝑖𝑑𝑠 ) = 𝐿𝑟 𝑖𝑑𝑟 + 𝐿𝑚 𝐼𝑑𝑠 (9)
𝜆𝑞𝑟 = 𝐿𝑙𝑟 𝑖𝑞𝑟 + 𝐿𝑚 (𝑖𝑞𝑟 + 𝑖𝑞𝑠 ) = 𝐿𝑟 𝑖𝑞𝑟 + 𝐿𝑚 𝐼𝑞𝑠 (10)
628 CHIRAG VADALIYA, AMIT N PATEL AND VINOD PATEL

Where 𝐿𝑠 = 𝐿𝑙𝑠 + 𝐿𝑚 and 𝐿𝑟 = 𝐿𝑙𝑟 + 𝐿𝑚 ; 𝐿𝑙𝑠 , 𝐿𝑙𝑟 and 𝐿𝑚 are stator and rotor leakage
inductance and mutual inductances, respectively.
Neglecting the power losses associated with the stator resistance, the active and reactive
stator powers are:
3
𝑃𝑠 = 2 (𝑣𝑞𝑠 𝑖𝑞𝑠 + 𝑣𝑑𝑠 𝑖𝑑𝑠 ) (11)
3
𝑄𝑠 = 2 (𝑣𝑞𝑠 𝑖𝑑𝑠 − 𝑣𝑑𝑠 𝑖𝑞𝑠 ) (12)

Fig.2 Equivalent circuit of DFIG in q-d axis

SIMULATION OF STATOR ORIENTED VECTOR… 629

3. DESIGN OF CONTROL SYSTEM

The objective of the RSC is to govern both the stator-side active and reactive powers
independently, while the objective of the GSC is to keep the dc-link voltage constant
regardless of the magnitude and direction of the rotor power. The GSC control scheme can
also be designed to regulate the reactive power.
3.1 Design of RSC controller
The RSC control scheme consists of two cascaded control loops. The inner current control
loops regulate independently the d-axis and q-axis rotor current components 𝑖𝑑𝑟 and 𝑖𝑞𝑟 ,
according to synchronously rotating reference frame. The stator-flux oriented reference
frame [10] is the most commonly used one. The outer control loops regulate both the stator
active power (and the generator rotor speed) and reactive power independently.
Aligning the q-axis of the reference frame along the stator-voltage position 𝑣𝑞𝑠 =𝑣𝑠 and 𝑣𝑑𝑠
is zero, since the amplitude of the supply voltage 𝑣𝑞𝑠 is constant. The active and reactive
power will be proportional to 𝑖𝑞𝑠 and 𝑖𝑑𝑠 respectively.
Hence, the active and reactive power from equation (11) and (12) are
3
𝑃𝑠 = 2 𝑣𝑞𝑠 𝑖𝑞𝑠 (13)
3
𝑄𝑠 = 2 𝑣𝑞𝑠 𝑖𝑑𝑠 (14)

The active and reactive power will be proportional to 𝑖𝑞𝑠 and 𝑖𝑑𝑠 respectively.
Since the stator is connected to the grid, and the influence of the stator resistance is small,
the stator magnetizing current 𝑖𝑚𝑠 can be considered constant. In the stator-flux oriented
reference frame, the d-axis is aligned with the stator flux linkage vector 𝜆𝑠 , namely, 𝜆𝑑𝑠 =
𝜆𝑠 and 𝜆𝑞𝑠 = 0. This gives the following relationships
𝐿𝑚
𝑖𝑞𝑠 = − 𝑖 (15)
𝐿𝑠 𝑞𝑟
𝐿𝑚
𝑖𝑑𝑠 = − (𝑖
𝐿𝑠 𝑚𝑠
− 𝑖𝑑𝑟 ) (16)
𝑣𝑞𝑠 −𝑟𝑠 𝑖𝑞𝑠
Where, 𝑖𝑚𝑠 = (17)
𝜔𝑠 𝐿𝑚

𝑣𝑑𝑟 = 𝑟𝑟 𝑖𝑑𝑟 + 𝜎𝐿𝑟 𝑝𝑖𝑑𝑟 − 𝑠𝜔𝑠 𝜎𝐿𝑟 𝑖𝑞𝑟 (18)

𝑖𝑚𝑠
𝑣𝑞𝑟 = 𝑟𝑟 𝑖𝑞𝑟 + 𝜎𝐿𝑟 𝑝𝑖𝑞𝑟 + 𝑠𝜔𝑠 (𝜎𝐿𝑟 𝑖𝑑𝑟 + 𝐿𝑚 2 𝐿𝑠
) (19)

𝐿 2
𝜎 = 1 − 𝐿 𝑚𝐿 (20)
𝑠 𝑟
630 CHIRAG VADALIYA, AMIT N PATEL AND VINOD PATEL

So, the active and reactive power are

3
𝑃𝑠 = − 2 𝜔𝑠 𝐿𝑚 2 𝑖𝑚𝑠 𝑖𝑞𝑟 /𝐿𝑠 (21)
3
𝑄𝑠 = 2 𝜔𝑠 𝐿𝑚 2 𝑖𝑚𝑠 (𝑖𝑚𝑠 − 𝑖𝑞𝑟 )/𝐿𝑠 (22)

Equations (21) and (22) indicate that active power is proportional to the 𝑖𝑞𝑟 and can be
regulated using 𝑣𝑞𝑟 (equation 19), and reactive power is proportional to the 𝑖𝑑𝑟 and can be
controlled using 𝑣𝑑𝑟 (equation 20). Consequently, the reference values of 𝑖𝑞𝑟 and 𝑖𝑑𝑟 can be
determined from the outer power control loops.

Fig. 3 schematic block diagram of RSC control.

The stator flux angle is calculated from

𝜆𝛼𝑠 = ∫(𝑣𝛼𝑠 − 𝑟𝑠 𝑖𝛼𝑠 )

𝜆𝛽𝑠 = ∫(𝑣𝛽𝑠 − 𝑟𝑠 𝑖𝛽𝑠 )

𝜆𝛽𝑠
𝜃𝑠 = tan−1
𝜆𝛼𝑠
Where, 𝜃𝑠 is stator-flux vector position
SIMULATION OF STATOR ORIENTED VECTOR… 631

Figure 3 shows the schematic block diagram of rotor side converter control. There are two
loops, outer loop is for controlling the active and reactive power and other is inner current
control loop for controlling the rotor excitation current 𝑖𝑑𝑟 and 𝑖𝑞𝑟 . The value of reference
reactive power is set to be zero. The reference active and reactive power are compared with
actual active and reactive power and the error is processed through PI controller, it will
generate reference rotor currents. The 𝑖𝑞𝑟 and 𝑖𝑑𝑟 error are processed through PI controller to
give 𝑣𝑞𝑟 and𝑣𝑑𝑟 , respectively. To ensure good tracking of these currents, compensation
terms are added to 𝑣𝑞𝑟 and 𝑣𝑑𝑟 to obtain the reference voltages 𝑣𝑞𝑟 ∗ and 𝑣𝑑𝑟 ∗ according to
𝑣𝑞𝑟 ∗ = 𝑣𝑞𝑟 + 𝑠𝜔𝑠 (𝜎𝐿𝑟 𝑖𝑑𝑟 + 𝐿𝑚 2 𝑖𝑚𝑠 /𝐿𝑠 ) (23)
𝑣𝑑𝑟 ∗ = 𝑣𝑑𝑟 − 𝑠𝜔𝑠 𝜎𝐿𝑟 𝑖𝑞𝑟 (24)
3.2 Design of GSC control
The objective of the grid-side converter is to keep the DC-link voltage constant regardless of
the magnitude and direction of the rotor power. A vector-control approach is used, with a
reference frame oriented along the stator (or supply) voltage vector position. The PWM
converter is current regulated, with the direct axis current used to regulate the DC-link
voltage and the quadrature axis current component used to regulate the reactive power. A
standard regular asymmetric sampling

Fig. 4 schematic of supply side converter.

PWM scheme [12] is used. Figure 4 shows the schematic of the supply-side converter. The
voltage balance across the inductors is
The voltage balance across the inductors is
𝑣𝑎 𝑖𝑎 𝑖𝑎 𝑣𝑎1
𝑑
(𝑣𝑏 ) = 𝑅 (𝑖𝑏 ) + 𝐿 𝑑𝑡 (𝑖𝑏 ) + (𝑣𝑏1 ) (25)
𝑣𝑐 𝑖𝑐 𝑖𝑐 𝑣𝑐1
632 CHIRAG VADALIYA, AMIT N PATEL AND VINOD PATEL

where L and R are the line inductance and resistance, respectively. Using park’s
transformation the above equation can be written into d-q reference frame (rotating at 𝜔𝑒 ) as
follows:
𝑑𝑖𝑑
𝑣𝑑 = 𝑅𝑖𝑑 + 𝐿 𝑑𝑡
− 𝜔𝑒 𝐿𝑖𝑞 + 𝑣𝑑1 (26)
𝑑𝑖𝑞
𝑣𝑞 = 𝑅𝑖𝑞 + 𝐿 𝑑𝑡
+ 𝜔𝑒 𝐿𝑖𝑑 + 𝑣𝑞1 (27)

The angular position of the supply voltage is calculated as

𝑣𝛽
𝜃𝑒 = ∫ 𝜔𝑒 𝑑𝑡 = tan−1
𝑣𝛼
Where, 𝑣𝛼 and 𝑣𝛽 are α, β stator-voltage component.
Neglecting the harmonics due to switching in the converter and the machine losses and
converter losses, the active power balance equation is as follows:
3
𝑣𝑑𝑐 𝑖𝑑𝑐 = 2 𝑣𝑑 𝑖𝑑 = 𝑃𝑟 ; 𝑣𝑞 = 0 (28)

From the equation 28, the DC link voltage may be controlled through 𝑖𝑑 control. The
reactive power flow from the source is given by
3
𝑄𝑟 = 2 𝑣𝑑 𝑖𝑞 ; 𝑣𝑞 = 0 (29)

Consequently, the reactive power from the power source to (from) the source-side converter
may be controlled through𝑖𝑞 .
In general, the reactive power from power source through the source-side converter is set to
zero (𝑖𝑞 = 0). The above vector control principles are illustrated in the generic scheme shown
in Figure 5. The DC link voltage is, in general, kept constant to take advantage of full
voltage for capacitor energy storage in the DC link.
Where,
𝑣𝑑1 ∗ = −𝑣𝑑 ′ + (𝜔𝑒 𝐿𝑖𝑞 + 𝑣𝑑 ) (30)
𝑣𝑞1 ∗ = −𝑣𝑞 ′ − (𝜔𝑒 𝐿𝑖𝑑 ) (31)
𝑣𝑑1 ∗ and 𝑣𝑞1 ∗ are the reference values for the supply-side converter, and the terms in
brackets constitute voltage-compensation terms.
SIMULATION OF STATOR ORIENTED VECTOR… 633

Fig 5. Schematic block diagram of GSC control

4. SIMULATION RESULTS
The 2MW/690V DFIG model is analyzed using PSIM under steady-state condition, the
machine parameters are given in appendix below. Figure10a-10f shows the output of DFIG
model in super-synchronous mode. The negative sign of active power means generated
active power. It is cleared from fig.6 that the machine speed in super-synchronous mode is
higher than the synchronous speed.
The vector control system based on the stator flux oriented reference frame is simulated in
PSIM, and a reference 2MW stator active power was applied to the outer control loop, and
the reactive power demand is set to zero.
634 CHIRAG VADALIYA, AMIT N PATEL AND VINOD PATEL

Fig.6a Machine speed (RPM) [X-axis: 1 div = 1 sec, Y-axis: 1 div = 500 rpm]

Fig.6b Stator Active power (pu) [X-axis: 1 div = 1 sec, Y-axis: 1 div = 1 pu]

Fig.6c Stator Reactive power (pu) [X-axis: 1 div = 1 sec, Y-axis: 1 div = 1 pu]
SIMULATION OF STATOR ORIENTED VECTOR… 635

Fig.6d Grid Current (Amp) [X-axis: 1 div = 20 msec, Y-axis: 1 div = 1k amp]

Fig.6e : Power Factor (Unity)

Fig.6f DC-link Voltage (Volt) [X-axis: 1 div = 1 sec, Y-axis: 1 div = 500 volt]
636 CHIRAG VADALIYA, AMIT N PATEL AND VINOD PATEL

The stator active power was able to track the applied reference value which confirms the
validity of the control system. The Grid side converter maintain 1000 V DC-link voltage as
shown in figure 6f and unity power factor as shown in figure 6e.

CONCLUSION
The 2 MW/690 V DFIG system has been simulated in PSIM. A bi-directional IGBT based
four quadrant AC-DC-AC converter with IGBT modules used in DFIG wind power
generation system is presented. The converter can operate at sub-synchronous and super-
synchronous modes. Based on DFIG model and field oriented control theory, the machine
side converter provides good decoupling between active and reactive powers, and the grid
side converter maintain DC-link voltage and power factor to be unity, which leads to high
power quality and higher efficiency in harnessing wind energy effectively.
Table 1 : Machine Parameters
Rated Power 2 MW
Rated voltage 690 V
Frequency 50 Hz
Stator resistance 2.6 mΩ
Rotor resistance 2.9 mΩ
Stator leakage inductance 87 µH
Rotor leakage inductance 87 µH
Magnetizing inductance 2.5 mH
No. of poles 4

REFERENCES

[1] Datta Rajib and Ranganathan V. T., Senior Member, IEEE , “Variable-Speed Wind Power
Generation Using Doubly Fed Wound Rotor Induction Machine—A Comparison With
Alternative Schemes”, IEEE Transactions on energy conversions, Vol. 17(3),. pp. 414-421,
September 2002.
[2] Hoffinann R., Mutschler P., Member, IEEE, “Comparison of Wind Turbines Regarding their
Energy Generation”, power electronics Specialists conference, Vol. 1, pp. 6-11, 2002.
[3] Zinger Donald S., Senior Member, IEEE, and Eduard Muljadi, Senior Member, IEEE,
“Annualized Wind Energy Improvement Using Variable Speeds”, IEEE Transection on industry
Applications, Vol. 33(6), November/December 1997.
SIMULATION OF STATOR ORIENTED VECTOR… 637

[4] Yanbo Che, Yu Wang, Chengshan Wang, Lin Ai, “Research on Grid-Connected Power Control
for Double-Fed Generator”, Power and Energy Engineering Conference, pp. 1-4, 2009.
[5] Muller S., Deicke M., De Doncker R.W., “ Doubly fed Induction Generator systems for wind
turbines”, IEEE transection on Industrial Applications, Vol. 8(3), pp. 16-33, May/June-2002.
[6] Bose Bimal K., “Modern Power Electronics And AC Drives”, USA, Prentice Hall PTR ,
Inc.2002.
[7] Krause Paul C., “Analysis of Electric Machinery.” McGraw-Hill, Inc.1986.
[8] Marques J., Pinheiro H., et al., “A Survey On Variable-Speed Wind Turbine System”, Federal
University of Santa Maria.
[9] Ekanayake Janaka B., Lee Holdsworth, XueGuang Wu, and Nicholas Jenkins, “Dynamic
Modeling of Doubly Fed Induction Generator for Wind Turbines”, IEEE transactions on power
system, Vol. 18, No. 2, May 2003.
[10] Tang Y. and Xu L., “A Flexible active and reactive power control strategy for a variable speed
constant frequency generating system”, IEEE Transection on Power Electronics, vol. 10, no. 4,
pp. 472-478, July 1995.
[11] Tapia Arantxa, Tapia Gerardo, Ostolaza J. Xabier, and Saenz Jose Ramon, “Modeling and
Control of Wind Turbine Driven Doubly Fed Induction Generator”, IEEE Transactions on
Energy Conversion, Vol. 18(2), June 2003.
[12] Guo Yougui, Zeng Ping, and Blaabjerg F, “Modelling and Simulation of Generator Side
Converter of Doubly Fed Induction Generator-Based Wind Power Generation System”, ICEEE
International Conference, Wuhan, China, November 2010.
[13] Xu Lie, Cartwright P, “Direct Active and Reactive Power Contorl of DFIG for Wind Energy
Generation”, IEEE Transactions on Energy Conversion, Vol. 21(3), September 2006.
[14] Ekanayake Janka B, Lee Holdsworth, Xue Guang Wu, and Nicholas Jankins, “Dynamic
Modeling of Doubly Fed Induction Generator Wind Turbines”, IEEE Transactions on Power
Systems, Vol. 18, (2), May 2003.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 639-659

DIRECT TORQUE CONTROL SCHEME FOR THREE PHASE

INDUCTION MOTOR USING FUZZY SPEED REGULATOR

PRASAD R NAIK, PRAGATI G SHANBHAG, RAJALAKSHMI V

MENASINAKAI AND SUMATHI KULKARNI
Department of Electrical & Electronics Engineering,
B. V. B College of Engineering & Technology,
HUBLI- 580 031, Karnataka, India.

Abstract
Induction motors are currently used in many industrial applications. Thus, their control techniques
have received a lot of interest. An efficient method of induction motor control is the Direct Torque
Control (DTC) which maintains the motor developed torque at a specific command value providing a
better controllability. Generally, PI controllers are employed for speed regulation. PI speed regulator
require extensive tuning which otherwise will lead to poor quality of speed regulation. Due to this
drawback of PI speed regulator, the direct torque control method using fuzzy logic controller has been
implemented. Fuzzy logic (FL) has an ability to effectively employ modes of reasoning that are
approximate rather than exact. FL is one of the tools used to model a multi-input, multi-output system
by defining system-specific fuzzy rules. Fuzzy rule base for FL speed regulator is defined using
triangular membership functions. Also, the FL speed regulator implemented is of Mamdani type. In
this paper, the mathematical model of DTC of induction motor using fuzzy logic speed regulator has
been developed and simulated using MATLAB/Simulink environment. Development of mathematical
models includes developing mathematical models for various components of drives viz., three phase
induction motor, three phase voltage source inverter, fuzzy speed regulator and direct torque control
scheme. The model so developed is used to simulate the entire fuzzy logic based DTC system.
Simulation study has been carried out using MATLAB/Simulink environment. Simulation results are
presented and discussed for the step change in speed, torque and speed reversal. Simulation results

© http: //[Link]
640 PRASAD R NAIK, PRAGATI G SHANBHAG, RAJALAKSHMI V MENASINAKAI
AND SUMATHI KULKARNI
have shown good agreement with the philosophy of fuzzy logic based DTC in terms of fast dynamic
response.

-----------------------------------
Keywords: Direct Torque Control, Fuzzy speed regulator, Induction Motor, Switching state, Voltage
Source Inverter.

1. INTRODUCTION
In present day industrial applications, induction machines are widely used because of its
advantages over dc motors such as power/weight ratio, lower inertia, higher speeds and less
maintenance. Thus, there is a need of efficient control methods of induction machines. The
basic parameters to be controlled are mainly flux and torque by controlling the speed. Many
such techniques are introduced that involve speed control of induction machine, broadly
classified as scalar control and vector control.
 Scalar Control
Scalar control, as the name indicates is due to magnitude variation of the control variables
and disregards the coupling effect of the machine.
Different Scalar control methods
 Pole Changing
For a given frequency, the synchronous speed is inversely proportional to the number of
poles.
Therefore, motor speed can be changed by changing the number of poles. The provision
for changing the poles should be carried out during design/manufacturing stage only and
this adds to the additional cost.
 Stator voltage control
By reducing stator voltage, speed of a high slip induction motor can be reduced by an
amount which is sufficient for speed control of some fan and pump drives. While torque
is proportional to the square of the applied voltage and current is proportional to voltage,
therefore as voltage is reduced to reduce the speed, for the same current, motor develops
641

lower torque. This method is suitable for applications where torque demand reduces with
speed such as fan and pump loads.
 Supply frequency control
In this type of controllers, voltage induced in stator is proportional to the product of
supply frequency and air-gap flux. If the stator drop is neglected, terminal voltage can be
considered proportional to the product of frequency and flux. Any reduction in the supply
frequency, without a change in the terminal voltage, causes an increase in air-gap flux
and reduction in speed.
 Voltage Source Inverter (VSI) control
Voltage Source Inverter allows a variable frequency supply to be obtained from a dc
supply. VSI can be operated as a stepped wave inverter or a PWM inverter. When
operated as a stepped wave inverter, power semiconductor switches are switched in the
sequence with a time difference of T/6 seconds and each transistor is kept on for the
duration T/2 seconds, where T is the time period for one cycle. Frequency of inverter
operation is varied by varying T and the output voltage of the inverter is varied by
varying dc input voltage or modulation index.
 Current Source Inverter (CSI) control
For a given speed, torque is controlled by varying dc link current by changing the values
of voltage applied. Therefore, when supply is ac, a controlled rectifier is connected
between the supply and inverter. When supply is dc, a chopper is interposed between the
supply and inverter.
 Rotor resistance control
Maximum torque is independent of rotor resistance, speed at which the maximum torque
is produced changes with rotor resistance. For the same torque, speed falls with an
increase in rotor resistance. Advantage of rotor resistance control is that the motor torque
capability remains unaltered even at low speeds. Rotor resistance control method is
applicable only to wound rotor induction motor.
Today, the scalar control methods are confined to small-scale applications due to cost-
effectiveness. Even though scalar control methods are easy, they have inferior
performance which is not suitable for industrial applications where accurate control of
machine parameters are required. As a result, majority of the applications are
incorporated with vector control methods.
642 PRASAD R NAIK, PRAGATI G SHANBHAG, RAJALAKSHMI V MENASINAKAI
AND SUMATHI KULKARNI
 Vector Control
Vector Control method is the effective technique in controlling the machine parameters as
it takes into account the coupling effect in the machine along with the magnitude of
control variables. These methods are most needed for applications of ac drives as the
control is smoother as well as efficient.
Different Vector control methods
 Direct Torque Control
The Direct Torque Control (DTC) method is characterized by its simple implementation
and a fast dynamic response. Furthermore, the inverter is directly controlled by the
algorithm, i.e. a modulation technique for the inverter is not needed. However if the
control is implemented on a digital system, the actual values of flux and torque could
cross their boundaries too far, which is based on an independent hysteresis control of flux
and torque. The main advantages of DTC are absence of coordinate transformation and
absence of current regulators.
 Field Oriented Control
Field Oriented Control (FOC), is a variable frequency drive (VFD) control method which
controls three-phase ac electric motor output by means of two controllable VFD inverter
output variables:
1. Voltage magnitude
2. Frequency
The Field Orientated Control (FOC) was originally developed for high performance
applications. This enabled smoother operation over full speed range and can generate full
torque at zero speed. But, today this method is becoming increasingly attractive for low
performance application due to FOC’s motor size, cost and power consumption reduction
superiority.
 Direct Self Control
The Direct Self Control (DSC) method is intended mainly for high power Adjustable
Speed Drives (ASDs) with VSIs. Typically, slow switches, such as GTOs, are employed
in such inverters, and low switching frequencies are required. Therefore, in DSC drives,
the inverter is made to operate in a mode similar to the square-wave with occasional zero
states thrown in. The zero states disappear when the drive runs with the speed higher than
643

rated, that is, in the field weakening area, where, as inall other ASDs, the inverter
operates in the square wave mode.
Advanced control of electrical machines require an independent control of magnetic flux
and torque. For this reason, Field Oriented Control (FOC) was introduced which meant a
huge turn in the field of electric drives, since with its type of control the robust induction
machines could be controlled with a high performance. Later, another control method
called Direct Self Control was introduced. This method intended mainly for high power
ASDs with voltage source inverters. But, both these control methods employed more
number of machine parameters during the control. During 1980s a new control technique
known as Direct Torque Control (DTC) was introduced, which was characterized by its
simple implementation and faster dynamic response. Presently, this method is found to be
efficient and easy to implement as it requires lesser number of parameters to be
monitored when compared to other two methods.

2. DIRECT TORQUE CONTROL

Fig. 1. Block diagram of Direct Torque Control

In the mid-1980s, a vector control technique, known as direct torque and flux control was
introduced for voltage-fed PWM inverter drives. The scheme, as the name indicates, is the
direct control of the torque and stator flux of a drive by inverter voltage space vector
selection through a lookup table. Referring to Fig. 1, the stator voltage (_vs) and stator
644 PRASAD R NAIK, PRAGATI G SHANBHAG, RAJALAKSHMI V MENASINAKAI
AND SUMATHI KULKARNI
current (_is) are taken as feedback to obtain actual values of stator flux, motor developed
torque, flux angle and rotor speed. In order to increase the dynamic speed response of the
drive, the actual speed and the command speed are processed in speed controller. The speed
controller can be realized either using PI controller or fuzzy logic. From the speed controller,
torque is obtained and is considered as command torque. The obtained values of flux and
torque are then compared with their respective command values to compute errors. These
errors are processed through hysteresis controllers. The output of hysteresis controllers are
fed to ‘Switching Table ’block. In this block, the appropriate switching states for voltage
source inverter are generated based on output of hysteresis controllers and the estimated flux
angle. This results in maintaining the stator flux and torque within the constrained band.

3. MATHEMATICAL MODELLING
3.1 Induction Machine
The concept of vector control has opened up a new possibility that induction motors can be
controlled to achieve dynamic performance as good as that of dc motors. In order to
understand and analyse vector control, the dynamic model of the induction motor is
necessary. The following equation set describe the induction motor dynamic behaviour:
𝑑𝛹̅̅̅𝑠
𝑣̅𝑠 = 𝑅𝑆 𝑖̅𝑠 + (1)
𝑑𝑡
̅̅̅̅𝑟
𝑑𝛹
0 = 𝑅𝑟 𝑖̅𝑟 + - ̅̅̅
𝛹𝑟 𝜔𝑚 (2)
𝑑𝑡
̅̅̅
𝛹𝑠 = 𝐿𝑆 𝑖̅𝑠 + 𝐿𝑚 𝑖̅𝑟 (3)
̅̅̅
𝛹𝑟 = 𝐿𝑟 𝑖̅𝑟 + 𝐿𝑚 𝑖̅𝑠 (4)
Where,
𝑅𝑆 = stator resistance in Ω ,
𝑅𝑟 = rotor resistance in Ω,
𝐿𝑆 = stator inductance in H,
𝐿𝑟 = rotor inductance in H,
𝐿𝑚 = mutual inductance in H,
𝜔𝑚 = rotor speed in rad/sec,
̅̅̅
𝛹𝑠 = 𝛹𝑑𝑠 + 𝑗𝛹𝑞𝑠 = stator flux in wb,
̅̅̅
𝛹𝑟 = 𝛹𝑑𝑟 + 𝑗𝛹𝑞𝑟 = rotor flux in wb
645

3.2 Voltage Source Inverter

Fig. 2. Basic configuration of Voltage Source Inverter

From Fig. 2, pole voltage appearing on each leg of inverter, is given by:
𝑉𝑑𝑐
𝑉𝑎0 = (2𝑆1 -1) 2
(5)

𝑉𝑑𝑐
𝑉𝑏0 = (2𝑆2 -1) (6)
2

𝑉𝑑𝑐
𝑉𝑐0 = (2𝑆3 -1) 2
(7)

Phase Voltage of the inverter can be calculated as:

1
𝑉𝑎𝑛 = 3(2𝑉𝑎0-𝑉𝑏0-𝑉𝑐0) (8)

1
𝑉𝑏𝑛 = (2𝑉𝑏0-𝑉𝑐0-𝑉𝑎0) (9)
3

1
𝑉𝑐𝑛 = 3(2𝑉𝑐0-𝑉𝑎0-𝑉𝑏0) (10)

The three phase voltages are further converted to two phase using following equations:

𝑉𝑑𝑠 = 𝑉𝑎𝑛 (11)

1
𝑉𝑞𝑠 = (𝑉 − 𝑉𝑐𝑛 ) (12)
√3 𝑏𝑛

3.3 Flux and Torque Estimator, Sector Identification

The circular trajectory of the command flux vector Ψ𝑠 ∗ with the hysteresis band rotates in an
anti-clockwise direction, as shown in Fig. 3. The actual stator flux Ψ𝑠 is constrained within
the hysteresis band and it tracks the command flux in a zig-zag path. The flux and torque
646 PRASAD R NAIK, PRAGATI G SHANBHAG, RAJALAKSHMI V MENASINAKAI
AND SUMATHI KULKARNI
estimators are used to calculate actual values flux and torque. The VSI voltage and current
vectors transformed into the d-q stationary reference frame are fed as inputs to this block.
The d-q axes stator flux linkage is estimated by computing the integral of difference between
the respective d-q input voltage and the voltage drop across the stator resistance,
𝛹𝑑𝑠 = ʃ 𝑉𝑑𝑠 - 𝑖𝑑𝑠 𝑅𝑠 (13)

𝛹𝑞𝑠 = ʃ 𝑉𝑞𝑠 - 𝑖𝑞𝑠 𝑅𝑠 (14)

The resultant flux linkage can be computed as:

|𝛹𝑠 | = √(𝛹𝑑𝑠 2 + 𝛹𝑞𝑠 2 ) (15)

Fig. 3. Trajectory of Stator flux vector in DTC

The location of the stator flux linkage on the vector plane should be known so that
appropriate voltage vector is selected depending upon the flux location. The sector S(k) in
which stator flux lies can be calculated as:
𝛹
Ɵe = 𝑡𝑎𝑛−1 ( 𝛹𝑞𝑠 ) (16)
𝑑𝑠

The electromagnetic torque developed by motor can be expressed as:

3𝑃
Te = ( 𝛹𝑑𝑠 𝑖𝑞𝑠 − 𝛹𝑞𝑠 𝑖𝑑𝑠 )
4

(17)
647

3.4 Hysteresis Controller and Switching State Selector

Hysteresis Controller
The actual calculated stator flux and developed torque are compared with the command flux
and torque values. The difference between them is the error values. These errors are
processed through hysteresis controllers. The hysteresis controllers are as shown in Fig. 4.

Fig. 4. Hysteresis Controller (a) Flux Hysteresis (b) Torque Hysteresis

 Flux Hysteresis: It is a two level controller. It works as follows:

H = 1 for E > +HB
H = -1 for E < -HB
where, E is the error in stator flux.
 Torque Hysteresis: It is a three level controller. It works as follows:
HTe = 1 for ETe > +HBTe
HTe = -1 for ETe < 1HBTe
HTe = 0 for -HBTe < ETe < +HBTe
where, ETe is the error in the torque.
Switching State Selector
The magnitude of stator flux and the developed torque can be directly controlled by proper
selection of space vectors of stator voltage, that is, selection of consecutive inverter states.
Specifically:
 Nonzero voltage vectors whose misalignment with the stator flux vector does not
exceed +/-90degree cause the flux to increase.
 Nonzero voltage vectors whose misalignment with the stator flux vector exceeds +/-
90degree cause the flux to decrease.
648 PRASAD R NAIK, PRAGATI G SHANBHAG, RAJALAKSHMI V MENASINAKAI
AND SUMATHI KULKARNI
 Zero states, 0 and 7, (of reasonably short duration) practically do not affect the vector
of stator flux which, consequently, stops moving.
 The developed torque can be controlled by selecting such inverter states that the stator
flux vector is accelerated, stopped or decelerated.

Table 1 : Inverter State Selector

Table 2 : Voltage Vectors

Fig. 5. Flux and Torque Variations Due to Applied Voltage Vector

Fig. 5 shows the variations of flux and torque due to applied voltage vector. By this we can
distinguish five different cases, they are:
 Both the flux and torque are to be decreased.
 The flux is to be decreased but the torque is to be increased.
649

 The flux is to be increased but the torque is to be decreased.

 Both the flux and torque are to be increased.
 The torque error is within the tolerance range.
Referring to Table I, for example, if H = 1, HTe = 1 and considering that flux is in sector-2,
the voltage vector applied is V3. From Table II it can be inferred that, by applying V3, the
generated switching states (S1, S2, S3) of inverter will be (0, 1, 0). In Fig. 5, the direction
refers to increase/decrease in magnitude of stator flux and torque whereas the size of the
arrow implies the extent of change in the magnitude.
3.5 Fuzzy Logic Speed Regulator
Fuzzy logic has been widely used in drive control applications in the past two decades. This
technique has been emphasized because it has the capability to efficiently control the non-
linear systems and is well-suited for systems demanding superior response. The fuzzy PI
speed controller has almost the same operation principles with the classic PI controller. The
basic difference of the two controllers is that the output of the fuzzy PI controller gives the
change in the reference torque value dTe , which has to be summed or integrated, to give the
T_e value. The FL controller has two inputs, the speed error e = !_r � !r and the change in
the speed error ce , which is related to the derivative de=dt of error. For a specific input
condition of the variables, fuzzy controller computes the output signal. The fuzzy controller
is an input/output static non-linear system which maps the pair values of e and ce according
to fuzzy rules and gives the following form:
k1e + k2ce = dTe (18) where, k1 and k2 are non-linear gain factors. The analytical block
diagram of the fuzzy PI controller is shown in Fig. 6.
The input variables e and ce are expressed in per unit values. This is achieved by dividing
the variables by specific scale factors. The output will also be expressed in per unit values.
The output of the controller is the change in the reference torque value dTe . Further, in
order to obtain the command torque T_e , the output of fuzzy controller is integrated.
The fuzzy sets of linguistic expressions of the variables and the membership functions (MFs)
of these variables areshown in Fig. 7 (a) and (b). The output of the controller is the change in
the reference torque value dTe. The MFs of the output variable in per unit values are shown
in Fig. 7 (c). The definition of the MFs depends on the system behaviour. All the MFs are
asymmetrical because near the origin (steady state), the signals require more precision.
650 PRASAD R NAIK, PRAGATI G SHANBHAG, RAJALAKSHMI V MENASINAKAI
AND SUMATHI KULKARNI
The next step in the analysis of fuzzy speed controller is the definition of fuzzy rules. In
fuzzy rules for the speed controller, the input variables e and ce can be divided into seven
fuzzy sets as shown in Table III. The MFs of the output variable dTe(pu) are shown in the
body of the matrix. In 7 x 7 = 49 possible rules in the matrix, a rule is read as:

Fig. 6. Block diagram of fuzzy speed regulator

IF e(pu)=NS , AND ce(pu)=PM , THEN Te(pu)=PVS
651
Fig. 7. Membership functions of the input variables (a) speed error e(pu) (b) change in speed
error ce(pu) and of the output variable (c) change in reference torque value dTe of the fuzzy
speed regulator
Table 3 : Fuzzy Rules

where, PB=Positive Big, PM=Positive Medium, PS=Positive Small, PVS=Positive Very

Small, Z=Zero, NVS=Negative Very Small, NS=Negative Small, NM=Negative Medium,
NB=Negative Big

4. SIMULATION RESULTS AND DISCUSSION

Fig. 8 shows the mathematical model of three phase induction motor using Eqn. (1), (2), (3)
and (4). The model computes the speed, torque developed, three phase stator flux and stator
current of the motor.

Fig. 8. Three phase Induction Motor model

652 PRASAD R NAIK, PRAGATI G SHANBHAG, RAJALAKSHMI V MENASINAKAI
AND SUMATHI KULKARNI

Fig. 9. Simulink model of DTC using fuzzy speed regulator

Fig. 9 shows the model of DTC using fuzzy speed regulator developed in MATLAB /
Simulink.

Fig. 10. Simulink model of fuzzy speed regulator

Fig. 10 shows the model of fuzzy speed regulator developed using membership functions
shown in Fig. 7 and the fuzzy rule base shown in III.
653

Fig. 11. Speed response of inductiom motor

Fig. 11 shows the speed response of induction motor. Under no-load condition, for ac supply
of 60 Hz, the rated speed is 377 rad/sec. The result shows that induction motor has attained
the rated speed well within 0.5 seconds. After 2 seconds, when the motor is loaded with a
load torque of 5 N-m, the speed is slightly reduced.

Fig. 12. Torque response of inductiom motor

Fig. 12 shows torque response of induction machine. It can be observed that the starting
torque of induction machine is very high. When the machine is made to run on load torque
of 5 N-m after 2 seconds, the corresponding increase in torque is shown.
654 PRASAD R NAIK, PRAGATI G SHANBHAG, RAJALAKSHMI V MENASINAKAI
AND SUMATHI KULKARNI

Fig. 13. Stator current of induction motor

Fig. 13 shows the stator current of induction motor. High starting current of induction motor
can be inferred from the result.

Fig. 14. d-q axis stator flux

655

For the command flux of 0.6 wb, the resultant stator flux is as shown in Fig. 14.

Fig. 15. Speed Response (a) Command speed (b) DTC with PI speed regulator (c) DTC with
fuzzy speed regulator
656 PRASAD R NAIK, PRAGATI G SHANBHAG, RAJALAKSHMI V MENASINAKAI
AND SUMATHI KULKARNI
Fig. 15 shows the dynamic speed response when the step change in command speed is
applied to motor keeping load torque constant to 1 N-m. Initially the speed is set to 200
rad/sec and after 3 seconds, the speed is raised to 250 rad/sec as shown in Fig. 15 (a). The
corresponding change in speed response of DTC with PI speed regulator and DTC with
fuzzy speed regulator is shown in Fig. 15 (b) and (c) respectively.

Fig. 16. Torque response (a) command torque (b) DTC with PI speed regulator (c) DTC with
fuzzy speed regulator
657

Torque response of induction motor is shown in Fig. 16. Initially the load torque is 1 N-m
and after 3 seconds, the torque is changed to 4 N-m. The corresponding change in motor
developed torque is shown. The torque response with PI speed regulator is shown in 16 (b).
It is observed that there is large overshoot during transient period with the use of PI speed
regulator whereas the use of fuzzy logic speed regulator has improved the torque response as
seen in 16 (c). Hence, from the results it can be inferred that the system with fuzzy logic
gives relatively better response.

Fig. 17. speed response (a) command speed (b) DTC with PI speed regulator (c) DTC with fuzzy
speed regulator
658 PRASAD R NAIK, PRAGATI G SHANBHAG, RAJALAKSHMI V MENASINAKAI
AND SUMATHI KULKARNI

Fig. 17 (a) very well depicts the speed reversal condition wherein the speed is reduced from
200 rad/sec to -200 rad/sec after 3 seconds. Fig. 17 (b) and Fig. 17 (c) show the speed
response of machine using conventional PI speed regulator and fuzzy speed regulator. From
the result, it is clear that the fuzzy speed regulator has better and smoother response to
change in command speed when compared to PI speed regulator.

Fig. 18. Torque response due to speed reversal (a) DTC with PI speed regulator (b) DTC with
fuzzy speed regulator

Fig. 18 shows the torque response during speed reversal. From result it can be observed that
the use of PI speed regulators leads to slower dynamic response compared to fuzzy speed
regulator. For the time period between 3 to 4.2 seconds for DTC with PI speed regulator and
3 to 3.75 seconds for DTC with fuzzy speed regulator, the torque is negative indicating that
the induction motor is operates as induction generator during speed reversal.

V. CONCLUSION
The paper presents Direct Torque Control scheme of induction motor using fuzzy speed
regulator. The developed technique is intended for an efficient control of torque and flux
659

without changing the machine parameters. The DTC scheme presented in this paper mainly
involves fuzzy logic to design the speed regulator. Since, fuzzy logic is adaptive in nature, it
is well-suited for applications requiring faster dynamic response and smoother control. The
developed control strategy has been analysed by simulating the entire system in
MATLAB/Simulink environment. The simulation results thus obtained have shown that
direct torque control scheme implemented using fuzzy logic speed regulator has superior
response than conventional PI speed regulator under wide range of operating conditions such
as change in command speed, step change in load and speed reverse operation.

ACKNOWLEDGEMENT
The authors would like to thank Prof. Ashok Shettar, Principal and Prof. A. B. Raju, Head of
the Department, Electrical & Electronics Engineering, B. V.B. College of Engineering and
Technology, Hubli, for the necessary facility provided.
APPENDIX
Induction motor parameters used for simulation.
Rated supply voltage, Vs= 400 volts
Rated speed, N= 1370 rpm
Lm = 0.07 H
Lr = 0.072 H
Ls = 0.072H
Rr=0.816 ohm
Rs= 0.435 ohm

REFERENCES

[1] Bose Bimal K, Modern Power Electronics and AC Drives, Prentice Hall of India, 2005.
[2] Trzynadlowski Andrej M, Control of Induction Motors, Academic Press, 2001.
[3] Ranganathan V T, Course notes on Electric Drives, 2004.
[4] Ned Mohan, Power Electronics-A first course-3rd Edition, Jhon Wiley and sons.
[5] Takahashi I. and Noguchi T., A new quick-response and high-efficient control strategy of an
induction motor, IEEE Trans. Ind. Applicat., vol IA-22, pp. 820-827, Sept./Oct. 1986.
[6] Georgios Adamidis and Koutsogiannis Zisis Direct Torque Control Using Space Vector
Modulation and Dynamic Performance of the Drive, via a Fuzzy Logic Controller for Speed
Regulation, Democritus University of Thrace, Greece.
International J. of [Link] & Advcs. in Engg.(IJMRAE),
ISSN 0975-7074, Vol. 6, No. II (April 2014), pp. 661-672

DETECTION OF DYNAMIC BIFURCATIONS OF SSR

IN A SERIES COMPENSATED SYSTEM

SHRUTHI RAMACHANDRA1, R. C.MALA2

AND NAGESHPRABHU3
1
[Link], M.I.T,Manipal,
2
[Link], M.I.T,Manipal
3
Professor, NMAMIT, Nitte

Abstract
This paper presents the effect of a fast acting static exciter and a Power System Stabilizer (PSS) on
Sub synchronous resonance in a Single Machine connected to an Infinite Bus (SMIB) system using
Bifurcation theory. Analysis is carried out on a SMIB system modeled as per IEEE FBM data.
Dynamic bifurcations of SSR are detected for three cases and the effect of exciter and PSS on the
stability region of the system is investigated.

-------------------------------
Keywords: Bifurcation, Hopf bifurcation, FBM (First Benchmark Model), SSR (Sub-Synchronous
Resonance), SMIB (Single Machine Connected to Infinite Bus)

© http: //[Link]
662 SHRUTHI RAMACHANDRA, R. [Link] AND NAGESHPRABHU

1. INTRODUCTION
Fixed series capacitors have been used since a long time for increasing power transfer
capability and to improve voltage stability in long transmission lines. However, the use of
series compensation may lead to sustained oscillations at sub-synchronous frequencies
known as Sub-Synchronous Resonance (SSR) that can cause turbine-generator shaft failure
and electrical instabilities [1].
IEEE sub synchronous resonance working group defines SSR as “An electric power system
condition where the electric network exchanges significant energy with a turbine-generator
at one or more of the natural frequencies of the combined system below the synchronous
frequency of the system following disturbances”. [2]
There are different methods proposed to analyze the SSR phenomenon. The most common
methods are frequency scanning technique, eigenvalue analysis and Electromagnetic
Transient Program (EMTP). The first two methods are based on linearized system models.
EMTP uses a detailed three phase modeling of the system and numerically integrates the
non-linear system equations. Bifurcation analysis is a geometric method of analysis of
nonlinear system dynamics. Local bifurcation theory has been applied to provide an
explanation for various nonlinear behaviors in power system and power system instabilities
such as voltage collapse and low-frequency electro-mechanical oscillations that occur in
electric power networks. The presence of different types of bifurcation in power system
models has been revealed. The most commonly encountered bifurcation in these models is
the Hopf bifurcation [3].
In the SSR area, Zhu [4] and Zhu et al. [5] were the first to employ bifurcation methods to
demonstrate the existence of a Hopf bifurcation in a single-machine-infinite bus bar power
system.
The paper is structured as follows: Section 2 briefs the mathematical modelling of SMIB
system in the d-q frame based on IEEE FBM data, modelling of turbine-shaft system,
differential equations of the static exciter, PSS. Section 3 explains dynamic bifurcation
theory. Numerical simulations are shown in Section 4. The paper is concluded in Section 5.
DETECTION OF DYNAMIC BIFURCATIONS OF… 663

2. MATHEMATICAL MODELLING OF SMIB SYSTEM

2.1. Mathematical model of a synchronous machine
Synchronous machine model 2.2 is considered for the analysis. It has two damper windings
on the quadrature axis and one damper winding on the direct axis. The field winding is
aligned along the direct axis [6][7]. The equations describing the synchronous machine
model 2.2 in direct axis and quadrature axis frame are as follows
d-axis equations
dΨd
dt
= −ωΨq − R a id ωb − ωb Vd (1)
𝑑𝛹𝑓 ′
1 𝑥𝑑
𝑑𝑡
= 𝑇 ′ (−𝛹𝐹 + 𝛹𝑑 ) + ( ′ 𝐸𝑓𝑑 ) (2)
𝑑 (𝑥𝑑 −𝑥𝑑 )
𝑑𝛹ℎ 1
𝑑𝑡
= 𝑇 ′′ (−𝛹ℎ + 𝛹𝑑 ) (3)
𝑑
′ ′′ ′ ′′
𝑥𝑑 −𝑥𝑑 𝑥𝑑 −𝑥𝑑 𝑥𝑑
𝛹𝑑 = 𝑥𝑑′′ 𝑖𝑑 + ( ′
𝑥𝑑
) 𝛹ℎ +( 𝑥𝑑
) ( ′ ) 𝛹𝑓
𝑥𝑑
(4)

q-axis equations.
𝑑𝛹𝑔 1
= (−𝛹𝑔 + 𝛹𝑞 ) (5)
𝑑𝑡 𝑇𝑞′

𝑑𝛹𝑘 1
𝑑𝑡
= 𝑇 ′′ (−𝛹𝑘 + 𝛹𝑞 ) (6)
𝑞

𝑑𝛹𝑞
𝑑𝑡
= 𝜔𝛹𝑑 − 𝑅𝑎 𝑖𝑞 𝜔𝑏 − 𝜔𝑏 𝑉𝑞 (7)
𝑥𝑞′ −𝑥𝑞′′ 𝑥𝑞 −𝑥𝑞′ 𝑥 ′′
𝛹𝑞 = 𝑥𝑞′′ 𝑖𝑞 + ( ) 𝛹𝑘 + ( ) ( 𝑞′ ) 𝛹𝑔 (8)
𝑥𝑞′ 𝑥𝑞 𝑥𝑞

2.2. Mathematical model of the transmission line

Fig 1. Single machine connected to an infinite bus bar.

A single machine connected to an infinite bus is as shown in the Fig.1. The terminal voltage
of the generator isVg ∠θ. The voltage at the infinite bus is Eb ∠0. Rt and Rl are the
 664 SHRUTHI RAMACHANDRA, R. [Link] AND NAGESHPRABHU

transformer and line resistance respectively. Xt is the transformer reactance and Xl line is the
line reactance. Xc is the series compensation.
The mathematical expressions describing the transmission line in d-q frame are
−Rωb
−ω
d id Xe id ωb Vd Ed VCd
[ ]
dt iq
=[ −Rωb ] [iq ] Xe + [Vq ] − [ E ] − [V ] (9)
ω q Cq
Xe

2.3. Mathematical modeling of the Turbine-Shaft system

IEEE FBM model consists of a synchronous generator rotor as a six mass system. Fig. 2
shows the spring mass damper representation of two masses of a turbine-shaft system

Dhi

HP IP

Khi
Dh Di

Fig. 2. Spring mass damper representation of two masses of a turbine shaft system.
The mechanical equations for a single mechanical mass is written as follows
𝑑𝑠 1
( 𝑑𝑡ℎ ) = [𝑇 − 𝐷ℎ (𝑠ℎ − 𝑠ℎ0 ) − 𝐷ℎ𝑖 (𝑠ℎ − 𝑠𝑖 ) − 𝑇ℎ𝑖 ] (10)
2𝐻ℎ ℎ
𝑑𝑇ℎ𝑖
𝑑𝑡
= 𝜔𝑏 𝐾ℎ𝑖 (𝑠ℎ − 𝑠𝑖 ) (11)

The per unit slip of a mass is written as

ωn −ωb
sn = ωb
(12)

The state variable corresponding to the rotor angle of the generator rotor is given by :
dδg
dt
= ωb (sg − sg0 ) (13)

[6]
2.4. Static Excitation system
The differential equations for the static exciter is given by
dVr 1
= [−Vr + K a (Vref − Vt + Vs )] (14)
dt Ta

Where Vs is the PSS output.

DETECTION OF DYNAMIC BIFURCATIONS OF… 665

V′t 1
Vt
= 1+sT (15)
m

[6]
2.5. Power System Stabilizer (PSS)

𝛿𝑚 𝑠𝐾𝑠 𝑇𝑤 1 + 𝑠𝑇1 1 + 𝑠𝑇3 𝑉𝑠

1 + 𝑠𝑇𝑤 1 + 𝑠𝑇2 1 + 𝑠𝑇4

Fig. 3. Lead Lag Power System Stabilizer

Lead-lag PSS structure used for the analysis is as shown in Fig. [Link] first block is a
washout filter used to reduce over response of damping during severe events. The next two
blocks are the lead-lag compensator. The output signal of the PSS is given as an input to the
exciter.

3. DYNAMIC BIFURCATION
Electrical power systems are very complex systems that show rich nonlinear dynamic
phenomena both under normal and emergency operating conditions. Solutions ofdynamic
systems can be either constant or dynamic in nature.
Bifurcation theory indicates a qualitative change in the behavior of a system with variation in
one or more parameters. The value of the parameter at which the system exhibits qualitative
change is known as bifurcation point. Local bifurcation theory has been applied to provide
an explanation for various nonlinear behaviors in power system and power system
instabilities [8].
Mathematically, if we have the following system:
dx
= F(x, μ) (16)
dt

When a scalar control parameter μ is varied, a Hopf bifurcation of an equilibrium solution of

a system is said to occur at μ= μ0 if the following conditions are satisfied:
1. F (x0: μ) = 0
2. The system matrix has a pair of purely imaginary eigenvalues ±𝑗 ⍵ while all of its
other eigenvalues have negative real parts at (xo; μ0)
666 SHRUTHI RAMACHANDRA, R. [Link] AND NAGESHPRABHU

3. Forμ ≈ μ0 , let the analytical continuation of the pair imaginary eigenvalues be𝜆̂ ± 𝑗⍵.
𝑑𝜆
Then𝑑𝜇 ≠0 atμ=μ0. This condition implies a transversal or nonzero speed crossing of

the imaginary axis and hence is called the transversality condition.

Branches of fixed points and periodic solutions meet at a hopf bifurcation point. Hence, a
Hopf bifurcation is classified as a dynamic bifurcation.

4. ANALYSIS
The mathematical model of a Single Machine connected to an infinite bus is modeled in
Matlab/Simulink environment as per the IEEE FBM data. Newton Raphson iterative method
is used to obtain the initial conditions for the transient analysis of the system. The generator
output power Pgis considered to be 0.9pu, with constant input mechanical torque. Series
capacitance is considered as the bifurcation parameter. It is varied from 0% to 90%
compensation i.e., from 0 to 0.45 with Xl=0.5.
Case 1: IEEE FBM is modelled with constant excitation system. Series compensation is
varied from 0 to 90% i.e., from Xc=0 to 0.45. Fig.4 shows the variation of real parts of eigen
values corresponding to torsional frequencies with varying parameter [Link] simulation
yielded five Hopf bifurcation points at Xc=0.06916, Xc=0.1915, Xc=0.2533, Xc=0.2898 and
at Xc=0.3147. The system finally loses its stability at Xc=0.3147 as mode 1 continues to be
unstable from this point [Link].5 shows variation in rotor angle with the varying
parameter Xc.

Fig. 4 Variation in real parts of eigen values with varying Xc.

DETECTION OF DYNAMIC BIFURCATIONS OF… 667

Fig.5. Variation in rotor angle with varying Xc under constant excitation.

Case 2:IEEE FBM when modeled with a fast acting static excitation system, and series
compensation varied from 0 to 0.45 yielded fourHopf bifurcation [Link].6 shows the
Hopf bifurcationsoccurring at Xc=0.02349, Xc=0.1953, Xc=0.2369 and at Xc=[Link]
system loses its stability from Xc=0.2537 onwards. Fig.5 shows variation in rotor angle with
the varying parameter Xc in the presence of a fast acting static excitation system.

Fig. 6 Variation in real parts of eigen values with varying Xc

668 SHRUTHI RAMACHANDRA, R. [Link] AND NAGESHPRABHU

Fig 7. Variation in rotor angle with varying Xc in a SMIB with a static exciter
Case 3:IEEE FBM with a fast acting static excitation system and a PSS yielded five Hopf
bifurcation points at Xc=0.068, Xc=0.1952, Xc=0.2394, Xc=0.2956 and at Xc=[Link].8
shows the variation of real parts of eigen values corresponding to torsional frequencies with
Xc. The system loses its stability via Hopf bifurcation i.e.,Xc=0.3297. Fig 9 shows variation
of rotor angle with the bifurcation parameter Xc. Fig 10shows the plot of rotor angle del v/s
time when Xc=0.3297 andFig. 11 shows the FFT analysis of rotor angle [Link] FFT
analysis indicates only one harmonic frequency which is mode 1 frequency. Fig. 12 shows
the plot of frequency v/s time for the same value of Xc. From Fig 10 and Fig 12, it can be
seen thatrotor angle as well as rotor speed exhibit sustained oscillations indicating the
presence of limit cycle.

Fig. 8 Variation in real parts of eigen values with varying Xc

DETECTION OF DYNAMIC BIFURCATIONS OF… 669

Fig 9. Variation in rotor angle with varying Xc (with a static excitation system and PSS).

Fig 10. Rotor angle v/s time at Xc=0.3297

Fig.11 FFT analysis of del

670 SHRUTHI RAMACHANDRA, R. [Link] AND NAGESHPRABHU

Fig 12. Frequency V/s time at Xc=0.3297

CONCLUSION
Detection of Dynamic bifurcation is done for a Single machine connected to infinite bus
system which is modeled as per IEEE FBM data. Under constant excitation, five Hopf
bifurcations were detected and the system finally loses stability via Hopf bifurcation
H5=0.3147.
The addition of static excitation system shifts the location of first Hopf bifurcation from
0.06916 to 0.02349. Four Hopf bifurcations were detected. The system loses its stability via
H4=0.2573 with a shrink in the stability region
The addition of Power System Stabilizer output to the static excitation system shifts the first
Hopf bifurcation to 0.068. The system loses its stability via H5=0.3297 with an expansion in
the stability region.

APPENDIX I
The data for the SMIB system considered is as per IEEE FBM based on 892.4 MVA and 500
KV base values.
Synchronous Machine Parameter
Open Circuit
′ " ′ "
𝑇𝑑0 𝑇𝑑0 𝑇𝑞0 𝑇𝑞0
DETECTION OF DYNAMIC BIFURCATIONS OF… 671

Time Constant

Value 4.3 0.032 0.850 0.050

Short Circuit
Time Constant 𝑇𝑑′ 𝑇𝑑" 𝑇𝑞′ 𝑇𝑞"
Value 0.4 0.0259 0.1073 0.0463

Network impedances in per unit based on the generator MVA

Parameter R Xt Xl XSYS Xc
value 0.02 0.14 0.50 0.06 0.35

Shaft inertias and spring constants for the FBM in per unit on the machine base

Inertia HP turbine IP turbine LPA LPB Generator Exciter

turbine turbine
Inertia 0.092897 0.155589 0.858670 0.884215 0.868495 0.0342165
Constant
(H)

Spring constantsfor the FBM in per unit on the machine base

Shaft Section HP - IP IP - LPA LPA - LPB LPB - GEN GEN - EXE

Spring
Constant 19.303 34.929 52.038 70.858 2.82
K in p.u
T /rad

Exciter: Ka=200; Ta=0.05;Tm=0.002

PSS parameters: Ks= 6; Tw=10; T1=0.6; T2=0.3; T3=0.9; T4=1
672 SHRUTHI RAMACHANDRA, R. [Link] AND NAGESHPRABHU

REFERENCES

[1] IEEE committee report, “Reader’s Guide to Subsynchronous Resonance”, in IEEE Transactions
on Power Systems. Vol. 7, No. 1, pp.150-158, February 1992.
[2] IEEE Subsynchronous Resonance Task Force of the Dynamic System Performance Working
Group Power System Engineering Committee, “First Benchmark Model for Computer
Simulation of Subsynchronous Resonance”, IEEE Transactions on Power Apparatus and
Systems, Vol. PAS-96. No.5. September/October 1977.
[3] Harb A. H. “Application of Bifurcation Theory to Subsynchronous Resonance in Power
Systems”, Ph.D Thesis, Virginia Polytechnic Institute and State University, December 1996.
[4] Zhu W., “Analysis of Subsynchronous Resonance in Power Systems”, Ph.D Thesis, Oregon
State University, June, 1994.
[5] Zhu W., Mohler R. R., Spee R., Mittelstadt W. A., and Maratukulam D., “Hopf Bifurcations in a
SMIB Power System with SSR”, IEEE Proceeding of the 1995 Summer Meeting, pp. 531-534,
PWRS.
[6] Pai M. A., Sen Gupta D. P., Padiyar K. R., “Small Signal Anaylysis of Power System”, Narosa
Publishing House, ISBN-978-81-7319-5945.
[7] “Power System Dynamics and Control”, [Link]
[8] Nayfeh A.H. and Balakumar Balachandran, “Applied Nonlinear Dynamics”, WILEY-VCH
Verlag GmbH & Co. KGaA, Weinheim, 2004.
 INTERNATIONAL JOURNAL OF MULTIDISCIPLINARY
RESEARCH AND ADVANCES IN ENGINEERING
(IJMRAE-Submission of Papers)
The Journal of Multidisciplinary Research and Advances in Engineering is an international journal
published in India quarterly (January, April, July & October); it aims to publish original research
papers devoted to all areas of various disciplines of engineering and their real life and industrial
applications. This Journal is basically devoted to multidisciplinary research in Engineering and
Technology, which can improve the present methods and technology being used in Industry.

International Journal of Multidisciplinary Research & Advances in Engineering (ISSN 0975-7074) is a

referred reviewed journal abstracted & indexed by Google and Google Scholar, CAS, a division of the
American Chemical Society (USA),EBSECO Publishin USA, Open Directory J., Russia, Open J-Gate
(New [Link]), ULRICH’s Periodical’s Directory (USA),Universe Digital Library etc.

This journal also welcomes review papers, research communication, research articles, innovative
ideas, real life applications that stimulate research engineering sciences, basic engineering, ,
physical sciences and allied areas.

Manuscript written in English on one side of the paper in 1.5 line spacing by MS-Word (with
text width 12.5 cm. and text height 19 cm.). Each page should be numbered, the first page of
manuscripts should contain (i) the title; (ii) the author’s name(s) and full mailing address with e-
mail; (iii) An abstract of not less than 150 words and not exceeding 300 words; (iv) Key-words

Authors are requested to submit manuscripts electronically at the e-mail address

smkhairnar2007@[Link] & anshirsath2010@[Link]

Manuscripts should not have been published or submitted elsewhere. An effort is made to publish
a paper duly recommended by a referee within a period of maximum three months.

Papers should be arranged in sections as follows: Title, Abstract, Key-words, Authors affiliation,
Introduction, Methods, Results, Discussion, Figures, Tables, and Legends for illustrations,
Acknowledgement (if any), References. Abbreviations should be spelt out when first used in the
text. The international system of units and standard abbreviations should be used.

Authors are requested to arrange printing charges of their papers @ US$ 15.00 per page from
their institutions or research projects. For authors in India this charge is Rs.300 per page. Ten
reprints of a paper and e-copy of the journal are provided to the authors ex-gratis. The hard copy
of the journal may be ordered prior to publication and will be charged extra at cost of Rs.1000 or
US$ 50.

The papers intended for publication in IJMRAE be submitted in soft / hard copy to:

Dr. S. M. Khairnar Website: [Link]

Principal Editor (IJMRAE), Email: smkhairnar2007@[Link]
A-63, Puru Society, Next to Sanjay Park, Tel.:0091-20-26610466
Air Port Road, Pune-411032, Mobile: 09850330076
Maharashtra State, INDIA.