Lecture Notes in Electrical Engineering 667

Lillie Dewan
Ramesh C. Bansal
Ujjwal Kumar Kalla   Editors

Advances in
Renewable
Energy and
Sustainable
Environment
Select Proceedings of NCRESE 2019
Lecture Notes in Electrical Engineering

Volume 667

Series Editors

Leopoldo Angrisani, Department of Electrical and Information Technologies Engineering, University of Napoli
Federico II, Naples, Italy
Marco Arteaga, Departament de Control y Robótica, Universidad Nacional Autónoma de México, Coyoacán,
Mexico
Bijaya Ketan Panigrahi, Electrical Engineering, Indian Institute of Technology Delhi, New Delhi, Delhi, India
Samarjit Chakraborty, Fakultät für Elektrotechnik und Informationstechnik, TU München, Munich, Germany
Jiming Chen, Zhejiang University, Hangzhou, Zhejiang, China
Shanben Chen, Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai, China
Tan Kay Chen, Department of Electrical and Computer Engineering, National University of Singapore,
Singapore, Singapore
Rüdiger Dillmann, Humanoids and Intelligent Systems Laboratory, Karlsruhe Institute for Technology,
Karlsruhe, Germany
Haibin Duan, Beijing University of Aeronautics and Astronautics, Beijing, China
Gianluigi Ferrari, Università di Parma, Parma, Italy
Manuel Ferre, Centre for Automation and Robotics CAR (UPM-CSIC), Universidad Politécnica de Madrid,
Madrid, Spain
Sandra Hirche, Department of Electrical Engineering and Information Science, Technische Universität
München, Munich, Germany
Faryar Jabbari, Department of Mechanical and Aerospace Engineering, University of California, Irvine,
CA, USA
Limin Jia, State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing, China
Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences, Warsaw, Poland
Alaa Khamis, German University in Egypt El Tagamoa El Khames, New Cairo City, Egypt
Torsten Kroeger, Stanford University, Stanford, CA, USA
Qilian Liang, Department of Electrical Engineering, University of Texas at Arlington, Arlington, TX, USA
Ferran Martín, Departament d’Enginyeria Electrònica, Universitat Autònoma de Barcelona, Bellaterra,
Barcelona, Spain
Tan Cher Ming, College of Engineering, Nanyang Technological University, Singapore, Singapore
Wolfgang Minker, Institute of Information Technology, University of Ulm, Ulm, Germany
Pradeep Misra, Department of Electrical Engineering, Wright State University, Dayton, OH, USA
Sebastian Möller, Quality and Usability Laboratory, TU Berlin, Berlin, Germany
Subhas Mukhopadhyay, School of Engineering & Advanced Technology, Massey University,
Palmerston North, Manawatu-Wanganui, New Zealand
Cun-Zheng Ning, Electrical Engineering, Arizona State University, Tempe, AZ, USA
Toyoaki Nishida, Graduate School of Informatics, Kyoto University, Kyoto, Japan
Federica Pascucci, Dipartimento di Ingegneria, Università degli Studi “Roma Tre”, Rome, Italy
Yong Qin, State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing, China
Gan Woon Seng, School of Electrical & Electronic Engineering, Nanyang Technological University,
Singapore, Singapore
Joachim Speidel, Institute of Telecommunications, Universität Stuttgart, Stuttgart, Germany
Germano Veiga, Campus da FEUP, INESC Porto, Porto, Portugal
Haitao Wu, Academy of Opto-electronics, Chinese Academy of Sciences, Beijing, China
Junjie James Zhang, Charlotte, NC, USA
The book series Lecture Notes in Electrical Engineering (LNEE) publishes the
latest developments in Electrical Engineering - quickly, informally and in high
quality. While original research reported in proceedings and monographs has
traditionally formed the core of LNEE, we also encourage authors to submit books
devoted to supporting student education and professional training in the various
fields and applications areas of electrical engineering. The series cover classical and
emerging topics concerning:
• Communication Engineering, Information Theory and Networks
• Electronics Engineering and Microelectronics
• Signal, Image and Speech Processing
• Wireless and Mobile Communication
• Circuits and Systems
• Energy Systems, Power Electronics and Electrical Machines
• Electro-optical Engineering
• Instrumentation Engineering
• Avionics Engineering
• Control Systems
• Internet-of-Things and Cybersecurity
• Biomedical Devices, MEMS and NEMS
For general information about this book series, comments or suggestions, please
contact [email protected].
To submit a proposal or request further information, please contact the
Publishing Editor in your country:
China
Jasmine Dou, Associate Editor ([email protected])
India, Japan, Rest of Asia
Swati Meherishi, Executive Editor ([email protected])
Southeast Asia, Australia, New Zealand
Ramesh Nath Premnath, Editor ([email protected])
USA, Canada:
Michael Luby, Senior Editor ([email protected])
All other Countries:
Leontina Di Cecco, Senior Editor ([email protected])
** Indexing: The books of this series are submitted to ISI Proceedings,
EI-Compendex, SCOPUS, MetaPress, Web of Science and Springerlink **

More information about this series at http://www.springer.com/series/7818

Lillie Dewan Ramesh C. Bansal
• •

Ujjwal Kumar Kalla

Editors

Advances in Renewable
Energy and Sustainable
Environment
Select Proceedings of NCRESE 2019

123
Editors
Lillie Dewan Ramesh C. Bansal
School of Renewable Energy and Efficiency Department of Electrical and Computer
National Institute of Technology Engineering
Kurukshetra University of Sharjah
Kurukshetra, India Sharjah, United Arab Emirates

Ujjwal Kumar Kalla

Department of Electrical Engineering
Government Engineering College Bikaner
Bikaner, Rajasthan, India

ISSN 1876-1100 ISSN 1876-1119 (electronic)

Lecture Notes in Electrical Engineering
ISBN 978-981-15-5312-7 ISBN 978-981-15-5313-4 (eBook)
https://doi.org/10.1007/978-981-15-5313-4
© Springer Nature Singapore Pte Ltd. 2021
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, expressed or implied, with respect to the material contained
herein or for any errors or omissions that may have been made. The publisher remains neutral with regard
to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Committees

Patrons

Dr. Renu Phulia, IAS (Director, DNRE and HAREDA)

Dr. Satish Kumar (Director, NIT Kurukshetra)

Co-patrons

Prof. Lillie Dewan (Coordinator, SREE and Professor, EED, NIT Kurukshetra)
Sh. O. D. Sharma (Project Director, DNRE/HAREDA)
Sh. P. K. Nautiyal (Scientific Engineer A, DNRE/HAREDA)

Conveners

Dr. Shelly Vadhera (Associate Professor, EED, NIT Kurukshetra)

Mr. Sukhchain Singh (Project Manager, DNRE/HAREDA)

Organizing Secretary

Dr. Gulshan Sachdeva (Assistant Professor, MED, NIT Kurukshetra)

Dr. Avadhesh Yadav (Assistant Professor, MED, NIT Kurukshetra)
Dr. Shashi Bhushan Singh (Assistant Professor, EED, NIT Kurukshetra)
Dr. Rahul Sharma (Assistant Professor, EED, NIT Kurukshetra)

v
vi Committees

Organizing Committee

Dr. Amit Kumar (Faculty, SREE, NIT Kurukshetra)

Mr. Gaurav Sharma (Faculty, SREE, NIT Kurukshetra)
Ms. Amandeep Kaur (Faculty, SREE, NIT Kurukshetra)
Preface

National Conference on Renewable Energy and Sustainable Environment

(NCRESE-2019) was organized by School of Renewable Energy and Efficiency,
NIT Kurukshetra, in association with Bureau of Energy Efficiency, Government of
India (GOI); Ministry of Power, Department of New and Renewable Energy,
Haryana; and Haryana Renewable Energy Development Agency, Panchkula, during
August 30–31, 2019. The aim of this conference was to bring together leading
academic scientists, researchers, scholars and students across the globe to exchange
and share their experiences and research results and development about all new
technologies in the field of renewable energy to help the environment professionals
to harness the full potential. The whole idea of the conference is to exchange
thoughts and ideas in the concerned fields and transform those in real time to solve
the problems. The conference was a rare opportunity for all individuals of the
environment community to know about the latest technologies and strategies. The
research papers for the conference were invited based on original work in the broad
areas of renewable energy and harvesting technologies, alternate energy resources,
wind energy, solar energy, smart grids, electric vehicle, energy storage, solar
thermal and bio-energy technologies. All of these submissions went through a
rigorous peer-review process commensurate with their tracks. Together, the papers
presented here represent a set of high-quality contributions to the literature on agile
research and experience addressing a wide range of contemporary topics. In all, the
accepted 43 research papers were reviewed by three members of the Program
Committee and presented in front of committee of three members in their respective
tracks. Materials from all of the sessions are available on the conference Web site at
www.ncrese2019.com. NCRESE-2019 attendees were also treated to a number of
high-profile keynote speakers.

vii
viii Preface

It was a mammoth effort to review these and bring them together into a coherent
program. We would like to thank everyone who contributed to this effort including
paper authors, session presenters, track chairs, Program Committee members,
volunteers, sponsors and keynote speakers. Without their support, the event would
not have been as successful.

Kurukshetra, India Lillie Dewan

Sharjah, United Arab Emirates Ramesh C. Bansal
Bikaner, India Ujjwal Kumar Kalla
August 2019
Contents

Voltage Stability and Performance Analysis of the Multi Machine

System Using UPFC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Akshay Kumar Dwivedi and Shelly Vadhera
Harmonic Analysis at Variable Load . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Vijit Srivastava and P. N. Gupta
Voltage and Transient Stability Enhancement in Power System
Using Unified Power Flow Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Jaswant Singh Bhati and Shelly Vadhera
Probabilistic Approach for Load Flow Analysis of Radial
Distribution System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Anuj Banshwar, Mohit Pathak, Bharat Bhushan Sharma,
Naveen Kumar Sharma, and Sumit Kumar
Harmonic Issues in Non-conventional Energy Supply
and Its Remedy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Vijit Srivastava and P. N. Gupta
Artificial Neural Network-Based Source Identification Producing
Harmonic Pollution in the Electric Network . . . . . . . . . . . . . . . . . . . . . 49
Ankit Tayal, Lillie Dewan, and J. S. Lather
Impacts of Distributed Generation on Distribution System
Based on the Backward and Forward Sweep Method . . . . . . . . . . . . . . 59
Indubhushan Kumar and Sandeep Gupta
Cost Analysis of Multi-rotor Wind Turbines . . . . . . . . . . . . . . . . . . . . . 69
Navjot Singh Sandhu and Saurabh Chanana
Annualized Cost of Wind Power Generation . . . . . . . . . . . . . . . . . . . . . 77
Sahil Bajaj and K. S. Sandhu

ix
x Contents

A New Technique for Wind Turbine Power Curve Incomplete

Bin Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Bhukya Ramdas, M. Saravanan, J. C. David Solomon, and K. Balaraman
Grid Integration of Wind Energy Conversion System with Variable
Speed Analysis Using Simulink Model . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Ravneet Kaur, Deepika Bhalla, and Naveen Kumar Sharma
A Comparison of Performance of DPC Technique for DFIG
with Using PI and PID Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Sachhidanand Veer Savarkar and Shashi Bhushan Singh
A Case Study on 24-h Simulation of V2G System . . . . . . . . . . . . . . . . . 131
Robin Chola and Shashi Bhushan Singh
A Review on Bidirectional DC-DC Converters for V2G
and G2V Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Robin Chola and Shashi Bhushan Singh
Grid Integrated with Wind Turbine System on Fault
Analysis Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Ravneet Kaur, Deepika Bhalla, and Naveen Kumar Sharma
Performance Improvement of Wind Turbine Induction Generator
Using Neural Network Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
Sanjay Dewangan and Shelly Vadhera
Solar Resource Assessment and Potential in Indian Context . . . . . . . . . 173
Vinod Nandal, Raj Kumar, and S. K. Singh
Implementation of Hill Climb Search Algorithm Based Maximum
Power Point Tracking in Wind Energy Conversion Systems . . . . . . . . . 191
Biaklian Tonsing, Shelly Vadhera, and Atma Ram Gupta
Analytical Review of Solar Cell as Globalized Approach . . . . . . . . . . . . 201
Srishtee Chaudhary and Rajesh Mehra
Active Battery-Balancing Using Ćuk Converter . . . . . . . . . . . . . . . . . . . 211
Nikhil Sarode and M. Divya
Analysis of Series-Connected PV Cells Using gEDA and ngSPICE . . . . 221
Ritesh Kumar and Shelly Vadhera
An Experimental Study on a Portable SPV-Integrated Forced
Convective Solar Dryer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
Debajit Misra
A Review on Energy-Efficient and Sustainable Urban Buildings . . . . . . 245
Bishnu Kant Shukla, Harun Or Rashid, Anit Raj Bhowmik,
and Pushpendra Kumar Sharma
Contents xi

Role of Optimization Algorithms in Enhancing the Performance

of Photovoltaic Thermal (PV/T) Systems . . . . . . . . . . . . . . . . . . . . . . . . 253
Sumit Kumar, Sourav Diwania, Anwar S. Siddiqui, and Sanjay Agrawal
Selection of the Best Material for Coating on Solar Cell
and Optical Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
Sanjay Kumar and Lillie Dewan
Microcontroller-based PID Temperature Controller Using Genetic
Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Bharat Bhushan Sharma, Anuj Banshwar, Mohit Pathak,
Naveen Kumar Sharma, and Aman Joshi
Optimal Energy Dispatch and Techno-economic Analysis of Islanded
Microgrid with Hybrid Energy Sources . . . . . . . . . . . . . . . . . . . . . . . . . 285
V. V. S. N. Murty and Ashwani Kumar
Performance Analysis of Carrier Modulation Based H-Bridge Inverter
for Hybrid Microgrid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
Mukh Raj Yadav and Navdeep Singh
Optimal Energy Management in Hybrid Microgrid with Battery
Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
Masood Rizvi, Bhanu Pratap, and Shashi Bhushan Singh
Optimal Placement of DG with Battery Energy Storage Using CPLS
and Combined Dispatch Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
Bharat Singh and Ashwani Kumar Sharma
Comparison Among Conventional and Adaptive Gain Scheduling PID
Controllers for Stabilizing an Inverted Pendulum System . . . . . . . . . . . 327
Vikram Chopra, Sunil K. Singla, and Lillie Dewan
A Review on Protection Devices and Protection Techniques
for DC Micro-grid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
Bholeshwar and R. S. Bhatia
Positive and Negative Vibe Classifier by Converting Two-Dimensional
İmage Space into One-Dimensional Audio Space Using Statistical
Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
Kushal Sharma and Sandeep Gupta
Coupled Field Method for Analyzing Short-Circuit Electromagnetic
Forces in Power Transformers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357
Karan Bali and Deepika Bhalla
Thermal Separation in 2D Vortex Tube for a Different Tube Length
and Cold Mass Flow Ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367
Ashish Kumar Gupt, Deepak Kumar, and M. K. Paswan
xii Contents

GA and PSO Based Optimization for Benchmark Thermal System

Using Wavelet-Based MRPID Controller . . . . . . . . . . . . . . . . . . . . . . . . 385
Abhas Kanungo, Monika Mittal, and Lillie Dewan
The Stablization of Soil Using Treated Bagasse Ash . . . . . . . . . . . . . . . 397
Manish Chaudhary, Mohit Rathore, Manoj Bhatt, and Deepak Verma
Status of Water Quality in Various Ponds and Lakes in India . . . . . . . 417
Narender Singh and S. K. Patidar
Water Quality Assessment of Brahma Sarover, Sannihit Sarover
and Saraswati Tirth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427
Narender Singh and S. K. Patidar
About the Editors

Dr. Lillie Dewan is currently a professor at the Department of Electrical

Engineering, National Institute of Technology Kurukshetra, Haryana, India. She
obtained her Bachelor’s in Electrical Engineering from Punjab University in 1983,
Master’s in Control System in 1987, and Ph.D. in Electrical Engineering from
Kurukshetra University in 2001. She has more than 35 years of teaching and
research experience. She has published papers in 61 journals and 74 conferences.
She has also supervised 14 doctoral and 40 postgraduate students, and she has
completed two MHRD sponsored projects. She is a life member of ISTE, SSI, and
member of IEEE.

Prof. Ramesh C. Bansal has more than 25 years of diversified experience of

research, scholarship of teaching and learning, accreditation, industrial, and aca-
demic leadership in several countries. Currently, he is a Professor in the Department
of Electrical and Computer Engineering at University of Sharjah. Previously he was
Professor and Group Head (Power) in the ECE Department at University of Pretoria
(UP), South Africa. Prior to his appointment at UP, he was employed by the
University of Queensland, Australia; University of the South Pacific, Fiji; BITS
Pilani, India; and Civil Construction Wing, All India Radio.
Prof. Bansal has significant experience of collaborating with industry and gov-
ernment organizations. He has made a significant contribution to the development
and delivery of BS and ME programmes for Utilities. He has extensive experience
in the design and delivery of CPD programmes for professional engineers. He has
carried out research and consultancy and attracted significant funding from Industry
and Government Organisations.
Prof. Bansal has published over 300 journal articles, presented papers at con-
ferences, books, and chapters in books. He has Google citations of over 10000 and
an h-index of 44. He has supervised 22 PhD, 4 Post Docs. His diversified research
interests are in the areas of Renewable Energy (Wind, PV, DG, Micro Grid) and
Smart Grid. Professor Bansal is an editor of several highly-regarded journals,

xiii
xiv About the Editors

IET-RPG and IEEE Systems Journals. He is a Fellow and Chartered Engineer

IET-UK, Fellow Engineers Australia, Fellow Institution of Engineers (India), and
Senior Member of IEEE-USA.

Dr. Ujjwal Kumar Kalla is currently an assistant professor and head, department
of electrical engineering, Government Engineering College Bikaner, Rajasthan,
India. He obtained his Master’s of Technology in Power Electronics, Machines and
Drives in 2010 and Ph.D. in Electrical Engineering in Power Electronics in 2015
from Indian Institute of Technology Delhi, India. He has more than 16 years of
teaching and research experience. He has published 35 research articles in reputed
journals and conferences. He is a senior member of IEEE, USA, fellow of
Institution of Engineers India (IEI), life member of ISTE and fellow of Institution of
Electronics & Telecommunication Engineers.
Voltage Stability and Performance
Analysis of the Multi Machine System
Using UPFC

Akshay Kumar Dwivedi and Shelly Vadhera

Abstract The multimachine system is one of the most common systems present
in the power system network. System expansion is necessary due to ever-increasing
load, and thus, the analysis of the response of multimachine becomes important too.
UPFC is the most common FACTS device used in power system in current scenario
as it can control all parameters of the power at same time. IEEE 14 bus test system
with three generators has been taken for the analysis, and also voltage stability is
calculated with the help of different indices to find out suitable position of UPFC to
be implemented in the system. All the simulation is carried out in the PSAT software
[1]. The loss with and without UPFC is calculated.

Keywords Voltage stability · Multimachine · FACTS · UPFC · PSAT

1 Introduction

UPFC can be considered as the most upgraded version of the FACTS devices. It is
also one of the most important FACTS devices as it can control all the parameters of
the network simultaneously (Fig. 1).
UPFC incorporates two VSCs, operated with DC storage capacitance also so
called DC link. These arrangements operate as a perfect AC-to-AC converter within
which the active power flows independently in any direction between the two
converters, and each converter is capable of generating reactive power on its own
terminal independently. Obviously, there can be no reactive power flow through the
UPFC DC link. UPFC is so called the addition of SSSC and STATCOM mutually
joined with common DC voltage link[2].

A. K. Dwivedi (B) · S. Vadhera

Department of Electrical Engineering, National Institute of Technology Kurukshetra,
Kurukshetra, Haryana 136119, India
e-mail: [email protected]
S. Vadhera
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 1

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_1
2 A. K. Dwivedi and S. Vadhera

Fig. 1 Basic UPFC scheme

The above diagram explains the typical concept of useful power flow control
functions and series voltage injection. The introduction of the typical voltage phasor
V pq to the pre-existing bus voltage, V 0 , at an angle that can be changed from 0°
to 360° is illustrated in Fig. 2a. Effects on voltage regulation when V pq = V 0 ,
induced in phase with V 0 , is given in Fig. 2b. A summation of series compensation
and voltage regulation is represented in Fig. 2c, and V pq represents the addition of a
voltage controlling component V 0 , and the line current leads the voltage component

Fig. 2 UPFC with different modes of operation

Voltage Stability and Performance Analysis … 3

V c that provides the series compensation by 90°. During the phase-shifting process
represented in Fig. 2d, the voltage V pq generated by the UPFC is the addition of phase-
shifting voltage component V a and voltage-regulating element V 0 . According to
the need of the system, UPFC controller can select one, two or all three variables as
its control mechanism.

2 Modelling of UPFC

To perform the steady-state analysis at fundamental frequency, the UPFC is repre-

sented by an equivalent circuit consisting of two voltage sources. The equivalent
circuit of UPFC is shown in Fig. 3.
The fundamental Fourier series component of switched voltage waveform is
represented by the synchronous voltage sources at AC converter terminals of the
UPFC.
The UPFC voltage sources are:

E vr = Vvr (cos δvr + j sin δvr ) (1)

E cr = Vcr (cos δcr + j sin δcr ) (2)

 
where Vvr and δvr represent the controllable magnitude Vvr ,min ≤ Vvr ≤ Vvr ,max
and phase angle (0 ≤ δvr ≤ 2π ) of voltage source that works as shunt converter,
respectively. The magnitude of voltage Vcr and phase angle δcr having the limits

Fig. 3 Circuit diagram of UPFC

4 A. K. Dwivedi and S. Vadhera
 
Vcr ,min ≤ Vcr ≤ Vcr ,max and (0 ≤ δcr ≤ 2π ), respectively, represents the series
converter variable. Power flow is determined by the phase angle of series-injected
voltage. If both δcr and θcr (nodal voltage angle) are in phase, then UPFC regulates the
terminal voltage, whereas when δcr becomes quadrature w.r.t. θm , it acts like a phase
shifter and regulates the reactive power flow [3]. In case δcr becomes quadrature with
line current, then it behaves like a variable series compensator and controls the flow
of real power at any value of δcr . UPFC can work as variable series compensator,
voltage regulator and phase shifter simultaneously. Power flow amount is controlled
by determining the magnitude of series-injected voltage [4, 5].
On the basis of equivalent circuit diagram Fig. 3 and Eqs. (3) and (4), the real and
reactive powers at bus k are formulated as [3]:

Pk = Vk2 G kk + Vk Vm [G km cos(θk − θm ) + Bkm sin(θk − θm )]

+ Vk Vcr [G km cos(θk − δcr ) + Bkm sin(θk − θcr )]
+ Vk Vvr [G vr cos(θk − δvr ) + Bvr sin(θk − θvr )] (3)

Q k = −Vk2 Bkk + Vk Vm [G km sin(θk − θm ) − Bkm cos(θk − θm )]

+ Vk Vcr [G km sin(θk − δcr ) − Bkm cos(θk − θcr )]
+ Vk Vcr [G vr sin(θk − δcr ) − Bvr cos(θk − θcr )] (4)

At bus m:

Pm = Vm2 G mm + Vk Vm [G km cos(θm − θk ) + Bmm sin(θm − θk )]

+ Vm Vcr [G mm cos(θm − δcr ) + Bkm sin(θm − θcr )] (5)

Q m = −Vm2 Bmm + Vk Vm [G km sin(θm − θk ) − Bkm cos(θm − θk )]

+ Vm Vcr [G mm sin(θm − δcr ) − Bmm cos(θm − θcr )] (6)

At series converter:

Pcr = Vcr2 G mm + Vk Vcr [G km cos(δcr − θk ) + Bkm sin(δcr − θk )]

+ Vm Vcr [G mm cos(δcr − θm ) + Bmm sin(δcr − θm )] (7)

Q cr = −Vcr2 Bmm + Vk Vcr [G km sin(δcr − θk ) − Bkm cos(δcr − θk )]

+ Vm Vcr [G mm sin(δcr − θm ) − Bmm cos(δcr − θm )] (8)

At shunt converter

Pvr = −Vvr2 G vr + Vk Vvr [G vr cos(δvr − θk ) + Bvr sin(δvr − θk )] (9)

Q vr = Vvr2 Bmm + Vk Vvr [G vr sin(δvr − θk ) − Bvr cos(δvr − θk )] (10)

Voltage Stability and Performance Analysis … 5

Power equations of UPFC in the linearized form are formulated with those of AC
network. It is applied when UPFC controls the following parameters [6]:
• Shunt converter voltage magnitude
• Flow of real power from bus k to m
• Bus m is taken as PQ bus, and reactive power is injected at bus m
System of equations in linearized form is given below:
⎡ ⎤
Pk
⎢ P ⎥
⎢ m ⎥
⎢ Q ⎥
⎢ k ⎥
⎢ ⎥
⎢ Q m ⎥
⎢ ⎥
⎢ Pkm ⎥
⎢ ⎥
⎣ Q km ⎦
Pbb
⎡ ∂ Pk ∂ Pk ∂ Pk ∂ Pk ∂ Pk ∂ Pk ∂ Pk ⎤⎡ ⎤
∂θk
|V |
∂θm ∂|Vvr | vr
|V |
∂|Vm | m ∂δcr
|V |
∂|Vcr | cr ∂δvr θk
⎢ ∂ Pm ∂ Pm ∂ Pm ∂ Pm ∂ Pm ⎥⎢
⎢ 0 |V | |V | 0 ⎥⎢ θm ⎥
⎢ ∂θk ∂θm ∂|Vm | m ∂δcr ∂|Vcr | cr
⎥⎢ |Vvr | ⎥
⎢ ∂ Qk ∂ Qk ∂ Qk
|V | ∂ Qk
|V | ∂ Qk ∂ Qk
|V | ∂ Qk
⎥⎢ |Vvr | ⎥
⎢ ∂θk ∂θm ∂|Vvr | vr ∂|Vm | m ∂δcr ∂|Vcr | cr ∂δvr ⎥⎢ |Vm | ⎥⎥
=⎢
⎢
∂ Qm ∂ Qm
∂θk ∂θm
0 ∂ Qm
|V |
∂|Vm | m
∂ Qm
∂δcr
∂ Qm
|V |
∂|Vcr | cr
0 ⎥⎢
⎥⎢ |Vm | ⎥ (11)
⎢ ⎥⎢ δ ⎥
cr ⎥
∂ Pkm ∂ Pkm ∂ Pkm ∂ Pkm ∂ Pkm
⎢ 0 |V | |V | 0 ⎥⎢
⎢ ∂θk ∂θm ∂|Vm | m ∂δcr ∂|Vcr | cr
⎥⎣ |Vcr | ⎥
⎣ ∂ Q km ∂ Q km
∂θk ∂θm
0 ∂ Q km
|V |
∂|Vm | m
∂ Q km
∂δcr
∂ Q km
|V |
∂|Vcr | cr
0 ⎦ |Vcr | ⎦
∂ Pbb ∂ Pbb ∂ Pbb ∂ Pbb ∂ Pbb ∂ Pbb ∂ Pbb δvr
∂θk
|V |
∂θm ∂|Vvr | vr
|V
∂|Vm | m
| ∂δcr
|V |
∂|Vcr | cr ∂δvr

3 Simulation Results

3.1 IEEE 14 Bus Test System

An IEEE 14 bus system is taken for this simulation. It has two generator buses, a
slack bus, 11 load buses and 14 transmission lines. Base MVA and voltage are taken
as 100 MVA and 69 kV, respectively. Test system is shown in Fig. 4.

3.2 P–V Curves

Change in bus voltage w.r.t. loading factor (λ) is calculated for the test system. The
whole study has been carried out in PSAT software. It is observed that line (9–14)
shows most insecurity during the analysis when L-index and FVSI are used. P–V
curve of the three lower most buses without UPFC is given in Fig. 5, while Fig. 6
6 A. K. Dwivedi and S. Vadhera

Fig. 4 IEEE 14 bus test system

Fig. 5 Three lower most V

without UPFC

Fig. 6 Three lower most V

with UPFC
Voltage Stability and Performance Analysis … 7

represents the lowest three buses when UPFC is implemented in IEEE 14 bus test
system (Tables 1 and 2).

3.3 Line Losses

Line losses of the most severe lines (9–14) and (13–14) have been calculated with
and without UPFC and given in Table 4.

3.4 Discussion

In Figs. 5 and 6, it can be observed that after implementation of UPFC at bus 14,
the P–V curve limit extends to greater value. Table 3 shows the change in the index
with and without UPFC as the load is varied at steps of 5%. Table 4 gives the active
as well as the reactive power loss of the line (9–14) and (13–14) with and without
UPFC, and it is clearly visible that losses decrease in greater extent when UPFC is
used in the system.

4 Conclusion

It can be observed from above simulation results that when UPFC is implemented in
the system, then active as well as the reactive power losses of the system decrease up
to 50% level, and the stability index for the system is also improved. P–V curve for
the system shows that with UPFC, system has greater stability for load variation. The
paper confirms that UPFC is a better FACTS device that enables one to control the
various parameters of the power system network simultaneously. Moreover, PSAT
enables user to smoothly conduct the steady state and transient analysis of power
system. Hence, paper is the pioneer one which conducts the performance analysis
and voltage stability of the power system using a UPFC in PSAT software.
 8

Table 1 Bus voltage without and with UPFC

Bus no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Voltage without UPFC 1.06 1.045 0.996 1.001 1.01 1.07 1.029 1.09 1.001 1.0019 1.0301 1.0423 1.0289 0.9639
Voltage with UPFC 1.06 1.045 1.01 1.009 1.014 1.07 1.034 1.09 1 1.0018 1.0306 1.0439 1.0344 0.9851
A. K. Dwivedi and S. Vadhera
Voltage Stability and Performance Analysis … 9

Table 2 Line index

Line L-index FVSI
14–9 0.116751 0.115201
14–13 0.223402 0.223322
13–12 0.026021 0.053321
13–6 0.1139092 0.113090
12–6 0.064941 0.0649211
10–9 0.0098220 0.0096617
10–11 0.094714 0.0947024
4–5 0.051780 0.0517801
4–2 0.043972 0.0439010
4–3 0.157253 0.150704
5–2 0.055584 0.055528
5–1 0.089680 0.089704
2–3 0.0109251 0.0109071
1–2 0.089042 0.088902

Table 3 Load variation versus FVSI

Change in load FVSI (13–14) FVSI (9–14) FVSI (13–14) FVSI (9–14) with
without UPFC without UPFC with UPFC UPFC
0 0.2234 0.11675 0.23075 0.109433
5 0.2322004 0.1248 0.230576 0.120011
10 0.241173 0.133091 0.24074 0.13062
15 0.2501994 0.14136 0.2457 0.14127
20 0.25929 0.1497 0.25078 0.1491
25 0.268393 0.1581 0.2524 0.1555
30 0.2775965 0.16656 0.26094 0.1604
35 0.28686 0.17508 0.26601 0.1690
40 0.296214 0.18366 0.27111 0.17768

Table 4 Line losses

Losses Losses without any Losses at line (9–14) Losses at line (13–14)
FACTS device with UPFC with UPFC
Real losses 0.30192 0.18449 0.19767
Reactive losses 0.94172 0.43292 0.45828
10 A. K. Dwivedi and S. Vadhera

References

1. Power System Analysis Toolbox Quick Reference Manual for PSAT version 2.1.2, 26 June 2008
2. Seifi A, Gholami S, Shabanpour A (2010) Power flow study and comparison of FACTS: Series
(SSSC), Shunt (STATCOM), and Shunt-Series (UPFC). Pacific J Sci Technol 11(1)
3. Abu-Siada A, Karunar C (2012) Improvement of transmission line power transfer capability,
case study. Electr Electron Eng Int J (EEEIJ) 1(1)
4. Kazemi A, Naghshbandy AH (2006) A hybrid approach in study of UPFC effects on the transient
stability of multi machine power systems. Int J Electr Eng 88(2):125–131
5. Chennapragada B, Kotamarti VK, Sankar SB, Haranath PV (2003) Power system operation
and control using FACT devices. In: 17th international conference on electricity distribution,
Barcelona, 12–15 May 2003
6. Chakraborty K, Majumdar S, Chattopadhyay PK, Nandi C (2011) Active line flow control of
power system network with FACTS devices of choice using soft computing technique. Int J
Comput Appl 25(9)
7. Babu VN, Sivanagaraju S (2012) A new approach for optimal power flow solution based on two
step initialization with multi-line FACTS device. Int J Electr Eng Inf 4(1)
Harmonic Analysis at Variable Load

Vijit Srivastava and P. N. Gupta

Abstract In this paper, harmonics due to variable load conditions can be discussed
with mitigation technique for a three-phase power supply and compare the harmonics
of source current caused by nonlinear variable loads and analyzed all the resultant
signals through wavelet transform by using Haar and Daubechies function. Results
through Matlab/Simulink simulation show that Total Harmonic Distortion (THD) in
source current is varied according to the varying load. These results are also verified
by Wavelet Function. “Wavelet Based Method” is also very effective for showing the
result of varying loads due to harmonics in nonconventional energy resources like
current harmonics in the wind energy system.

Keywords APF · Variable loads · Harmonic analysis · Simulation results ·

Wavelet signal analysis

1 Introduction

Day by day the demand of the energy (electricity) will be increases but the resources
are limited so it is very difficult to make a balance between demand and supply ratio
for electricity. Each country will also try to reduce the effect of greenhouse so they
will no longer support traditional methods of energy generation like from coal, oil or
natural gas etc. Making clean and pollution free environment switch to nonconven-
tional energy resources like solar energy, wind energy, biomass, etc. but the using
of nonconventional method facing a lot of problems like voltage regulation, power
quality issues, stability etc. Power quality is the major issue in distribution system
because of using non linear loads in the system. Nonlinear loads injected harmonics
into the system and give rise to nonsinusoidal voltages and currents [1]. Due to the

V. Srivastava (B) · P. N. Gupta

J.K. Institute, University of Allahabad, Allahabad, India
e-mail: [email protected]
P. N. Gupta
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 11

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_2
12 V. Srivastava and P. N. Gupta

variation in load, supply current is also varying according to varying load when load
decreases supply current also decreases and when it increases supply current also
increases. Variable load with active power filters are modeled for harmonic analysis
and also to reduce the total harmonic distortion due to nonlinear loads. The perfor-
mances of active power filters have been determined by operating on three-phase
three-wire system [2–6].
Resultant signals are also analyzed by wavelet transform method by using Haar
and Daubechies functions with 2D analysis [7].

2 Proposed Technique

2.1 Active Power Filter

In electrical power supply system harmonics is one of the biggest issues. Harmonics
are developed in the system due to the nonlinear load and it polluted the supply
current. It is the integral multiple of fundamental frequency which causes the distorted
waveform results are additional power loss, heating effect and also reduces the life
of the component, extra burden on the wires etc. Hence harmonics elimination is
required for smooth conduction of electricity in the supply system mainly in distri-
bution and transmission system. So the proposed technique is active power filter is
used for harmonic reduction this is one of the simple and best methods. Principle
of active power filter is to inject harmonic currents equal in magnitude but opposite
in phase to those harmonics that are present in the system due to non linear load.
The characteristics of nonlinear load would be changed with APF. Applications of
active power filters have become more popular and have attracted great attention
as compare to other methods of mitigation technique. It removes the drawback of
passive power filters, like component aging and resonant problems [8–11].

2.2 Variable Load

Now, when we are using the varying load condition with APF with different time
duration THD should also vary according to the given time duration which was
reduced by APF circuit and shows that supply current is varied according to the
varying load [12] (Fig. 1).
Control strategies of APF
Harmonic Analysis at Variable Load 13

I(supply) I(load)

Varying non
Z
linear load

I(injected)
AC

Power Energy storage

Conditioner

Fig. 1 Block diagram of Shunt APF

I(load) = I(supply) + I(injected)

I(injected) = I(harmonics) + I(reactive part)
I(load) = I(real) + I(reactive part) + I(Harmonics)
I(real) = I(load) − I(harmonics) − I(reactive part)
= I(supply)

3 Matlab Model/Simulation Result Discussions

3.1 Active Power Filter with Variable Load

Now, when we are using varying load in one simulation for 1.5 sec of total time
duration in which circuit carry full load for 0–0.5 sec time duration and circuit carry
50% of full load for 0.6–1 sec time duration and circuit carry 1.5 times of full load
for 1.1 to 1.5 sec time duration (Fig. 2).
In the above condition, supply current is also varying according to varying load
when load decreases supply current also decreases and when it increases supply
current also increases.
All 1000 samples correspond to lower frequency of the analysis, carry relevant
information. While disturbances occur at instant 500, 730 of samples. This shows
good time and poor frequency resolution (Table 1).
THD in the supply current will be decreasing when load increases and it will
increase when load decreases.
14 V. Srivastava and P. N. Gupta

(a)

(b)
30

20

10
Amplitude

-10

-20

-30
0 0.5 1 1.5
Time
(c)

Fig. 2 Matlab lab model of active power filter with variable load system. a Circuit model of
APF with varying load. b Simulation waveform APF with varying load. c Single-phase simulation
waveform of APF with varying load. d Wavelet-based analysis of APF with varying load by using
Haar function. e Wavelet-based analysis of APF with varying load by using 2db (Daubechies) and
4 level decomposition
Harmonic Analysis at Variable Load 15

(d)

(e)

Fig. 2 (continued)
16 V. Srivastava and P. N. Gupta

Table 1 THD comparison

Time instant Percentage of THD
with variable load
0.1 4.41
0.3 4.28
0.5 7.74
0.7 8.13
0.9 8.10
1.1 2.96
1.3 2.96
1.4 2.96

4 Conclusion

According to the result analysis, when considering the varying load condition it
shows that variations in load show corresponding variation in supply current, supply
current increases when load is increased and it will decrease when load decreases
and the THD of supply current will be decreases when load is increases and it
will be increases when load is decreases. According to the result analysis, when
considering the wavelet approach it gives the better result for analysis of the signal.
This control algorithm performance can be evaluated under varying load condi-
tion. Discrete wavelet analysis provides sufficient information both for analysis and
synthesis of the original signal. It also provides the information of that point where
disturbances occur.

References

1. Newman MJ, Zmood DN, Holmes DG (2002) Stationary frame harmonic reference generation
for active filter systems. IEEE Trans Ind Appl 38(6), 1591–1599
2. Wang F, Duarte J, Hendrix M (2011) Grid-interfacing converter systems with enhanced voltage
quality for microgrid application; concept and implementation. IEEE Trans Power Electron
26(12):3501–3513
3. de Araujo Ribeiro R, de Azevedo C, de Sousa R (2012) A robust adaptive control strategy
of active power filters for power-factor correction, harmonic compensation, and balancing of
nonlinear loads. IEEE Trans Power Electron 27(2):718–730
4. Wei X (2010) Study on digital pi control of current loop in active power filter. In: Proceedings of
2010 International Conference on Electrical and Control Engineering, June 2010, pp 4287–4290
5. Cheng P-T, Bhattacharya S, Divan D (2001) Operations of the dominant harmonic active filter
(DHAF) under realistic utility conditions. IEEE Trans Ind Appl 37(4), 1037–1044
6. Odavic M, Biagini V, Zanchetta P, Sumner M, Degano M (2011) Onesample- period-ahead
predictive current control for high-performance active shunt power filters, Power Electronics.
IET 4(4):414–423
7. Soman KP, Ramachandran KI (2004) Insight into wavelets-from theory to practice
8. Balbo N, Penzo R, Sella D, Malesani L, Mattavelli P, Zuccato A (1994) Simplified hybrid
active filters for harmonic compensation in low voltage industrial application. In: Proceedings
of 1994 international conference on harmonics in power systems, pp 263–269
Harmonic Analysis at Variable Load 17

9. Rastogi M, Mohan N, Edris AA (1995) Hybrid-active filtering of harmonic currents in power

systems. IEEE Trans Power Deliv 10:1994–2000
10. Fujita H, Yamasaki T, Akagi H (2000) A hybrid active filters for damping of harmonic resonance
in industrial power system. IEEE Trans Power Electron 15:215–222
11. Satyanarayana G, Lakshmi Ganesh K, Narendra Kumar C, Vijaya Krishna M (2013) A crit-
ical evaluation of power quality features using Hybrid Multi-Filter Conditioner topology. In:
2013 international conference on green computing, communication and conservation of energy
(ICGCE), pp 731–736, 12–14 Dec 2013
12. Gonzatti RB, Ferreira SC, Silva CH, Silva LEB, Lambert-Torres G, Fernandez Silva LG
(2012) PLL-less control strategy applied to hybrid active series power filter. In: 10th IEEE/IAS
international conference on industry applications (INDUSCON 2012), Fortaleza, Brazil
Voltage and Transient Stability
Enhancement in Power System Using
Unified Power Flow Controller

Jaswant Singh Bhati and Shelly Vadhera

Abstract The power system is getting more and more complex by the use of multi-
machine systems. With multi-machine systems in use, the need for voltage and tran-
sient stability arises. In this paper, standard test system, having two machines and
four buses is taken into account to show the better performance of the efficacious
flexible alternating current transmission system (FACTS) device named as unified
power flow controller (UPFC) in enhancing the voltage and transient stability in a
power system network. Shunt branch of UPFC help in improving the voltage stability
by exchanging the reactive power with the system whereas a series branch of UPFC
help in improving transient stability by injecting variable voltage and exchanging
real power with the system. Various types of symmetrical and unsymmetrical faults
are created in the standard system and its behavior is observed. The test system is
simulated using MATLAB/Simulink.

Keywords FACTS · UPFC · Voltage stability · Transient stability · Faults

1 Introduction

Unified power flow controller (UPFC) is a voltage source based flexible a.c. trans-
mission system (FACTS) device. It has two converters one in the shunt and other
in the series, hence it is called a shunt-series connected FACTS device and is used
here for the purpose to improve the voltage and transient stability of the system.
The effect of UPFC on transient stability investigation of a longitudinal system is
exhibited in [1]. In this paper, a mathematical model is developed to study the rela-
tionship between the UPFC and transmission system. The impact of UPFC on a
longitudinal system for the enhancement of its transient stability is analyzed in [2].
A mathematical model is developed to describe the interdependence of longitudinal

J. S. Bhati (B) · S. Vadhera

Department of Electrical Engineering, National Institute of Technology, Kurukshetra, Haryana
136119, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 19

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_3
20 J. S. Bhati and S. Vadhera

system parameters and operating condition with UPFC. Reference [3] discusses tran-
sient power flow in UPFC experimentally as well as theoretically. It reveals that the
outflow of active power from a series device of UPFC is stored as line inductance in
transient state. In order to enhance the transient and voltage stability, a coordinated
excitation controller and UPFC is designed with a one-machine, one-load power
system in [4]. To improve the stability, a model of UPFC for reference identifica-
tion of series part based on state variables is designed in [5]. Here, the UPFC is
connected to an infinite bus and as such control of voltage magnitude is neglected.
By harmonizing the active and reactive power, a massive reformation in the first
swing transient stability is obtained. An H∞ scheme is applied to a single-neuron
radial based function neural network in order to design the control strategy of UPFC
[6]. This type of controller of UPFC provides improvement in transient stability to
an expanded boundary of operating conditions. Reference [7] describes the dynamic
regulation model (DRM) for VSC based on FACTS controller. A comparison of
DFIG and DDPMSG is done on their construction, operating principle and modeling
in [8]. The result shows the effectiveness of UPFC on the stability of DDPMSG when
it is installed. Reference [9] exhibits a cascaded control arrangement to stabilize the
a.c. current of the transmission line with MMC-UPFC which is based on voltage
limit control. With suitable transformer connections for MMC-UPFC, the sequence
currents except positive sequence are quelled with the help of resultant current loops.
The results show the effectiveness of UPFC in maximizing the controllable region
during unbalanced grid situation. In this work, UPFC is implemented in test system
and its effect on the system is observed in different fault conditions. It has improved
the voltage as well as transient stability of the system in all the cases.
The organization of this paper is done in the following manner that
MATLAB/Simulink model description is given in Sect. 2 whereas simulation results
related to voltage, active power, reactive power with and without UPFC along with
fault clearing time are presented in Sect. 3, Sect. 4 includes the conclusion of the
work done.

2 MATLAB/Simulink Model Description

2.1 Two Machine Four Bus System Without UPFC

The system has two generating stations, four buses, two transformers, four three-
phase loads, four different transmission lines (i.e., 280, 150, 150, and 50 km) and
one three-phase fault is taken for the study of the voltage and transient stability of the
test system network. This model (as shown in Fig. 1) is simulated without installing
UPFC and results are recorded in both graphical and tabular formats.
Voltage and Transient Stability Enhancement in Power System Using … 21

B1

Generating Two Winding

Station 1 Transformer Line 2-1
Three 150 KM Three
Phase Phase
Load Fault

B2 B4 Three
Phase
Load
Line 1 Line 2-2
280 KM 150 KM

Line 3 B3
50 KM

Generating Two Winding

Station 2 Transformer
Three Three
Phase Phase
Load Load

Fig. 1 Model representation of two machine four bus system

B1

Generating Two Winding

Station 1 Transformer Line 2-1
Three 150 KM Three
Phase UPFC Phase
Load Fault

B2 B4 Three
Phase
Load
Line 1 Line 2-2
280 KM 150 KM

Line 3 B3
50 KM

Generating Two Winding

Station 2 Transformer
Three Three
Phase Phase
Load Load

Fig. 2 Model representation of two machine four bus system with UPFC

2.2 Two Machine Four Bus System with UPFC

In this model (Fig. 2) one FACTS device named UPFC is installed at the line between
bus B1 and B2 near to bus B2 to improve the voltage and transient stability and to
reduce the fault clearing time of the system. This model also contains the same
equipment used in Fig. 1 except for UPFC.

2.3 Different Types of Faults

The faults in power systems can be classified into two broad categories:
1. Symmetrical Faults
2. Unsymmetrical Faults
22 J. S. Bhati and S. Vadhera

Symmetrical faults are also called balanced faults. When all the three phases get
simultaneously short-circuited then this type of fault is considered as a symmetrical
fault. The probability of occurrence of this type of fault in the system is very less
(i.e., 2–5%) but the severity of these faults is very high and they can damage the
system even though it is in a balanced condition. Examples of symmetrical faults are
three-phase and three-phase to ground fault.
Unsymmetrical faults are also called as unbalanced faults. Both open circuit faults
and short circuit faults come under the category of unsymmetrical faults. The prob-
ability of occurrence of these faults is given in such a manner: line to ground fault
(85%), double line fault (8%) and double line to ground fault (3–5%). Both LG and
LL faults are less severe compared to LLG fault which has more severity. In this
paper, both symmetrical and unsymmetrical faults are studied and performed on the
test system. The effect of faults on this system are observed and UPFC performs its
function very efficiently by reducing the fault clearing time and by improving the
voltage profile of the system.

3 Simulation Results

3.1 System Without UPFC

The simulation is run for 10 s without UPFC. Different types of faults (i.e., symmet-
rical and unsymmetrical faults) are considered. Types of faults considered here are
line to ground (LG) fault, double line (LL) fault, double line to ground (LLG) fault,
three-phase (LLL) fault and three-phase to ground (LLLG) fault. Here the prime
focus is on the voltage profile, active and reactive power of the test system.

3.2 System with UPFC

Similarly, the simulation is run for 10 s with UPFC by creating all types of faults.
The UPFC is placed near to bus B2. The voltage at bus B2 is captured to show the
effect of UPFC for voltage profile improvement. The same parameters are calculated
and analyzed as discussed in Sect. 3.1.
The simulation results are captured in Tables 1 and 2 and in Figs. 3, 4, 5, 6, 7 and
8. Table 1 tabulates the data of voltage, active and reactive power with and without
UPFC. Figures 3, 4, 5, 6 and 7 shows the graphical representation of voltage at
bus B2 with and without UPFC. Table 2 tabulates the fault clearing time for all the
types of faults considered here with and without UPFC whereas, Fig. 8 presents the
comparison of fault clearing time with and without UPFC.
Table 1 Effect of faults on voltage, active and reactive power of different buses
Type of fault LG LL LLG LLL LLLG
Parameters Bus No. Without With Without With Without With Without With Without With
UPFC UPFC UPFC UPFC UPFC UPFC UPFC UPFC UPFC UPFC
Voltage (per B1 1.007 1 1.007 1 1.007 1 1.007 1 1.007 1
unit) B2 1.007 1.069 1.007 1.069 1.007 1.069 1.007 1.069 1.007 1.069
B3 1.002 1.015 1.002 1.015 1.002 1.015 1.002 1.015 1.002 1.015
B4 1.015 1.015 1.015 1.016 1.015 1.016 1.015 1.016 1.015 1.015
Active power B1 1337 1363 1338 1363 1338 1363 1337 1363 1337 1363
(MW) B2 663.9 597.8 664 597.9 664 579.9 663.9 597.9 663.9 597.8
B3 989.9 1004 990.1 1004 990.1 1004 990.1 1004 990.1 1004
B4 563.3 652.4 563.4 652.4 563.4 652.4 563.3 652.4 563.3 652.3
Reactive B1 −251.2 −133.2 −251.4 −133 −251.4 −133 −251.3 −133 −251.4 133.2
power B2 −121.3 −27 −121.4 −26.99 −121.4 −26.99 −121.4 −26.99 −121.4 −27
Voltage and Transient Stability Enhancement in Power System Using …

(MVAR)
B3 −32.5 −114.4 −32.63 −114.4 −32.62 −114.4 −32.61 −114.4 −32.63 −114.4
B4 −39.33 −95.7 −39.37 −95.67 −39.37 −95.75 −39.4 −95.75 −39.35 −95.7
23
24 J. S. Bhati and S. Vadhera

Table 2 Effect of UPFC on

Type of fault Case Fault clearing time (s)
fault clearing time
LG Without UPFC 6.68
With UPFC 5.20
LL Without UPFC 6.92
With UPFC 5.81
LLG Without UPFC 5.98
With UPFC 4.75
LLL Without UPFC 5.52
With UPFC 4.98
LLLG Without UPFC 5.02
With UPFC 4.01

Fig. 3 Voltage improvement at bus 2 with LG fault

Fig. 4 Voltage improvement at bus 2 with LL fault

Voltage and Transient Stability Enhancement in Power System Using … 25

Fig. 5 Voltage improvement at bus 2 with LLG fault

Fig. 6 Voltage improvement at bus 2 with LLL fault

Fig. 7 Voltage improvement at bus 2 with LLLG fault

26 J. S. Bhati and S. Vadhera

Fault Clearing Time (Seconds)

With UPFC
7
Without UPFC
6
5
4
3
2
1
0
LG LL LLG LLL LLLG
Type of Fault

Fig. 8 Fault clearing time for different faults with and without UPFC

3.3 Fault Clearing Time

Time taken by the system between fault initiation and fault clearance is considered as
fault clearing time of that system. Two machine four bus system had been simulated
two times once without UPFC and next time with UPFC. It is found that if there is
any fault in the given system then it takes usually more time to clear the fault when
there is no UPFC installed. But when UPFC is introduced in the system it’s fault
clearing time gets reduced significantly.
All the symmetrical faults (i.e., LLL fault and LLLG fault) and unsymmetrical
faults (i.e., LG fault, LL fault and LLG fault) are introduced in the test system. In both
the cases, i.e., with and without UPFC the effect on the system have been observed
and it is found that using UPFC one can reduce the fault clearing time in the system
thereby improving the transient stability of the system.

4 Conclusion

Voltage and transient stability play a crucial role in the power system are achieved
using UPFC. It has reduced fault clearing time of LG fault by 22.15% which is the
highest time reduction among all the faults and for LLL fault it reduced fault clearing
time by 9.78% which is the smallest reduction among all the faults.
The contribution of this paper is enlisted below:
1. Simulation of two machine four bus system with and without UPFC is performed
2. Analysis of the effect of all symmetrical and unsymmetrical faults for the test
system under consideration
3. Graphical and numerical results for voltage, active power and reactive power are
recorded
4. Fault clearing time for a different type of faults are calculated
Voltage and Transient Stability Enhancement in Power System Using … 27

Simulation results show the contribution of UPFC in enhancing the voltage and
transient stability and reducing the fault clearing time for various type of faults.

References

1. Mihalic R, Zunko P, Povh D (1996) Improvement of transient stability using unified power flow
controller. IEEE Trans Power Del 11(1):485–492
2. Kang YL, Shresta GB, Lie TT (2001) Application of an NLPID controller on a UPFC to
improve transient stability of a power system. IEE proceedings—generation, transmission and
distribution, vol 148, no. 6, pp 523–529
3. Fujita H, Watanabe Y, Akagi H (2001) Transient analysis of a unified power flow controller and
its application to design of the DC-link capacitor. IEEE Trans Power Electron 16(5), 735–740
4. Chen H, Wang Y, Zhou R (2001) Transient and voltage stability enhancement via coordinated
excitation and UPFC control. IEE proceedings—generation, transmission and distribution, vol
148, no. 3, pp 201–208
5. Gholipour E, Saadate S (2005) Improving of transient stability of power systems using UPFC.
IEEE Trans Power Del 20(2):1677–1682
6. Mishra S (2006) Neural-network-based adaptive UPFC for improving transient stability
performance of power system. IEEE Trans Neural Netw 17(2), 461–470
7. Jiang X, Chow JH, Edris A, Fardanesh B, Uzunovic E (2010) Transfer path stability enhancement
by voltage-sourced converter-based FACTS controllers. IEEE Trans Power Del 25(2), 1019–
1025
8. Liu H, Li X, Qin G, Hao S (2017) Stability of grid connected system of two types of wind
turbines with UPFC. J Eng 2017(13), 2178–2183
9. Hao Q, Man J, Gao F, Guan M (2018) Voltage limit control of modular multilevel converter
based unified power flow controller under unbalanced grid conditions. IEEE Trans Power Del
33(3):1319–1327
Probabilistic Approach for Load Flow
Analysis of Radial Distribution System

Anuj Banshwar, Mohit Pathak, Bharat Bhushan Sharma,

Naveen Kumar Sharma, and Sumit Kumar

Abstract This paper presents a probabilistic approach for load flow method for the
radial distribution systems (RDS). The uncertainties existing in the line data and the
load data have been taken into consideration. The proposed power flow method will
be useful for planning purposes of radial distribution systems where the uncertainty
is there in the assumed data. The sensitivity index and voltage stability index are also
calculated in order to find the most sensitive node.

Keywords Radial distribution system · Probabilistic load flow · Direct load flow ·
Bus injected to branch current · Branch current bus voltage · Voltage stability index

1 Introduction

Power flow (or load flow) analysis is absolutely an important and fundamental tool in
power system planning. This method is used to determine the operating state of any
system, to compare different planning schemes, and to provide the preliminary state
for other advanced applications. The growing integration of distributed generations
(DGs) in transmission as well as distribution systems inevitably results in growing
amount of uncertainties to the power system operative states. Therefore, adopting a

A. Banshwar (B)
Government Polytechnic Puranpur, Pilibhit, Uttar Pradesh, India
e-mail: [email protected]
M. Pathak
Noida Institute of Engineering & Technology, Greater Noida, Uttar Pradesh, India
B. B. Sharma
School of Automation, Banasthali Vidyapith, Rajasthan, India
N. K. Sharma
IKG Punjab Technical University, Main Campus, Punjab, India
S. Kumar
Punjab State Power Corporation Limited, Punjab, India

© Springer Nature Singapore Pte Ltd. 2021 29

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_4
30 A. Banshwar et al.

probabilistic framework is vital to ensure accurate and precise power flow analysis
[1].
Actually, the proposal of probabilistic framework can be traced back to later
1970s; the aim was to incorporate uncertainties into the load flow problem [2, 3].
Thereafter, many efforts were made. The probabilistic load flow (PLF) requires inputs
with probability density function (PDF) or cumulative density function (CDF) to
obtain states of the system and power flows in terms of these density factors so that
the uncertainties in the system can be included. The PLF can be solved numerically,
i.e. using a Monte Carlo (MC) methods, or analytically or a combination of them.
There are various load flow algorithms which are proposed for the analysis of
transmission systems as well as for radial distribution system (RDS). However, most
of the proposed algorithms are based on the consideration of fixed input parameters
which are used to obtain the solution of RDS. But in actual practice, there are various
uncertainties which are existing in the input parameters and need to be considered in
order to find efficient solution; otherwise, the results obtained are only single-point
solutions and may not depict the real picture of the system operating conditions. The
probabilistic approach is used so that the uncertainties are being taken into account.
A simple load flow method which used direct approach has been proposed by Teng
in [4]. Das [5] developed bus injected to branch current (BIBC) and branch current
bus voltage (BCBV) matrices followed by a simple matrix multiplication in order
to make the solution possible by using the topology of the network and Kirchhoff’s
laws. All the above methods provided the load flow solution without considering the
uncertainties into account. These uncertainties need to be accounted for otherwise
the results obtained are merely a snapshot of the actual operating conditions.
In [6], various probabilistic load flow (PLF) techniques have been reviewed and
some improvements which are made in the PLF techniques are presented along with
the applications and the various extensions of the PLF algorithm. In [7], the various
uncertainties have been taken into consideration by PLF analysis and a new method
was presented by which the accuracy has been increased due to the multi-linear
simulation. In [8], the load uncertainty can be modelled through a simple Gaussian
distribution function which has been linearized at multiple points so as to obtain the
various intervals.
The algorithm proposed in this paper will take into consideration the data
uncertainties using probabilistic approach.

2 Load Flow Calculation of RDS

Consider a seven bus test radial distribution system as shown in Fig. 1.

By applying KCL in Fig. 1, the branch currents are expressed as

B1 = I1 + I2 + I3 + I4 + I5 + I6 + I7 (1.1)
Probabilistic Approach for Load Flow Analysis of Radial Distribution System 31

Fig. 1 A 7 bus test radial distribution system

B2 = I3 + I4 + I5 + I6 + I7 (1.2)

B3 = I4 + I5 (1.3)

B4 = I5 (1.4)

B5 = I6 + I7 (1.5)

B6 = I7 (1.6)

In the general form

[B] = [BIBC][I ] (2)

From Fig. 1, the following relations are obtained:

V2 = V1 − B1 Z 12 (3.1)

V2 = V1 − B1 Z 12 (3.2)

V3 = V2 − B2 Z 23 (3.3)

V4 = V3 − B3 Z 34 (3.4)
32 A. Banshwar et al.

V5 = V4 − B4 Z 45 (3.5)

V6 = V3 − B5 Z 36 (3.6)

V7 = V6 − B6 Z 67 (3.7)

After manipulations, the above voltage equations can be written as

V3 = V1 − B1 Z 12 − B2 Z 23 (4.1)

V4 = V1 − B1 Z 12 − B2 Z 23 − B3 Z 34 (4.2)

V5 = V1 − B1 Z 12 − B2 Z 23 − B3 Z 34 − B4 Z 45 (4.3)

V6 = V1 − B1 Z 12 − B2 Z 23 − B5 Z 36 (4.4)

V7 = V1 − B1 Z 12 − B2 Z 23 − B5 Z 36 − B6 Z 67 (4.5)

In matrix form
⎡ ⎤ ⎡ ⎤ ⎡ ⎤⎡ ⎤
V1 V2 Z 12 0 0 0 0 0 B1
⎢V ⎥ ⎢V ⎥ ⎢Z 0 ⎥ ⎢B ⎥
⎢ 1 ⎥ ⎢ 3 ⎥ ⎢ 12 Z 23 0 0 0 ⎥⎢ 2 ⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎢ ⎥
⎢ V1 ⎥ ⎢ V4 ⎥ ⎢ Z 12 Z 23 Z 34 0 0 0 ⎥⎢ B3 ⎥
⎢ ⎥=⎢ ⎥=⎢ ⎥⎢ ⎥ (5)
⎢ V1 ⎥ ⎢ V5 ⎥ ⎢ Z 12 Z 23 Z 34 Z 45 Z 36 0 ⎥⎢ B4 ⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎢ ⎥
⎣ V1 ⎦ ⎣ V6 ⎦ ⎣ Z 12 Z 23 0 0 Z 36 0 ⎦⎣ B5 ⎦
V1 V7 Z 12 Z 23 0 0 0 Z 67 B6

In general form

[V ] = [BCBV][B] (6)

Now, distribution load flow (DLF) matrix is obtained as follows:

[V ] = [BCBV][BIBC][I ] (7)

or

[V ] = [DLF][I ] (8)

The step-by-step algorithm is given below:

Probabilistic Approach for Load Flow Analysis of Radial Distribution System 33

1. The BIBC matrix is obtained by coping the entries of the ith column and then
reset the entry corresponding to the mth row and jthcolumn as 1. Repeat this
procedure so that all the line sections are included in Eq. (2) matrix.
2. The BCBV matrix is obtained by coping the entries in the ith row to the jth row
and then reset the entry corresponding to jth row and mth column with the line
impedance (Zth). Repeat this procedure so that all the line sections are included
in Eq. (6) matrix.
3. Calculate load current at each bus


 m Pi + j Q i
Ii = (9)
Vim

4. The elements of the V are computed using the equation

 
V m+1 = [DLF] I m (10)

where

[DLF] = [BCBV][BIBC]

5. Update the bus voltages at all buses

  
V m+1 = V O + V m+1 (11)

6. The above procedure is repeated till the following condition is satisfied

tol > (12.66∗0.05) (12)

3 System Data Modelling Uncertainty

The uncertainty in the line data is accounted by the changes in the resistance and
reactance values, and the uncertainty in the load data is accounted by the changes in
active and reactive power values. Therefore, these uncertainties need to be modelled
and the probabilistic approach is used for accounting these uncertainties. It has been
assumed that the uncertainty is modelled with the help of the equation given below:
  2 
 1 yj − μ
f yj = √ exp − (13)
σ 2π 2σ 2

The three cases have been discussed here, and the corresponding results are
obtained:
1 For varying line data and constant load data.
34 A. Banshwar et al.

Fig. 2 Actual voltage profile

of the system without
consideration of uncertainty

Fig. 3 Voltage profiles for

varying line data and
constant load data

2 For varying load data and constant line data.

3 For both varying line data and load data.
The load flow calculations give us the voltage magnitudes at various buses for
different cases which are also shown in Figs. 2, 3, 4 and 5. Here it has been assumed
that the uncertainty existing in the data varies 5% around the mean value. The voltages
at various buses are plotted for different cases so that the effect of uncertainties in
the data has been analysed.

4 Sensitivity Analysis and VSI

The sensitivity is an important factor to determine the stability of the system under
consideration. Therefore, it is very important to find the most sensitive node of the
system so that stability of the system is maintained. The sensitivity analysis is done
Probabilistic Approach for Load Flow Analysis of Radial Distribution System 35

Fig. 4 Voltage profiles for

varying load data and
constant line data

Fig. 5 Voltage profiles when

both the line data and load
data are varying

to see how the reactive power changes with change in voltage which is given by

dQ r (Vs − 2Vr cos δ)

SVQ = = (14)
dVr X cos δ − R sin δ

A simple assumption is made here because the angles are so small that their effect
may be neglected without any loss of accuracy; therefore, the above expression
reduces to
dQ r (Vs − 2Vr )
SVQ = = (15)
dVr X−R

The results obtained after ignoring the effect of angles are almost equivalent to
that obtained when the effect of angles has been taken into consideration. The VSI
also helps us to find out the sensitive nodes of the system. The VSI helps the operating
personnel to know the stability of the system nodes so that one can bring the voltage
36 A. Banshwar et al.

Fig. 6 Reactive
power–voltage sensitivity
curve

level within the prescribed range. Therefore,

V (i) RP(i) + XQ(i)

VSI(i) = − (16)
2 V (i)

If the value of sensitivity at a particular node is less, it means that node is the
more sensitive as compared to other nodes of the system. The sensitivity values at
various nodes are shown in Fig. 5, and it has been noted that node 32 is the most
sensitive node; nodes 11, 15, 17, 20, and 26 are also tending towards instability. The
voltage stability index has also been calculated, and the results are plotted as shown
in Fig. 6. It has been noted that node 32 is the most sensitive node. Therefore, it
has been concluded that both the sensitivity calculations and voltage stability index
calculations have revealed that node number 32 is the most sensitive node of the
radial distribution system.

5 Conclusion

The load flow analysis has been carried out on an IEEE 33 bus radial distribution
system. Three cases are considered in which the uncertainties in the input data have
also been taken into consideration. The voltage profile for different cases has been
drawn. The sensitivity analysis and voltage stability index calculations are also done
in order to find out the most sensitive node of the system.
Probabilistic Approach for Load Flow Analysis of Radial Distribution System 37

References

1. Sheng H, Wang X (2019) Probabilistic power flow calculation using non-intrusive low-rank
approximation method. IEEE Trans Power Syst 34:3014–3025
2. Borkowska B (1974) Probabilistic load flow. IEEE Trans. Power Apparatus Syst 93:752–759
3. Allan RN, Borkowska B, Grigg CH (1974) Probabilistic analysis of power flows. Proc Inst Electr
Eng 121:1551–1556
4. Teng JH (2003) A direct approach for distribution system load flow solutions. IEEE Trans Power
Syst 18:882–887
5. Das B (2006) Consideration of input parameter uncertainties in load flow solution of three-phase
unbalanced radial distribution system. IEEE Trans Power Syst 21:1088–1095
6. Chen P, Chen Z, Jensen BK (2008) Probabilistic load flow: a review. In: Proceedings: third
international conference on electric utility deregulation and restructuring and power technologies
(DRPT-2008), pp 1586–1591
7. Silva AML, Arienti VL (1990) Probabilistic load flow by a multilinear simulation algorithm.
Proc Inst Electr Eng 137:276–282
8. Chaturvedi A, Prasad K, Ranjan R (2006) Use of interval arithmetic to incorporate the uncertainty
of load demand for radial distribution system analysis. IEEE Trans Power Syst 21:1019–1021
Harmonic Issues in Non-conventional
Energy Supply and Its Remedy

Vijit Srivastava and P. N. Gupta

Abstract Day by day the demand for electricity will be increase because of
increasing population and industrial use. This greatly increases the demand for
electrical utility. Conventional resources are not enough for balancing the demand
and supply ratio at the present time. Now it is time to switch to non-conventional
sources so that balancing the above ratio. With the effect of harmonics in sources and
increasing loads, the power quality of the system is degraded; this is the major issue
of non-conventional resources like in wind energy. In this paper, various harmonic
mitigation techniques have been proposed for a three-phase power supply like PPF
which greatly reduces harmonics but suffers from the resonance and reactive power
issues; APF, according to the simulation result of this circuit, is better with the
PPF, and HAPF gives the best results as compared to other technique of harmonic
mitigation.

Keywords PPF · APF · HAPF · Harmonic analysis · Simulation results

1 Introduction

For making balance the demand and supply ratio of electricity people has moved
towards the non-conventional source of energy like solar energy, wind energy,
hydropower etc. Available conventional sources of electricity are not sufficient
concerns about demand and supply of world energy needs and also not concern
about global warming so it is time to move towards the non-conventional sources
of energy but the use of renewable energy sources facing the biggest problem of
power quality issues like harmonics. For making clean and pollution-free environ-
ment, switch to non-conventional energy resources; but the use of non-conventional

V. Srivastava (B) · P. N. Gupta

J.K. Institute, University of Allahabad, Allahabad, India
e-mail: [email protected]
P. N. Gupta
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 39

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_5
40 V. Srivastava and P. N. Gupta

Fig. 1 Block diagram of

PPF Non Linear
Source
Load

method is facing a lot of problems like voltage regulation, power quality issues, and
stability. Power quality is the major issue in the distribution system because of using
nonlinear loads, power electronics devices, inverter in the system [1]. Nonlinear
loads injected harmonics into the system and give rise to non-sinusoidal voltages
and currents. Due to the variation in load, supply current is also varying according
to varying load when load decreases, supply current also decreases and when it
increases, supply current also increases, and it affects the entire distribution system
like additional loss, overheating and also reduce the life of the system [2]. In this
paper to overcome of this problem of power quality issues, the proposed different
harmonic mitigation techniques like PPF, APF, HAPF are discussed along with their
simulated results and also concern the THD [3, 4].

2 Technique Used

2.1 Passive Power Filter

Passive power filter is the combination of series RLC which are connected in parallel
with the load. This circuit is very attractive because of its simplicity and cost-effective.
The main feature of this circuit has an ability to compensate reactive power and elimi-
nate harmonics, but the circuit is suffering from the resonance, instability, mistuning.
Parallel combination of PPF with the power system is used [5, 6] (Fig. 1).

2.2 Active Power Filter

In electrical power supply system, harmonics is one of the biggest issues. Harmonics
is developed in the system due to the nonlinear load and it polluted the supply
Harmonic Issues in Non-conventional Energy Supply and Its Remedy 41

I(supply) I(load)

Varying non
Z
linear load

I(injected)
AC

Power Energy storage

Conditioner

Fig. 2 Block diagram of Shunt APF

current. It is the integral multiple of fundamental frequency which causes the distorted
waveform, and the results are additional power loss, heating effect and also reduces
the life of the component and extra burden on the wires. Hence, harmonics elimination
is required for smooth conduction of electricity in the supply system mainly in
distribution and transmission system. So the proposed technique is active power filter
which is used for harmonic reduction; this is one of the simple and best methods.
The principle of active power filter is to inject harmonic currents which equal in
magnitude but opposite in phase to those harmonics that are present in the system due
to nonlinear load. The characteristics of nonlinear load would be changed with APF.
Applications of active power filters have become more popular and have attracted
great attention as compared to other methods of mitigation technique. It removes the
drawback of passive power filters like component aging and resonant problems [7,
8] (Fig. 2).
Control strategies of APF

I(load) = I(supply) + I(injected)

I(injected) = I(harmonics) + I(reactive part)
I(load) = I(real) + I(reactive part) + I(Harmonics)
I(real) = I(load) − I(harmonics) − I(reactive part)
= I(supply)
42 V. Srivastava and P. N. Gupta

Fig. 3 Block diagram of

HAPF Non Linear
Source
Load

Shunt
Active Filter

2.3 Hybrid Active Power Filter

Hybrid active power filter is the combination of PPF and APF so it has advantages of
both circuits; it greatly reduces harmonics as well as resonance issues. This circuit
is mostly used because of its simplicity and cost-effective [9, 10] (Fig. 3).

3 MATLAB/Simulation Result Discussion

To verify the results discuss different technique for harmonic mitigation, their circuit
and their simulated resultant waveform

3.1 Power Supply

Power supply consists of three-phase supply feeding a nonlinear load which is

injected 31.82% THD in first phase, 31.59% THD in second phase, and 31.86%
THD in third phase at instant of 0.1 s in supply (Fig. 4).

3.2 Passive Power Filter

This circuit greatly reduces the harmonics from the supply current 2.77% THD in
first phase, 2.79% THD in second phase, 2.85% THD in third phase, but the circuit
is suffered because of resonance problem (Fig. 5).
Harmonic Issues in Non-conventional Energy Supply and Its Remedy 43

(a)
200

150

100

50
Amplitude

-50

-100

-150

-200
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Time
(b)

(c)

Fig. 4 MATLAB model of supply and load system. a Circuit model of power supply. b Simulation
waveform of three-phase supply with harmonics. c THD graph of first-phase signal
44 V. Srivastava and P. N. Gupta

300

200

100
Amplitude

-100

-200

-300
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Time
(a)

(b)

Figure 5. a Simulation waveform of three-phase supply with the reduced harmonics. b THD graph
of first-phase signal of PPF

3.3 Active Power Filter

This technique greatly reduces harmonics as well as resonance problem; it has 2.77%
THD in first phase, 2.79% THD in second phase, and 2.85% THD in third phase
(Fig. 6).
Harmonic Issues in Non-conventional Energy Supply and Its Remedy 45

20

15

10

5
Amplitude

-5

-10

-15

-20
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Time
(a)

(b)

Fig. 6 a Simulation waveform of I abc . b THD graph of first-phase signal of APF

46 V. Srivastava and P. N. Gupta

3.4 Hybrid Active Power Filter

This circuit greatly reduces harmonics as well as resonance problem; it has 0.74%
THD in first phase of supply, 0.83% THD in second phase of supply, and 0.96%
THD in third phase of supply (Fig. 7).

200

150

100

50
Amplitude

-50

-100

-150

-200
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5

Time

(a)

(b)
Fig. 7 a Simulation waveform of I abc . b THD graph of first phase signal of HAPF
Harmonic Issues in Non-conventional Energy Supply and Its Remedy 47

4 Conclusion

According to the result analysis, PPF is suitable for its simplicity, cost-effective and
greatly reduces harmonics, but circuit suffers from the resonance; APF gives a better
result as compared to the PPF circuit for harmonics reduction as well as for resonance
problem; HAPF gives the best result as compared to APF or PPF; this circuit greatly
reduces harmonics as well as resonance problem and improves the power quality by
removing the polluted supply.

References

1. Balbo N, Penzo R, Sella D, Malesani L, Mattavelli P, Zuccato A (1994) Simplified hybrid

active filters for harmonic compensation in low voltage industrial application. In: Proceedings
of 1994 International Conference on Harmonics in Power Systems, pp 263–269
2. Valcárcel M, Mayordomo JG (1993) Harmonic power flow for unbalanced systems. IEEE Trans
Power Deliv 8(4):2052–2059
3. Wang F, Duarte J, Hendrix M (2011) Grid-interfacing converter systems with enhanced voltage
quality for microgrid application; concept and implementation. IEEE Trans Power Electron
26(12):3501–3513
4. Prabhakar N, Mishra M (2010) Dynamic hysteresis current control to minimize switching for
three-phase four-leg VSI topology to compensate nonlinear load. IEEE Trans. Power Electron.
25(8):1935–1942
5. Jou HL, Wu JC, Wu KD (2001) Parallel operation of passive power filter and hybrid power filter
for harmonic suppression. In: IEE Proceedings—Generations, Transmission and Distribution,
vol 148, pp 8–14 (2001)
6. Smith BC, Arrillaga J, Wood AR, Watson NR (1998) A review of iterative harmonic analysis
for AC-DC power systems. IEEE Trans Power Deliv 13(1):180–185
7. Xia D, Heydt GT (1982) Harmonic power flow studies, Part I—Formulation and solution,
Part II—Implementation and practical application. IEEE Trans Power Apparatus Syst, PAS-
101:1257–1270 (1982)
8. Xu W, Jose JR, Dommel HW (1991) A multiphase harmonic load flow solution technique’.
IEEE Trans Power Syst PS-6:174–182
9. Rastogi M, Mohan N, Edris AA (1995) Hybrid-active filtering of harmonic currents in power
systems. IEEE Trans Power Deliv 10:1994–2000
10. Fujita H, Yamasaki T, Akagi H (2000) A hybrid active filters for damping of harmonic resonance
in industrial power system. IEEE Trans. Power Electron 15:215–222
Artificial Neural Network-Based Source
Identification Producing Harmonic
Pollution in the Electric Network

Ankit Tayal, Lillie Dewan, and J. S. Lather

Abstract Massive use of power electronic devices which are nonlinear and mostly
unbalanced in nature has influenced the power quality (PQ) in electric distribution
networks. They not only create significant harmonic pollution in the electric power
system but also degrade the system along the grid. Prediction of the sources feeding
the electrical system with polluted disturbances is a pivotal point in the estimation of
the power quality. This paper proposes an fast and accurate approach based on artifi-
cial neural network (ANN). Unique identification of various types of devices along
with distinct harmonic signatures is achieved via ANN utilizing feature extraction
through input current waveform. To achieve feature extraction, multilayer percep-
tron with certain parameters is constructed and trained through backpropagation
algorithm with performance compared and evaluated. The ANN incorporating MLP
architecture with supervised learning is simulated in MATLAB. The results validate
the ability of the proposed architecture for efficient classification of device signature
for harmonic contributions with reasonably fast response.

Keywords Electric power quality · Artificial neural network · Harmonic

signature · Total harmonic distortion · Harmonic source identification · Smart grid

1 Introduction

Measurement and identification of harmonic sources require specialized instrumen-

tation. The complete measurement of harmonic data is useful for location and source
identification. The shunt capacitors are not considered towards harmonic source
recognition as it helps apriori elimination of redirected harmonic flow [1].
Electric power engineers and researchers have worked upon different aspects
of harmonics and inter-harmonics to measure, monitor, control, or even mitigate
it. Owing to their inherent nonlinear mapping characteristics, the artifical neural
networks find their applications for power system problems incorporating nonlinear

A. Tayal (B) · L. Dewan · J. S. Lather

National Insititute of Technology Kurukshetra, Kurukshetra, Haryana 136119, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 49

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_6
50 A. Tayal et al.

loads and with few harmonic measuring instruments [2]. Cascade correction networks
(CCN) can detect multiple harmonic sources in the power system if the current
injection-based power flow is provided by distributed control system [3]. Online
tracking and monitoring of multiple harmonic sources can be achieved by current
measurement at the incoming supply point present in any electrical installation [4].
Application of data mining process can differentiate between linear and nonlinear
loads. At the same time, it can identify various types of harmonic sources present in
residential setup [5].
State estimation method provides a vital insight about type of harmonic sources on
the basis of actual harmonic measurements necessary for all other harmonic-related
studies [6]. Location of harmonic sources can be detected by using state estimation
technique with least-square estimators with the help of observability analysis method
[7]. Adaptive wavelet neural network (AWNN) produces quick estimation of the
dominant harmonics, whereas identifying these via rotational invariance technique
contributed it to manage time-varying signals with greater reliability [8]. Harmonic
distortion analysis on each side of PCC shall be vital in the future studies and esti-
mation to have an estimate of contributions of each utility or customer for possible
contamination via harmonic distortions [9, 10].
Industrial and residential loads behave like harmonic voltage sources. The exper-
imental computation made at both sides of installation of capacitor bank validates
the concept that distortions in the currents and voltages are sensitive to the amount
of shunt compensation utilized [11].
A hardware model was designed to develop various datasets from different
harmonic sources measuring from Fluke-41 power quality analyser. Estimation of
harmonics by a particular source and identification of harmonic sources employed
in an electric setup was employed using pattern matching and estimation abilities of
ANN. Harmonic sources condition the applied voltage on the basis of their unique
impedance and characteristic; hence, the current data generated is called “harmonic
signature.” The identification of harmonic signature data using ANN architectures
with multiple layers was developed. Once trained, this model could identify the
devices on the basis of the current harmonics of the sources.
This paper is systematically structured in the following parts: Part 2 addresses
the hardware implementation for extracting the harmonic signatures from various
sources. It also explains the MLP-based NN models and feature extraction using
NNs. Part 3 discusses simulation results and provides conclusion and future work.

2 Hardware Implementation and Signature Identification

Model

In this paper, neural network was applied for harmonic source identification based
upon the unique signature of different sources. Different harmonic loads condi-
tion the applied voltage based upon their nonlinear impedance characteristic, and
Artificial Neural Network-Based Source Identification … 51

Fig. 1 Overall hardware implementation scheme

hence, current flowing through them contains unique harmonic signature. The peri-
odic components of harmonics especially in current waveforms inhibit rich features
for identifying harmonic signatures. The overall hardware implementation is shown
in Fig. 1.
The complete process was implemented in the following steps.

2.1 Data Acquisition and Computing

A laboratory-based hardware implementation was set up using four commonly used

devices. First, the data was measured for individual device. Since all the four sub-
components were commonly arranged in parallel, hence datasets with different
combination of devices were computed using algebraic combination of the data
of such sub-components. Fluke 41 power quality analyzer was utilized for capturing
the waveforms for major incoming sources in the laboratory as shown in Fig. 2.
With four devices available, the total combinations of different devices inducing
harmonics may vary from 0 to 16 (24 = 16) different states, each representing different
arrangments of components/devices switched on or off, i.e., from no source to all
the four sources generating harmonics in the circuit. The employed power quality

Fig. 2 Experimental
Schematics for Setup
installation
52 A. Tayal et al.

Table 1 Experimental data

Title Amps
as obtained from harmonic
analyzer ID 1
Type Harmonics
X Scale 48.95
X Resolution 0
X Size 50
X Unit Hz
X Label Frequency
Y Scale 0
Y Resolution 0
Y Size 0
Y Unit A
Y Label RMS
0 0
48.95 0
97.9 139
146.85 168
195.8 −171
244.75 −15
293.7 −155
342.65 166
391.6 −106
440.55 7
489.5 167
538.45 −149
587.4 −86
636.35 48
ss685.3 −63
734.25 105

analyzer computes the harmonic components through FFT in the current waveform
considering both amplitude and phase (Tables 1 and 2).
In the experiments, the harmonics in the current waveforms appeared in the appre-
ciable amounts from the fundamental components to upto fifteenth harmonic level.
Therefore, here only the odd harmonics were considered for proceeding further recog-
nition process taking the consolidated effect of amplitude as well as phase as per Eq.
(1):
  
k+1
x m = I( k+1
) cos ( k+1
) , for k = 1, 3, 5, 7, 9, 11, 13, 15; m = (1)
2 2 2
Artificial Neural Network-Based Source Identification … 53

Table 2 Modified dataset

Title Amps
considering only odd
harmonics as even harmonics ID 1
does not contribute much in Type Harmonics
total output of the systems
X Scale 48.95
X Resolution 0
X Size 50
X Unit Hz
X Label Frequency
Y Scale 0
Y Resolution 0
Y Size 0
YUnit A
Y Label RMS
48.95 0
146.85 168
244.75 −15
342.65 166
440.55 7
538.45 − 149
636.35 48
734.25 105

where x m is the mth input, I m represents the magnitude component, whereas m

accounts for the phase angle of the mth odd current harmonic, respectively (Table 3).
For a complete dataset taking into account of all possible combinations of all
the devices, i.e., no source, possible combination of two or three sources or all four
sources generating harmonics, different datasets were taken into consideration from
time to time. Since the data had to incorporate in neural network under no source
condition, rather than putting 0 in all the rows, authors had to use 0.00001 so as
to put some value near equal to 0 but not zero; otherwise, neural networks provide
ambiguous results (Table 4).

2.2 Characteristics of Harmonic Signatures

During the experiment, all the devices were connected in parallel; hence, no interfer-
ence was measured while the data acquisition took place from other devices. Also,
the wiring resistances of the devices connected on the source were assumed to be
negligibly small; hence, source voltage was least affected by these four loads.
Figure 3 represents one dataset of different signatures for particular harmonic for
54 A. Tayal et al.

Table 3 Final dataset for

Title Amps
individual source
ID 1
Type Harmonics
X Scale 48.95
X Resolution 0
X Size 50
X Unit Hz
X Label Frequency
Y Scale 0
Y Resolution 0
Y Size 0
YUnit A
Y Label RMS
48.95 0 0
146.85 168 126.1732
244.75 45 −9.41274
342.65 166 −133.588
440.55 7 −3.91038
538.45 −149 −127.033
636.35 48 23,31,076
734.25 105 −93.8851

the components installed in laboratory arrangement extracted by the clamper circuit

using Fluke 41 power quality analyser.
The drive panel showed the highest THD value because other electronic equip-
ments draw arbitrary amount of current from the panel’s power supply when
connected through the drive panel. On the other hand, resistive load such as the
fluorescent light shows lower THD value as compared to the other sources.
Since all harmonic sources exhibit nonlinear characteristics in nature; hence, arti-
ficial neural networks are considered to be best-suited to accomplish the recognition
of harmonic source.

2.2.1 Pattern Recognition Using Multilayer Perceptron Based NN

The pattern recognition process was carried out using MLP-based artificial NNs from
various harmonic datasets extracted from the process as depicted in earlier steps.
The configuration of MLP was selected that incorporated an input layer and output
layer with intervening middle layer. The parameters, i.e., number of neurons and
hidden layers, are chosen to best describe the mapping problem. The number of
harmonic considered was taken as the number of input layer neurons as a natural
Table 4 Final dataset considering all possible combinations of all devices
No Source A B C D A, B A, C A, D B, C B, D C, D A, B, C A, B, D A, C, D B, C, D A, B, C, D
0.0001 0.02 0.01 0.02 0.01 0.03 0.04 0.03 0.03 0.02 0.03 0.05 0.04 0.05 0.04 0.06
0.0001 21.711 −15.15 43.206 −44.36 6.56 64.92 −22.65 22.06 −59.51 −1.152 49.77 −37.8 20.56 −16.3 5.411
0.0001 −25.62 −21.69 −48.58 328.38 −47.32 −74.2 302.8 −70.3 306.7 279.8 −95.9 281.1 254.2 258.2 232.5
0.0001 11.18 −3.25 34.3 −301.8 7.94 45.48 −290.7 31.05 −305.1 −267.5 42.2 −293.9 −256.3 −270.8 −259.6
0.0001 0.926 0.0001 −23.9 239.7 0.927 −22.97 240.64 −23.9 239.7 215.8 −22.97 240.64 216.74 215.81 216.74
Artificial Neural Network-Based Source Identification …

0.0001 −12.25 4.33 18.19 −1.483 −7.92 5.945 −13.73 22.53 2.85 16.72 10.27 −9.41 4.46 21.04 8.79
0.0001 −13.99 0.0001 −12.81 73.83 −13.99 −26.7 59.86 −12.8 73.8 61.03 −26.79 59.85 47.04 61.03 47.04
0.0001 −6.6 0.0001 6.92 −1.33 −6.598 0.33 −7.93 6.92 −1.33 5.596 0.325 −7.93 −1.01 5.59 −1.01
55
56 A. Tayal et al.

Fig. 3 Fluke 41 power quality analyzer data for drive panel, monitor, UPS, and fluorescent light

choice. Since the problem of harmonic source device detection is not more involved,
thus only one hidden layer is chosen. However, the number of neurons in the hidden
layers was chosen on the basis of hit and trial approach. The simulations were carried
out for the number of hidden layer neurons chosen from 4 to 60 and results with 20
hidden neurons were found to suffice, since there is always a tradeoff between the
mapping accuracy and the complexity of ANN. Similar to case of input layer neurons,
the number of output layer neurons was chosen to be equal to the number of harmonic
source devices and thus was fixed at 4. Finally, the signum function was used at the
output level so as to convert all positive values to two levels: High, i.e., +1, refers
to inference that corresponding device is present and low, i.e., −1, when the same is
absent.
Artificial Neural Network-Based Source Identification … 57

2.3 Traınıng and Performance Evaluatıon

The complete dataset was divided into two parts: 67% of the total data was utilized for
training the MLP architecture as well as validating its learnt behavior, while the rest
of 33% data was utilized for prediction purpose, i.e., whether the MLP has achieved
generalization and is able to correctly identify the device responsible for harmonic
degradations.

3 Simulation Results and Conclusıon

In this work, a unique method utilizing MLP with optimized hidden layer neurons is
proposed for predicting the device responsible for harmonic degradation in consid-
ered electrical setup via study of current waveforms at point of supply. Figure 4 shows
the plot of mean square of error with number of epochs using backpropagation algo-
rithm. An error signal is the difference between the required output and ANN output.
The results showed that BP algorithm exhibits great accuracy to identify the devices
based on their harmonic signatures.
With this research, authors have presented a novel way of offline-based harmonic
source identification in an electrical setup. In future, research may be done for online
signature recognition using big data and cloud computing. Research may also be
focused upon automatic detection of devices using artificial intelligence and machine
learning.

Learning Curve
40

35

30
Mean Square Error ----->

25

20

15

10

0
0 50 100 150 200 250 300 350 400 450
No. of Epochs ----->

Fig. 4 Plot of mean square error with number of epochs using backpropagation algorithm
58 A. Tayal et al.

References

1. Arrillaga J (1997) Power system harmonic analysis. Wiley, New York

2. Hartana RK, Richards GG (1990) Harmonic source monitoring and identification using neural
network. IEEE Trans Power Syst 5(4):1098–1104
3. Lin W-M, Lin C-H, Tu K-P, Wu C-H (2005) Multiple harmonic source detection and equipment
identification with cascade correlation network. IEEE Trans Power Syst 20(3), July 2005, pp
2166–2173
4. Srinivasan D, Ng WS, Liew AC (2006) Neural-network-based signature recognition for
harmonic source identification. IEEE Trans Power Delivery 21(1):398–405
5. Fernandes RAS, da Silva IN, Oleskovicz M (2011) Data mining applied to harmonic current
sources identification in residential consumers. IEEE Latin Am Trans 9(3):302–310
6. Heydt GT (1989) Identification of harmonic sources by a state estimation technique. IEEE
Trans Power Deliv 4(1), 569–576
7. Saxena D, Bhaumik S, Singh SN (2014) Identification of multiple harmonic sources in
power system using optimally placed voltage measurement devices. IEEE Trans Ind Electron
61(5):2483–2492
8. Jain P, Tiwari AK, Jain SK (2017) Harmonic source identification in distribution system using
ESPRIT-THP method. Trans Inst Measur Control. https://doi.org/10.1177/0142331217721316
9. Safargholi F, Malekian K, Schufft W (2017a) On the dominant harmonic source identification—
Part I: review of methods. IEEE Trans Power Deliv. https://doi.org/10.1109/TPWRD.2017.275
1663
10. Safargholi F, Malekian K, Schufft W (2017b) On the dominant harmonic source identification—
Part II: application and interpretation of methods. IEEE Trans Power Deliv. https://doi.org/10.
1109/TPWRD.2017.2751673
11. Pomilio JA, Deckmann SM (2007) Characterization and compensation of harmonics and
reactive power of residential and commercial loads. IEEE Trans Power Deliv 22(2):1049–1055
Impacts of Distributed Generation
on Distribution System Based
on the Backward and Forward Sweep
Method

Indubhushan Kumar and Sandeep Gupta

Abstract A conventionally large portion of the generation is derived from fossil fuel,
i.e., coal, gas, nuclear, large hydropower plants. However, due to increased demand
growth, higher T&D cost, fast depletion of fossil-fuels, heightened environmental
concerns, utility generation paradigm is shifting from large central-station stations
to various small scale generation units. These new technologies allow the generated
electricity in small-sized plants known as Distributed Generation. The objective of
this paper is to analyze the impact of DG in distribution system mainly on power
losses, voltage profile. A MATLAB programming is developed based on the back-
ward and forward sweep method using the data of 33-IEEE Bus 12.66 kV, distribution
test system for optimal placement of Distributed Generation. The technique further
tested on 33-IEEE bus systems to show the impacts of DG.

Keywords Distributed generation · Power quality · Decentralized energy

systems · Power mitigation

1 Introduction

Distributed generation category can be based on the non-renewable as well as

renewable energy-based technologies. Non-renewable based technologies often
include reciprocating engine, combustion gas turbine, fuel cell, micro-turbine, micro
combined heat and power (CHP), etc.; while renewable-based technologies include
a wind turbine, solar photovoltaic systems, biomass gasification, geothermal, small
hydro power, etc. As per the Electric Power Research Institute (EPRI), generation
from a few kilowatts up to 50 MW can be considered as distributed generation [1, 2]. It
is suggested by International Energy Agency (IEA) that a generating plant severing

I. Kumar · S. Gupta (B)

Department of Electrical Engineering, JECRC University, Jaipur, Rajasthan, India
e-mail: [email protected]
I. Kumar
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 59

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_7
60 I. Kumar and S. Gupta

Table 1 Ranges of
Types of distributed generation Range
distributed generation
technology [5] Micro 1 to <5 kW
Small 5 to <5 MW
Medium 5 to <50 MW
Large 50–300 MW

a customer on-site or providing support to the distribution network, connected to

a synchronous grid at distribution level voltages can be considered as Distributed
Generation [3, 4]. After consideration of the technological constraints, the range of
this generation technology can be shown in Table 1.
The literature review presented is best to analyze the DG impacts on Power Loss,
voltage profile improvement as well as power quality and reliability improvement
which decides its optimal placement [6]. Chun-Lien Su [7] has proposed a new prob-
abilistic methodology to provide reliable voltage evaluation result on the connection
of DG on a distributed network. This algorithm is sufficient to handle uncertainties
in changing loads along with its output power and voltage control devices operation.
Fernando Bastião et al. [8] has analyzed voltage profile and gradient on the basis of
power in-feed/out-feed of DG. Atwa et al. [9] has investigated the applicability of
a connected radial distribution system during Grid Connected and Islanding mode
for different types of DG units. Noroozian et al. [10] has investigated the effects
of Distributed generation resources with optimal capacity inappropriate parts of the
distribution network using Cuckoo optimization algorithm (COA). Ochoa et al. [11]
has worked on adopting passive but better power factor and substation settings as
well as Smart Grid control for the reduction of the reactive support for DG to achieve
power quality.

2 Research Methodology

2.1 Network Confıguratıon

The distribution system is fed only at one point and configuration of the system may
be radial or mesh. The branches of distribution network have a wide range of R and X
values and also R/X ratio is higher compared to the transmission system. The radial
configuration has only one power source for a group of customers. A power failure,
short-circuit, or a downed power line would interrupt power in the entire line. If N
is the number of buses in system, N B is the number of branches, then for a radial
system:

N = NB + 1 (2.1)
Impacts of Distributed Generation on Distribution System Based on the Backward … 61

Mesh network is configured with different loops through the service area and
returns to the original point. Putting switches in strategic locations, the utility can
supply power to the customer from either direction. If one source of power fails,
switches are thrown (automatically or manually), and power can be fed to customers
from the other source. A loop in a mess can be opened by adding a dummy or fictitious
bus. This breaking point is called loop breaking point (LBP).
For this configuration, N < N B , i.e., the number of loop in mess network is given
as:

NLP = NB − N + 1 (2.2)

where N LP = Loop Breaking Point.

In the network configuration, branch numbering, as well as ordering of branch,
is essential for the distribution system [12]. The branch numbering process of a
particular network requires the construction of a tree of the network. The branch
numbering is starting from the first layer of the tree or can be started from any layer
only after numbering all the branches in the previous layer. The ordering of buses
of a tree allows the calculation of power injection at Loop Breaking Point (LBP) in
a simplified way. This configuration is analyzed to enhance efficiency, the balance
of load, and maintain proper voltage profile in the distribution system so that the
reliability of the system can be increased.

2.2 Load Flow Equatıons

A Steady-state condition of a system is essential for proper operation of a distribution

system for various load demands. Load flow technique is widely used to find this
steady-state operating condition. So it is necessary to reconfigure the system or use a
corrective approach to converge the solution for the secure and reliable distribution
system. For a single-source network, load flowequations can be solved iteratively
using recursive equations [12, 13]. Consider a model of the branch of a network
is shown below in Fig. 1.

3 Results and Dıscussıon

The proposed load flow method is tested with various IEEE bus systems with or
without the integration of DG at a different location and of different ratings. A
MATLAB programmed is developed based on the backward and forward sweep
method using the data of IEEE given.
62 I. Kumar and S. Gupta


Fig. 1 model branch of a network [14]

3.1 33-Buses Network, 12.66 kV System:

The schematic diagram of a 12.66 kV, 33-bus distribution test system is illustrated.
This system has 5 tie lines with voltage at root bus is considered to be 1.0 p.u. with
tolerance 0.0001. Simulation has been done with assumed initial power injection at
all bus to be zero and the integration of DG has tested.

3.2 Sımulatıon Results

To analyze the impacts of DG on the distribution system, the allocation of DG is an

important issue otherwise negative impact shows the adverse impact on the effect on
the overall distribution system. To find the DG impacts on 33-IEEE bus distribution
system for voltage profile as well as for loss minimization, same rating of DG has
tested by allocating at various bus systems as shown in Fig. 2. This allocation is done
with the help of sensitive buses to voltage (low voltage magnitude) and ranked as
per voltage interval as well as minimum loss bus as shown in Table 2. It has done

Fig. 2. Single line diagram of 33-bus distribution network

Impacts of Distributed Generation on Distribution System Based on the Backward … 63

Table 2 Voltages of buses ranked from minimum to maximum

Before DG allocation After DG allocation
Bus no Voltage (p.u) Rank Bus no Voltage (p.u) Rank
18, 17, 16, 15 0.9064–0.9119 1 33, 32, 31, 30 0.9223–0.9277 1
14, 33, 32, 0.9120–0.9170 2 29, 18, 17 0.9278–0.9341 2
13, 31
30, 12, 29, 11 0.9171–0.9253 3 16, 15, 14, 13 0.9342–0.9369 3
10, 9, 28, 8 0.9254–0.9400 4 28, 12, 11 0.9370–0.940 4
27, 7, 26 0.9401–0.9469 5 10, 9, 8, 27 0.941–0.9510 5
6, 5, 25, 24 0.9470–0.9724 6 7, 26, 6, 25, 5 0.9511–0.9710 6
4, 23, 3 0.9725–0.9827 7 24, 4, 23 0.9711–0.9810 7
22, 21 0.9827–0.9921 8 3, 22, 21 0.9811–0.9924 8
20, 19, 2, 1 0.9922–1 9 20, 19, 2, 1 0.9925–1.0000 9

with DG size of 100 kW.

3.2.1 Case 1: Effect on Voltage Profile

This effect has stimulated by allocating the DG to the most voltage-sensitive bus to
know a change in voltage profile as shown in Table 3.
This can be easily analyzed with the graphical representation given below for both
the cases as shown in Fig. 3.

3.2.2 Case 2: Effect on Power Losses

After placing of DG on the voltage-sensitive bus, it is observed that reactive as well

as the active power of system has reduced as shown in Table 4.
From Table 4, It is observed that after placing the DG at different bus viz. 15, 16,
17, 18; Bus number 17 is the lowest active and reactive power compare to other buses.
This lowest loss provides additional information for allocation of buses optimally.
Thus DG allocation on this bus provides a portion of active and reactive power locally,
thus helping to decrease current along a section of distribution line, which in turn,
results in a boost in voltage magnitude at the customer site.
It is evident that the base case (without DG) total active and reactive power losses
are 0.213 and 0.143 p.u., whereas these losses for DG size of 0.1 MVA at unity power
factor results in the reduction of active and reactive losses. Also, active and reactive
power generation is changed with integration of DG. This loss minimization helps in
improvement of voltage profile discussed above in Figs. 4 and 5, respectively. Table
5 shows the comparison of losses and generation.
64 I. Kumar and S. Gupta

Table 3 Voltage magnitude at different bus

Bus no. Without DG With DG Bus no. Without DG With DG
1 1 1 18 0.906492 0.934021
2 0.997001 0.99727 19 0.996472 0.996742
3 0.982738 0.984449 20 0.992895 0.993165
4 0.975132 0.977909 21 0.99219 0.992461
5 0.967605 0.971493 22 0.991553 0.991824
6 0.948912 0.955241 23 0.979152 0.980869
7 0.945271 0.952194 24 0.972479 0.974208
8 0.939917 0.948844 25 0.969153 0.970889
9 0.932899 0.94475 26 0.946981 0.953324
10 0.926347 0.94114 27 0.944415 0.950776
11 0.92535 0.94069 28 0.932966 0.939408
12 0.923584 0.939968 29 0.924741 0.931241
13 0.916415 0.936919 30 0.92118 0.927706
14 0.913757 0.935785 31 0.917016 0.923571
15 0.911927 0.935587 32 0.9161 0.922661
16 0.910035 0.935245 33 0.915816 0.922379
17 0.907104 0.934125

Fig. 3 Voltage magnitude versus bus number

4 Conclusıon

The paper is concluded that the location and size of DG are crucial factors in the
application of DG for loss minimization and voltage profile respectively. The instal-
lation of DG units at non-optimal places can result in an increase in system loss;
implying an increase in costs and resulting low or overvoltage in the network, having
an effect opposite to desired. In this project work, the only optimal location of DG has
determined with comparative approach without considering the size. The size of DG
affects losses and voltage profile of the distribution system. This optimal allocation
Table 4 Active power loss (p.u) and reactive power loss (p.u)
Without DG With DG Without DG With DG
Branch no Active power Reactive Active power Reactive Branch no. Active power Reactive Active power Reactive
(pu) power (pu) (pu) power (pu) (pu) power (pu) (pu) power (pu)
1 0.012493 0.006368 0.010455 0.005329 18 0.000161 0.000154 0.000161 0.000154
2 0.052985 0.026987 0.04342 0.022115 19 0.000832 0.00075 0.000832 0.000749
3 0.020523 0.010452 0.015685 0.007988 20 0.000101 0.000118 0.000101 0.000118
4 0.019313 0.009837 0.014558 0.007415 21 0.000044 0.000058 0.000044 0.000058
5 0.039534 0.034127 0.029616 0.025566 22 0.003183 0.002175 0.003172 0.002167
6 0.002243 0.007414 0.001128 0.003729 23 0.005146 0.004063 0.005128 0.004049
7 0.005865 0.001938 0.002497 0.000825 24 0.001288 0.001008 0.001283 0.001004
8 0.005347 0.003842 0.001728 0.001242 25 0.002605 0.001327 0.002569 0.001308
9 0.004628 0.003293 0.001354 0.000963 26 0.003335 0.001698 0.003288 0.001674
10 0.000737 0.000244 0.000192 0.000064 27 0.01132 0.00998 0.011161 0.00984
11 0.001208 0.000399 0.000271 0.00009 28 0.007846 0.006836 0.007736 0.006739
12 0.003809 0.002997 0.000676 0.000532 29 0.003902 0.001988 0.003847 0.00196
13 0.001099 0.001447 0.000143 0.000188 30 0.001596 0.001578 0.001574 0.001555
14 0.000648 0.000577 0.000038 0.000034 31 0.000214 0.000249 0.000211 0.000245
15 0.00058 0.000423 0.000035 0.000025 32 0.000013 0.000021 0.000013 0.00002
16 0.000646 0.000863 0.000056 0.000075 33 0.000010 0.000019 0.000010 0.00001
Impacts of Distributed Generation on Distribution System Based on the Backward …

17 0.000054 0.000042 0.000009 0.000007

65
66 I. Kumar and S. Gupta

Fig. 4 Active power (in p.u.) versus bus number

Fig. 5 Reactive power versus branch number

Table 5 Comparison of
Type Without DG With DG
losses and generation
Active power loss (p.u) 0.213299 0.162979
Reactive power loss (p.u) 0.143252 0.107029
Total active generation (p.u) 3.968299 3.517979
Total reactive generation (p.u) 2.443252 2.407829
Impacts of Distributed Generation on Distribution System Based on the Backward … 67

and sizing can be done with different optimization techniques like genetic algorithm,
optimum power flow method, fuzzy logic, etc.

References

1. Lamarre L (1993) The vision of distributed generation. EPRI J 18(3):6–18

2. Gupta S, Sharma A (2018) Global scenario of solar photovoltaic (SPV) materials. In: Inter-
national conference on advanced computational and communication paradigms (ICACCP),
Lecture notes in electrical engineering (LNEE), 475, pp 126–133. Springer, Berlin
3. IEA-2002 (2002) Distributed generation in liberalized electricity markets. Paris, p 128
4. Gupta S, Kumar I (2019) Steady state performance of DFIG with stability assessment in wind
system. In: International conference on manufacturing, advance computing, renewable energy
and communication (MARC-2018), Lecture notes in electrical engineering (LNEE), vol 553.
Springer, Berlin
5. Ackermann T, Andersson G, Sodder L (2001) Distributed generation: a definition. Elect Power
Syst Res 57:195–204
6. Khatod DK, Pant V, Sharma J (2012) Evolutionary programming based optimal placement of
renewable distributed generators. IEEE Trans Power Syst 28(2):683–695
7. Su C-L (2010) Stochastic evaluation of voltages in distribution networks with distributed
generation using detailed distribution operation models. IEEE Trans Power Syst 25(2), 786–795
8. Bastiao F, Cruz P, Fiteiro R (2008) Impact of distributed generation on distribution networks.
In: IEEE 5th international conference on the European electricity market, pp 1–6
9. Atwa YM et al (2011) Adequacy evaluation of distribution system including wind/solar DG
during different modes of operation. IEEE Trans Power Syst 26(4):1945–1952
10. Noroozian R, Molaei S (2012) Determining the optimal placement and capacity of DG in intel-
ligent distribution networks under uncertainty demands by COA. In: IEEE Iranian conference
on smart grids, pp 1–8
11. Ochoa LF, Keane A, Harrison GP (2011) Minimizing the reactive support for distributed gener-
ation: enhanced passive operation and smart distribution networks. IEEE Trans Power Syst
26(4), 3134–3142
12. Kathori DP, Nagrath V (2010) Modern power system analysis. TMH Publication. ISBN(13):
978-0-07-107775-0
13. Rajkumar Viral N, Khatod DK (2012) Optimal planning of distributed generation systems in
distribution system: a review. Renew Sustain Energy Rev 16(7):5146–5165
14. CIGRE (1998) Impact of increasing contribution of dispersed generation on the power system.
CIGRE Study Committee no. 37, Final Report, Sept 1998
Cost Analysis of Multi-rotor Wind
Turbines

Navjot Singh Sandhu and Saurabh Chanana

Abstract Power production of a wind turbine is dependent upon its rotor size, and at
present, wind turbines with large rotor diameter are available in the market. But, the
major problems associated with such large size conventional turbines are their cost
and noise pollution. Due to these reason, researchers have diverted their attention
towards lower sized equivalent multi-rotor wind turbines. Keeping it in view, in this
paper, an effort has been made to compare the cost of conventional turbine with
equivalent multi-rotor wind turbines. A new cost model as proposed has been used
for the cost comparison. Results as presented justify the suitability of multi-rotor
wind turbine in place of single-rotor configuration.

Keywords Multi-rotor wind turbine · Wind energy · Wind turbine · Wind turbine
cost

1 Introduction

Mostly fossil fuels are used for the power generation, and such resources are likely
to vanish with time. In addition, these are also resulting in to pollution and global
warming which are harmful to society. In order to reduce such harmful effects,
scientists are exploring renewable sources of energy [1] such as wind power, solar
power and tidal power. Out of these sources, wind power is the fastest-growing
resource. In the past few years, there was a tremendous growth of grid-connected
wind power capacity worldwide. However, cost of energy is the major concern in
any renewable power generation. Cost comparison of wind power generation with

N. S. Sandhu (B)
School of Renewable Energy and Efficiency, National Insitute of Technology, Kurukshetra,
Haryana 136119, India
e-mail: [email protected]
S. Chanana
Department of Electrical Engineering, National Insitute of Technology, Kurukshetra, Haryana
136119, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 69

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_8
70 N. S. Sandhu and S. Chanana

others resources was taken by [2, 3]. The development progress [4, 5] in the industry
of wind turbine over the last decade and the effect of rotor diameter on normalized
cost of energy per unit area was discussed by Musgrove [5]. Researchers [6] tried
to investigate the wind power impact on the price of electricity by the help of a
cost model as proposed. It is also found that the electricity price increases under the
effect of over-forecasting and decreases under the effect of under-forecasting. It was
observed that the cost of energy (COE) was computed by the use of annual energy
production and relative percentage cost [7]. An approach known as life cycle cost
technique was presented [3], in which four cost factors, i.e., capital cost, operation and
maintenance cost, fuel cost, damage prevention, or damage cost and salvation value
were used to calculate the cost of energy. With the help of facts and figures, researcher
[8] tried to explain the progress of modern wind energy technology and the significant
reduction in the case of cost of energy (COE). Different factors as identified [9],
such as government support, financial incentives and availability of local expertise,
may be useful to develop off shore wind energy in India. New developments in
wind energy technologies like gearless direct drive generator, pitch control, tower
construction, and the impact of these developments on the energy cost of wind power
was taken up by the researcher in [10]. Authors [11] provided some information
regarding the efficiency of wind power plant and the effect of wind turbine size
over the price/performance ratio of wind power plant. Few researchers [4] tried
to illustrate the effect of annual mean wind speed on the cost of energy of wind
power plant. Effect of capacity factor on the cost of energy (COE) of a wind power
plant was presented by the research [12]. A set of equations were proposed by the
researcher [13] based upon the review of methodologies presented by other authors,
and these were used to compute the wind energy production in a given area. A
simple method as proposed may be useful to find the annualized wind generation
cost of a wind power plant. The comparison of energy captured per unit area with
different turbine technologies with varying electricity price and wind speed was
conducted by the author [14]. It was found that in the optimal financing structure,
the choice of turbine and form configuration can be drastically different for different
sites with different wind speeds. Authors [15] analysed the direct costs of wind
power generation and discussed various types of indirect costs associated with wind
power developments and also used the concept of overnight construction cost for
the calculation of cost of energy (COE). Economics analysis was presented [16] to
compare the generation cost of electricity from both wind and gas turbine systems
and computed the generation cost per unit energy using initial capital cost, annual
energy yield, capital recovery factor, fuel cost/unit energy output and ‘operation and
maintenance’ cost/unit energy output. It was observed that the generation cost of
wind energy compares favourably with that of conventional gas turbine. In addition,
the continuously diminishing capital cost of wind energy systems and falling interest
rates promise to make it even more economically viable energy resource. Some of
the researchers [17] tried to compute the cost of energy for a number of rated wind
speeds for a given annual mean wind speed and Weibull shape factor using capital
cost, charge rate and cost of operation & maintenance. Turbine rotational speed is
assumed to vary in proportion to the rated speed for simplicity. Net present worth
Cost Analysis of Multi-rotor Wind Turbines 71

method was used by the author [18] for the computation of month-wise cost of energy
(COE). Impact of many parameters such as capital cost, capacity factor, maintenance
and operational cost, and size of wind turbine were taken up [19] on the economic
viability of wind power plant. It was observed that profitability is very sensitive to
changes in capital cost, capacity factor, electricity escalation rate and first installation
cost. Cost of energy (COE) was computed by the researcher [20] in terms of capacity
factor, investment costs, labour costs and operational and maintenance costs. It was
found that solution is possible to get maximum capacity factor with low value of cost
of energy (COE). A study was included regarding the generation costs of onshore
wind energy projects in Europe by the authors and also analysed the effect of rise of
cost of raw material on the cost of energy (COE) of wind power plant.
However, so far, no attempt has been made to compute the cost of energy in case
of multi-rotor wind turbines. In this paper, an economic model has been developed
to find out the cost of multi-rotor wind turbine.

2 Cost Model

Cost of energy (COE) may be defined as:

(WTPC ∗ (1 + G)) ∗ FCR + C O&M

COE = (1)
AEP
where
COE Cost of energy.
WTPC Wind turbine project cost.
FCR Fixed charge rate.
G Gestation factor.
C O&M Operational and maintenance cost of wind power plant (land rent, insurance,
administration, miscellaneous, and maintenance).
AEP Annual energy production of wind power plant.

I ∗ (1 + I )n
FCR = (2)
(1 + I )n − 1

I = Interest rate.
n = Life span of wind turbine (in years)
 I 
G=  
12
 −t ∗ gm (3)
1 − 1 + 12I

gm Gestation period in months.

72 N. S. Sandhu and S. Chanana

Fig. 1 Flowchart for computing the project cost of multi-rotor wind turbine

t Wind turbine time span including gestation period.

t (n*12) + gm .

For the computation of cost of energy, there is a need to compute the cost compo-
nent and annual energy yield of the turbine. In case equivalent multi-rotor wind
turnine gives same annual energy yield, then the COE is totally dependent upon the
wind turbine project cost. In this section, a new cost model has been proposed to
compute the project cost of an equivalent multi-rotor wind turbine.
Figure 1 describes the process to determine various cost components of a multi-
rotor wind turbine.
C = Single-rotor wind turbine project cost (SRWTPC), it can be determined by
using the per MW data [21–23].
SRWTFC1; Single-rotor wind turbine fixed cost 1 = 0.69*C, (it includes rotor
blade, rotor hub, rotor bearing, pitch system, gearbox, tower, foundation and other
components).
SRWTFC2; Single-rotor wind turbine fixed cost 2 = 0.11*C, (it includes electrical
infrastructure, road construction and transport cost).
SRWTRC; Single-rotor wind turbine reserve cost = 0.20*C, (Required to meet
operational and maintenance cost of the project)

MRWTPC = MRWTFC1 + MRWTFC2 + MRWTRC (4)

MRWTFC1; Multi-rotor wind turbine fixed cost 1 = K 1 *SRWTFC1.

MRWTFC2; Multi-rotor wind turbine fixed cost 2 = SRWTFC2.
MRWTRC; Multi-rotor wind turbine reserve cost = K 2 *SRWTRC.
Cost Analysis of Multi-rotor Wind Turbines 73

K 1 , K 2 are the multiplying factors required to determine the multi-rotor wind

turbine project cost (MRWTPC). K 1 and K 2 are dependent upon the configuration
of multi-rotor wind turbines.

3 Results and Discussions: Case Study

Wind turbine cost of two multi-rotor wind turbines having three and four rotors wind
as determined is given in Table 1. Three identical rotors with power rating of 1 MW
are used for a three-rotor wind turbine, and these are to be replaced by 750 kW in
case of four-rotor wind turbine. Typical values of K 1 and K 2 for three-rotor wind
turbine are computed to be 0.8346 and 1.15. While for four-rotor wind turbine, these
are 0.7224 and 1.22, respectively. Project cost of single-rotor wind turbine as given
in Table 1 (4433000 USD) is considered as the reference value to start with the cost
comutation of multi-rotor wind turbine. This process is defined in Sect. 2 (Fig. 1).
Cost results as obtained for different components of wind turbines (single- as well
as multi-rotor) are tabulated in Table 1. İt is observed that:
1. There is a saving of 505824 USD in case single-rotor wind turbine is replaced
by an equivalent three-rotor wind turbine.
2. There is a saving of 849042 USD in case single-rotor wind turbine is replaced
by an equivalent four-rotor wind turbine.

Table 1 Breakdown of wind turbine components cost

Project cost of 3000 kW wind project [21–25] 4433000 USD
Cost of single-rotor wind turbine of 3000 kW 3058770 USD
Sr. no. Wind turbine Component costs Component costs Component costs
component ($) of single-rotor ($) of three-rotors ($) of four-rotors
wind turbine wind turbine wind turbine
1 Rotor blades 627,604 581,740 376,990
2 Rotor hub 102,214 141,336 140,690
3 Rotor bearing 72,780 49,308 25,174
4 Pitch system 163,477 158,611 95,108
5 Gearbox 366,272 366,272 366,272
6 Tower 726,125 508,287 508,287
7 Foundation 709,280 409,502 354,640
8 Other 291,018 291,018 291,018
components
9 Support structure – 46,872a 51,549a
(spars + cables)
Total cost 3,058,770 2,552,946 2,209,728
a Prices as per USA market
74 N. S. Sandhu and S. Chanana

3. Adoption of four-rotor turbine gives a net saving of 343218 USD if replaced in

place of three-rotor wind turbine.

4 Conclusions

Keeping in view the recent developments in terms of equivalent multi-rotor wind

turbines, there is a need to analyse their cost. In this paper, an effort has been made
to compare the cost of conventional turbine with two different equivalent multi-rotor
wind turbines. A new cost model as given in Sect. 2 has been proposed. Proposed
model can be used to compute the cost of multi-rotor wind turbines. All this needs
the cost component of conventional wind turbine and configuration of multi-rotor
wind turbine. Cost model as proposed has been used for the cost comparison. Results
as presented justify the suitability of multi-rotor wind turbine in place of single-rotor
configuration.

Acknowledgements Authors are thankful for providing the financial support to carry out this
research by “TEQIP III” of Government of India which is being implemented as a World Bank-
assisted project.

References

1. Abbaspour M, Atabi F (1997) Wind energy estimation for production of electricity in Iran. Int
J Eng 10(1):19–26
2. Kaygusuz K (2004) Wind energy: progress and potential. J Energy Sourc 26:95–105
3. Sahin DA (2004) Progress and recent trends in wind energy. Progr Energy Combust Sci 30:501–
543
4. Quraeshi S (1987) Costs and economics of wind turbine generators for electrical power
production. Solar Wind Technol 4(1):55–58 (1987)
5. Musgrove JP (1983) Wind energy conversion—an introduction. Proc A Phys Sci Measure
Instrum Manage Econ Rev 130(9):506–516
6. Martinez-Anido BA (2016) The impact of wind power on electricity prices. Renew Energy 94:
474–487
7. Fuglsang, P., Thomsen, K.: Site-Specific Design Optimization of 1.5–2.0 MW Wind Turbines.
ASME: Journal of Solar Energy Engineering, vol. 123, pp. 296 – 303 (2001).
8. Islam RM, Mekhilef S, Saidur R (2013) Progress and recent trends of wind energy technology.
Renew Sustain Energy Rev 21:456–468
9. Mani S, Dhingra T (2013) Offshore wind energy policy for India—key factors to be considered.
Energy Policy 56:672–683
10. Hoffman LD, Molinski ST (2009) How new technology development will lower wind energy
costs. In: CIGRE/IEEE PES joint symposium: integration of wide scale renewable resources
into the power delivery system, pp 1–6
11. Follings JF, Pfeiffer EA (1998) Economics of wind power plants. J Wind Eng Indus Aerodyn
27:263–274
12. Bean P, Blazquez J, Nezamuddin N (2017) Assessing the cost of renewable energy policy
options—a Spanish wind case study. Renew Energy 103:180–186
Cost Analysis of Multi-rotor Wind Turbines 75

13. Schallenberg Rodriguez J (2013) A methodological review to estimate techno economical wind
energy production. Renew Sustain Energy Rev 21:272–287
14. Atamturk A, Lai WJ, Richartz EL, Li H, Gu K (2009) An economic analysis of wind energy
harvest. In: Power and energy engineering conference APEEC. IEEE
15. Timilsina RG, Kooten VCG, Narbel AP (2013) Global wind power development: economics
and policies. Energy Policy 61:642–652
16. Marafia HA, Ashour AH (2003) Economic of off shore/on shore. Wind energy systems in
Qatar. Renew Energy 28:1953–1963
17. Xing-Jia Y, Zuo-Xia X, Lei C, Hong-Xia S (2007) Analysis of 1MW variable speed wind
turbine parameter optimal design based on cost modeling method. In: Industrial electronics
and applications ICIEA. IEEE, pp 817–821
18. Fazelpour F, Soltani N, Soltani S, Rosen AM (2015) Assessment of wind energy potential and
economics in the North Western Iranian Cities of Tabriz and Ardabil. Renew Sustain Energy
Rev 45:87–99
19. Kaldellis KJ, Gavra JT (2000) The economic viability of commercial wind plants in Greece: a
complete sensitivity analysis. Energy Policy 28:509–517
20. Dalabeeh KSA (2012) Techno economic analysis of wind power generation for selected
locations in Jordan. Renew Energy 101:1369–1378
21. http://www.renewablesfirst.co.uk
22. http://www.conserve-energy-future.com
23. http://www.local.gov.uk
24. Fingersh L, Hand M, Laxson A (2006) Wind turbine design cost and scaling model. National
Renewable Energy Laboratory
25. Blanco IM (2009) The economics of wind energy. Renew Sustain Energy Rev 13:1372–1382
Annualized Cost of Wind Power
Generation

Sahil Bajaj and K. S. Sandhu

Abstract Production of effective, efficient power keeping in mind its economical

aspect for the population is a big challenge today. For any country to progress,
economic power generation to fulfill the demands of the consumers is necessary.
Different renewable energy resources like solar, hydro, wind, etc., may be exploited
to facilitate the adequate amount of power generation for the masses. Wind energy
power generation may have an upper hand among these renewable energy sources.
Wind power generation may be classified as an effective and economical means of
power generation for the population. For producing effective and efficient energy,
wind power plant needs to be economically viable. In this paper, an attempt has been
made to determine annualized cost of generation (C g ) from wind energy. This has
been computed by selecting the location (Muppandal) kanyakumari in India. Further
the effect of life span has been seen on the annualized factor as well as the cost of
wind power generation.

Keywords Wind energy · Economics of wind energy · Annualized cost of

generation (cg ) · Economic analysis · Wind turbine

1 Introduction

Globally, there is an incline toward wind energy technology. Today, power generated
worldwide by the use of wind energy conversion system is very high. In total, as
of now, around 105 countries are fulfilling their power needs using wind energy
technology. 600 GW of wind energy is produced worldwide as of 2019 [1–5]. As
there is increase in the demand of energy needed, the establishment of more and more

S. Bajaj (B)
School of Renewable Energy and Efficiency, NIT Kurukshetra, Kurukshetra, Haryana, India
e-mail: [email protected]
K. S. Sandhu
Department of Electrical Engineering, National Insitute of Technology Kurukshetra, Kurukshetra,
Haryana 136119, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 77

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_9
78 S. Bajaj and K. Sandhu

economically viable wind power plants should be encouraged. Wind power produced
using these plants should be consumer friendly and should be economically viable
without compromising the quality of energy derived. Economically, sound wind
power plants will help in meeting the energy demands of the future.
Many studies and research work have been performed to explore wind energy
economics so that wind energy can economically be incorporated in the clean and
efficient production of electricity. It can easily be inferred that wind energy economics
include constituting factors that can either be controlled or can be kept in mind while
establishing a wind power generating system[6] so as to produce an economically
viable source of electricity. Milborrow [7] studied the economics of wind power
and its comparison with conventional thermal plant. It includes the expected prices
and of wind energy up to year 2000. Environmental, social, and economic aspects
of wind energy were studied by Surugiu et al. [8] in 2000. van den Bergh et al.
[9] explored the possible contribution of evolutionary economics to environmental
policy-making, in particular with respect to innovations in energy technology. It
can hence be seen that wind energy was found to be most economical in nature.
In 2007, a new analytical method was proposed by Joshi et al. [10] which finds
reliability, average failure rate, and the operation and maintenance cost of wind
turbine generator (WTG). In 2008, Benitez [11] reviewed the economics of wind
power with energy storage. Sangmesh et al. [12] presented a new methodology to
select a wind turbine generator from the view point of performance and economic
consideration. Wind energy had experienced dramatic growth over the past decade;
hence, ecological and economic cost-benefit analysis of offshore wind energy was
done by Snyder et al. [13] in 2009. Marıa [14] observed the capital cost and the
variable cost associated with generation of wind energy. In the same year, Alper
et al. [15] developed a novel approach that simultaneously optimizes a wind and it is
financing for an improved economic analysis of wind energy project. Varun et al. [16]
studied the economic and environmental impacts of renewable energy systems and
found out that wind and small hydro are the most sustainable source for the electricity
generation. Kabir et al. [17], again in 2011, compared life cycle energy, emission, and
economic analysis of 100 kW name plate wind power generation. Valenzuela et al.
[18], in 2011, found a probabilistic model for assessing the long-term economics
of wind energy. The assessment indicators and economic-financial management of
projects implemented for onshore wind energy systems were characterized by de
Oliveira et al. [19] in 2011. Followed by, Green et al. [20] described economics
for offshore wind energy. In 2012 [21], IRENA published a cost analysis series for
wind energy technology. Schallenberg [22] estimated the Techno-Economical factors
for wind energy production and also gave a methodology to find annualized wind
generating costs so as to assess the economics [23–35] of wind power plant. This
paper includes an economic assessment technique for wind power plant applied on a
wind site of india (muppandal in kanya kumari). For different life spans of the project.
The case study done may be fruitfull in determining more economically viable sites
for establishment of a wind power plant. The next section includes the methodology
followed for determining the economics of a wind power plant.
Annualized Cost of Wind Power Generation 79

2 Methodology for Economıc Analysıs of Wınd Power

Plants

Economical analysis of wind power plants may be done basicaly using two types of
analysis; first type is the absolute analysis that helps in determining the costs incurred
are higher or lower than the benefits achieved [36–39]. This analysis also counts for
the economic viability of the wind power plant. Second approach considered for
economic analysis of wind power plant is the relative analysis. Relative analysis
shows and determines if the benefits are higher and at what amount or costs it also
helps in ranking the project in terms of costs and benefits [40–42]. The aim of this
paper is to determine for what costs incurred in a wind power plant may help in
analyzing economic viability of it.
The wind electricity generation cost ($/kWh) can be annualized using [22]:

a ∗ I + Co&m
Cg = (1)
h eq
AEP
h eq = (2)
W Tpower

where C g is wind electricity generation cost ($/kWh) for the location, a is annuity
factor, given by:
r
a=
1 − (1 + r )−LT

where r is the interest rate, in this case 8%. LT is the lifetime (a);
I is the investment cost ($/kw). C O&M is the operation and maintenance cost
annually ($/kW/year a). heq is the annual equivalent hours at the location referred.
The above equations will be used to compute the annualized cost of electricity from
wind for two different lifespans, i.e., 20 and 25 years, respectively. With the help
of the methodology found to compute the economics of wind power plant. The
following analysis done may give a fruitful idea for economic working and viability
of wind power plant.

3 Cost of Annualized Electricity Generated from Wind

Energy

For computation of cost of electricity generated from wind energy, a case study
may be considered. A reference wind site (Muppandal in Kanyakumari) may be
considered with a annual mean wind speed of 5.88 m/s. For calculation of investment
cost, the following cost model may be considered (I cc ) [43].
80 S. Bajaj and K. Sandhu

Table 1 Specification of the

Name of the wind turbine Vestas-100
wind turbine
Power rating 1800 kW
Cut-in speed V c 4 m/s
Cut-out speed V o 20 m/s
Rated speed V r 12 m/s
Rotor diameter 100 m
Hub height 80 m
Swept area 7858 m2

Icc = 0.2856d 3 + 0.3526d 2.92 + 2.9605d 2.5 + 0.0055d 3.5 + 0.0866d 2.94
+ 0.0151d 2.887 + 0.6236d 2.65 − 0.0137d 2.5 + 2.54d 1.953
+ 947.914d 0.85 + 139.699d + 0.00002046P 3 − 0.0485P 2 +
547.1451P + 2.9756 ∗ (dh)1.1736 + 1.1357Ah + 72, 997.4398;

where I cc is the initial cost of investment for wind power plant($). I cc per Kw may
be further used for computing I ($/kw), P is the power rating of a wind power plan,
h is the hub height, d is the rotor diameter, and A is the swept area of the turbine
[44] and may be computed as π /4(d)2 . A reference wind turbine vestas 100 may be
considered with following specifications (Table 1).
For the calculation of AEP, availability and power curve of a wind turbine may
be used (Table 1.). Available wind speeds for 8760 h for the given site was collected
from reference [21].
For this computation, the value of AEP [26, 45, 46] was found using MATLAB
programming and the following power functions were also programmed for the
computation of the same.

P = 0; V < cut − in speed of wind turbine.

P = 0.5 ∗ density ∗ V 3 ∗ A ∗ Cp ; Vc < V < Vr .

P = Prated ; V > or = Vr and V < Vc .

P = 0; V > Vc .

where C p is coefficient of performance and its value may be assumed as 0.40 [21].
Density may be assumed 1.22 kg/m3 . In the next section, the results obtained using
Eqs. (1) and (2) are analyzed for economic viability of a given site with the reference
wind turbine for two life spans (20 and 25 years).
Annualized Cost of Wind Power Generation 81

Table. 2. Computed values for different characteristics of a wind power plant

Name of the referred wind site with mean Muppandal in Kanyakumari
speed
Reference wind turbine Vestas-100
Power rating of wind turbine (WT power) 1800 kW
Investment cost, I ($/kW) 1967.45
Operation and maintenance cost, C o&m 51$/kW/year
($/kW/year) [24]
Rate of interest [24] 8%
Annuity factor 0.102 (20 years) and 0.092 (25 years)
Annual energy production, AEP (kWh per 8,216,880
year)
Lifespan (years) 20 25
Annualized wind electricity generation 0.055 0.052
cost ($/kWh)

4 Results and Discussions

Table 2 shows the simulation results obtained using the wind turbine with specifica-
tions ahown in Table1. These values of cost of energy were computed for two different
lifespans that is 20 and 25 years, respectively. The results show that the project with
larger lifespan is more economically viable as compared to others. Further annuity
factor as shown varies with the life span of the wind turbines.

5 Conclusions

This paper includes the discussions on various factors which affect the economic
viability of wind power plant also the various costs incurred in the establishment of
wind power plant have been highlighted. It is also seen that the method for compu-
tation of annualized cost of generation (C g ) from wind energy is very beneficial for
computing the economics of a wind power plant. As this is the cost incurred annually
for a wind power plant. This computation may lead to a viable economic compar-
ison between different sites based on different wind turbines, lifetimes, and interest
rates. The paper may be used for fruitful economic analysis of a wind power plant.
In the above presented study, C g for different lifespans is computed and it is seen
that the project with a larger life span is more economically viable as compared to
other because of high investment costs. It is also seen that there may be a decrease
in the cost computed if the wind turbine selected is of less power rating or the site
selected may be having a higher mean wind speed. In any case, the methodology
may be useful in computing the economics of a wind power plant. Hence, the above
82 S. Bajaj and K. Sandhu

presented important entities may be useful in establishment of an economically viable

wind power plant.

Acknowledgements Authors would like to thank SCHOOL OF RENEWABLE ENERGY AND

EFFICIENCY, NIT KURUKSHETRA for successful completion of this work

References

1. Boyle G (2010) Renewable energy, 2nd edn. Oxford University Press

2. Zobaa AF, Bansal RC (2011) Handbook of renewable energy technology. World Scientific
Publishing Co. Ltd.
3. Fugslang P et al. (1998) cost optimization of wind turbines for large scale offshore wind farms.
Riso National Laboratory, Roskilde
4. Fugslang P et al (2001) Site specific design optimization of 1.5–2 Mw wind turbines. ASME
123
5. Sousa de Oliveira W et al (2012) Optimization model for economic evaluation of wind farms—
how to optimize a wind energy project economically and technically. Int J Energy Econ Policy
2(1):10–20
6. Negam HM et al (2000) Structural design optimization of wind turbine towers. Pergamon,
Comput Struct, 649–666
7. Milborrow D (1994) Economics of wind power and comparisons with conventional thermal
plant. Renew Energy 5(1):692–699
8. Surugiu L, Paraschivoiu I (2000) Environmental, social and economic aspects of wind energy.
The American institute of aeronautics and astronatltics, Inc.
9. Jeroen J, van den Bergh CJM, Faber A, Idenburg AM, Oosterhuis FH (2006) Survival of the
greenest: evolutionary economics and policies for energy innovation. Environ Sci 3(1):57–71
10. Joshi DR, Jangamashetti SH (2007) A new method to estimate the o&m costs for the financial
planning of the wind power projects based on wind speed. In: lET-UK international confer-
ence on information and communication technology in electrical sciences ICTES. Dr. MG.R.
University, Chennai, Tamil Nadu, India. 20–22 Dec, 45–49
11. Benitez, LE, Benitez PC, Cornelis van Kooten G (2008) The economics of wind power with
energy storage. Energy Econ 30:1973–1989
12. Soddamani SS et al (2008) Economic index for selection of wind turbine generator at site.
IEEE international conference on sustainable energy technologies
13. Snyder B et al (2009) Ecological and economic cost-benefit analysis of offshore wind energy.
Renew Energy 34:1567–1578
14. Blanco MI (2009) Renewable and sustainable energy. Reviews 13:1372–1382
15. Atamturk A (2009) An economic analysis of wind energy harvest. 978-1-4244-2487-0/09.
IEEE
16. Varun et al (2009) Energy, economics and environmental impacts of renewable energy systems.
Renew Sustain Energy Rev 13:2716–2721
17. Kabir MR et al (2011) comparative life cycle energy, emission, and economic analysis of 100
kW nameplate wind power generation. Renew Energy 37:133–141
18. Valenzuela J et al (2011) A probabilistic model for assessing the long-term economics of wind
energy. Electric Power Syst Res 81:853–861
19. Sousa de Oliveira W et al (2011) Economic evaluation applied to wind energy projects.
Multidisc J Sci Technol J Sel Areas Renew Sustain Energy (JRSE)
20. Green R et al (2012) The economics of offshore wind. Energy Policy 39:496–502, 2011.
Renewable energy technology. Cost analysis series. IRENA, Volume: Power Sector, Issue 5/5
21. Ahilian T (2012) Forecast and performance of wind turbines. Am J Appl Sci 9(2):168–176
Annualized Cost of Wind Power Generation 83

22. Rodriguez JS (2013) A methodological review to estimate techno-economical wind energy

production. Renew Sustain Energy Rev 21:272–287
23. Kumar R, Gupta RA, Bansal AK (2013) Economic analysis and power management of a
stand- alone wind/photovoltaic hybrid energy system using biogeography based optimization
algorithm. Swarm Evol Comput 8:33–43
24. Mone C et al (2014) Cost of wind energy review. National Renewable Energy laboratory
25. Cheng M, Zhu Y (2014) The state of the art of wind energy conversion systems and technologies:
A review. Energy Convers Manage 88:332–347
26. Mani S et al (2013) Offshore wind energy policy for India—key factors to be considered.
Energy Policy 56:672–683
27. Shannon L (2013) Ferrell et al Decommissioning wind energy projects: an economic and
political analysis. Energy Policy 53:105–113
28. Sesto E (1999) Wind energy in the world: reality and prospects. Renew Energy 16:888–893
29. Sahin AD (2004) Progress and recent trends in wind energy. Prog Energy Combust Sci 30:501–
543
30. Energy KKW (2004) Progress and potential. Energy Sourc 26(2):95–105
31. Kennedy S (2005) Wind power planning: assessing long term costs and benefits. Energy Policy
33:1661–1675
32. Kalledis JK et al (2000) The economic viability of commercial wind plants in Greece: a
complete sensitivity analysis. Energy Policy 28:509–517
33. Afanasyeva S et al (2016) Technical, economic and uncertainty modelling of a wind farm
project. Energy Convers Manage 107:22–33
34. Kaldellis JK, Gavras TJ (2000) The economic viability of commercial wind plants in Greece
a complete sensitivity analysis. Energy Policy 28:509–517
35. Collecut GR, Flay RGJ (1996) The economic optimization of horizontal axis wind turbine
design. J Wind Eng Ind Aerodyn 61:87–97
36. Bajaj S, Sandhu KS (2014) Wind turbine economics: a study. In: 6th IEEE India international
conference on power electronics (IICPE-2014), National Institute of Technology Kurukshetra,
8–10 Dec 2014
37. Bajaj S, Sandhu KS (2014) Matlab/simulink modeling for cost estimation of wind turbine.
National conference on Nano materials and instrumentation. National Institute of Technology
Kurukshetra, 9–10 Mar 2014
38. Bajaj S, Sandhu KS (2014) Economic analysis of wind turbine using new cost model. In:
The 2014 international conference on power system and engineering. Interlaken ,Switzerland,
22–24 Feb
39. Bajaj S, Sandhu KS (2014) Economic analysis of wind power plant student conference on
engineering sciences. MNIT Allahabad 28th–30th May 2014
40. Varela-Vazques P et al (2015) Socioeconomic impact of wind energy on peripheral regions.
Renew Sustain Energy Rev 50:982–990
41. Harrison R, Hau E, Snel H (2000) Large wind turbines. Design and economics. Baffins Lane
Wiley, Chichester
42. Gonzales JS, Rodriguez AGG, Mora JC, Santos JR, Payan MB (2010) Optimization of wind
farm turbines layout using an evolutive algorithm. Renew Energy 35:1671–1681
43. Muche T et al (2016) Economically optimal configuration of onshore horizontal axis wind
turbines. Renew Energy 90:469–480
44. Yao X–J et al (2007) Analysis of 1 MW variable speed wind turbine parameter optimal design
based on cost modelling method. In: Second IEEE conference on industrial electronics and
applications 2007
45. Kothari DP (2000) Renewable energy scenario in India. In: IEEE Power Engineering Society
Winter Meeting, pp 634–636
46. Ajao KR, Mahamood MR (2009) Wind energy conversion system: the past, the present and
the prospect. J Am Sci, 17–22
A New Technique for Wind Turbine
Power Curve Incomplete Bin Evaluation

Bhukya Ramdas, M. Saravanan, J. C. David Solomon, and K. Balaraman

Abstract The power curve is the performance indicator of the wind turbine. The
current industry standard warranted by the International Electrotechnical Commis-
sion (IEC) 61400-12-1 (IEC 61400-12-1: first edition 2005-12, wind turbines—part
12-1: power performance measurements of electricity producing wind turbines [1])
defines a unique procedure to evaluate a wind turbine’s power curve (for both stall and
pitch regulated wind turbine). The IEC 61400-12-1 (IEC 61400-12-1: first edition
2005-12, wind turbines—part 12-1: power performance measurements of electricity
producing wind turbines [1]) standard gives a leverage to estimate any incomplete bin
of power curve from two adjacent complete bins using linear interpolation method
whenever the incomplete bin prevents the completion of test due to data insuffi-
ciency. The incomplete bin interpolation approach could add additional uncertainty
in the power performance assessment due to the nonlinear power performance char-
acteristics of wind turbines. Especially in stall regulated wind turbine, the approach
over estimates the power in incomplete bin about 32% and leading to 1.2% over
estimation of annual energy production (AEP). This paper proposes to use scaled
down time averaging window data sets instead of traditional interpolation method to
evaluate the incomplete bins of IEC power curve by employing the statically strong
data points. In this study, thirty second average data sets have been used to evaluate
the incomplete bin. The technique has been implemented in both stall and pitch regu-
lated wind turbines power curve data bases and found that the results are reasonable
and realistic. The proposed technique improves the power estimation by 30.4% and
the AEP by 1.1% in a stall regulated wind turbine when compare to interpolation

B. Ramdas (B) · M. Saravanan · J. C. D. Solomon · K. Balaraman

National Institute of Wind Energy (NIWE), Chennai, Tamil Nadu, India
e-mail: [email protected]
M. Saravanan
e-mail: [email protected]
J. C. D. Solomon
e-mail: [email protected]
K. Balaraman
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 85

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_10
86 B. Ramdas et al.

method. The paper strives to demonstrate that the incomplete bins of IEC database
can be filled with the data sets typically averaged over thirty seconds to derive actual
IEC power curve.

1 Introduction

The power curve is the relation between the wind speed (V ) available across the wind
turbine’s rotor and the electrical power produced by the wind turbine (P(v)). P(v) is
proportional to V 3 , and the nature of the relationship between P(v) and V 3 is nonlinear.
The emphasis of the nonlinearity can be observed when the turbine operates in the
cut-in to the grid and power regulation of portions near to the rated wind speed.
The inequality factor between V 3 and P(v), turbine power control strategies and
turbulence [2] associated with time histories create the nonlinear relationship between
each bin of the power curve. As per the IEC 61400-12-1 [1], each bin requires at least
three data points (10 min average values), otherwise the bin treated as incomplete and
leads to linear interpolation. The power estimated in an incomplete bin using linear
interpolation method does not hold realistic as it is evaluated from two adjacent bins
where the nonlinear relationship exists between them. Especially in stall regulated
wind turbine, the method over estimates the power about 32% (near rated wind speed).
Table 1 presents the difference between the actual and interpolated bin results of stall
regulated wind turbine. Figure 1 presents the typical power curve of stall regulated

Table 1 IEC power curve bin and interpolation results comparison in stall regulated wind turbine
power curve
Actual power Interpolation results Difference (%)
Wind speed (P.U) Power (P.U) Wind speed (P.U) Power (P.U)
0.96 1.08 0.96 1.44 32

Fig. 1 In complete bin (in black circle) of stall regulated wind turbine at near power regulation
A New Technique for Wind Turbine Power Curve Incomplete Bin … 87

wind turbine near rated wind speed. Over estimation of power is leading to higher
annual energy estimation of 1.2% at 11 m/s annual average wind speed.
The inadequate measurement in certain bins leads to power curve bin interpolation
and the factors, which limit the capture of adequate data base as per IEC 61400-12-1
[1] standard in India have been detailed below.
The wind turbine power curve measurement as per IEC 61400-12-1 [1] has great
difficulties form very first step of site selection process. 1. Site Selection: The test site
selection which has terrain slope characteristics as per IEC 61400-12-1 [1] Annexure
‘B’ is in itself a tedious task. During the pre-feasibility study,many sites are rejected
due to the irregular terrain shape. 2. Measurement Sector: The power curve measure-
ments are carried out in sites which have high wind potential. Large numbers of wind
turbines are installed at these sites, and during the measurement, only a very narrow
band of sectors would be available to record the wake free data [3]. 3. High Wind
Speed Measurement: Challenges exist on the field to capture power curve data base
which covers the wind speed ranges from at least 1 m/s below cut in wind speed to
1.5 times the wind speed at 85% of the rated power of the wind turbine (ten minute
average data sets) is produced. 4. Test Turbine Disconnection: During the measure-
ment period, many power system network feeders in India may be disconnected due
to constraints in power system operation and maintenance due to high wind power
penetration in to the electrical grid. 5. Faults: A test turbine fault, feeder faults,
feeder shutdown due to neighboring operating wind turbines faults, and measure-
ment system faults consume lot of windy season and stretches the measurement
period. All the above explained issues elevate the challenges to complete the test
within the contractual time and further impose challenges to capture the adequate
data base as per the IEC 61400-12-1 [1] standard. The inadequate measurement
strictly insists to estimate power curve of incomplete bins using linear interpolation
method.
The paper uses data from a wind turbine power curve measurements carried out on
pitch and stall regulated wind turbines. The measurements were done in accordance
with IEC-61400-12-1 [1] standard at sites in Tamil Nadu, India. The results are
evolved from the data acquired at high sampling rate about 35 Hz. For study purposes,
during measurement, the data sets have been averaged and stored for 10 min (IEC
procedure) and 30 s simultaneously. The reason for acceptance of 30 s average data
set for incomplete bin evaluation is demonstrated based on pitch regulated wind
turbine measurement for simple reason that the data base has addition signals for
ease power curve diagnostics. Later, the technique tested with stall regulated wind
turbine data base. The power curve assessment requires very accurate wind speed,
power measurements. The measurement procedure and instrumentation technique
adopted for the research are detailed below.
88 B. Ramdas et al.

2 Measurement

The measurements was carried out in a sites which met the requirements of
IEC 61400-12-1 [1] standard Annexure ‘B’. Figure 2 illustrates the wind speed
distribution with respect to wind direction at site (pitch regulated wind turbine).
The predominant wind direction of site is 270° (western direction) as observed in
Fig. 2. The meteorological mast was positioned in predominant wind direction of site
at 270° (western direction) at a distance of 200 m with respect to test turbine during
testing. The test site is completely free from obstacles in the predominant direction.
However, the test turbine acts as an obstacle to meteorological mast; therefore, data
sets from 51° to 129° direction have been excluded from the analysis. Figure 3
illustrates elevation at site around 2 km radius of test turbine. The site elevation
varies from 65 to 105 m above the sea level. The terrain slopes are within the limits
as specified in IEC 61400-12-1 [1] standard Annexure ‘B’. Therefore, the terrain
effects on wind speed measurement in each measurement sector are negligible [4,
6].

2.1 Hub Height Wind Speed and Reference Height Wind

Speed Correlation

The hub height anemometer and reference height (ref-height) anemometer have been
instrumented on meteorological mast as specified in IEC 61400-12-1 [1] standard

Fig. 2 Wind rose at site

A New Technique for Wind Turbine Power Curve Incomplete Bin … 89

Fig. 3 Terrain elevation view

Annexure ‘G’ G.2 to eliminate wake effects due to anemometer booms. Figure 4 illus-
trates anemometer position on meteorological mast. However, anemometer at hub
height and the reference anemometer (below 1.5 m the hub height) have been corre-
lated to check the quality of measurement (10 min). Figure 5 shows the regression
curve between two anemometers on metrological mast.
The correlation coefficient is calculated as 0.9998. Therefore, the quality of wind
speed measurement has been good. The hub height anemometer data sets have been
used for the analysis.

1500 mm HUB HEIGHT

ANEMOMETER

HUB HEIGHT FROM

GROUND LEVEL

RHT PRESSURE
REF-
ANEMOMETER

SAFETY ROPE

Fig. 4 Sensor position on meteorological mast

90 B. Ramdas et al.

Fig. 5 Wind speed correlation

3 Sensitivity Analysis and Data Correlation

Before preceding to compilation of 30 s data sets in the incomplete bins, a sensitivity

analysis on 30 s data sets has been performed. It is necessary to first examine which
time averaging window data sets are appropriate to fill the incomplete bins. This has
been substantiated by comparing the IEC power curve with 5 min, 2 min, 1 min and
30 s average power curve (valid data points) and the results are as shown in Fig. 6.
The criteria for power curve comparison have been kept as uncertainty band of IEC
power curve. The uncertainty of electrical power has been evaluated by considering
the uncertainty components as specified by IEC 61400-12-1 [1] standard. From the
analysis found that all the power curves are reasonable and within the uncertainty
band of IEC power curve. The 30 s and 1 min power curves at lower wind speed little
outside the uncertainty band of IEC power due to time lagging phenomenon. This
has been addressed in Fig. 6. Meteorological mast and turbine configuration plays
vital role in selection of valid 30 s average data sets; from the analysis, it is evident
that the inclusion of time lagging data leading to deviation of 30 s power curve at low
wind speed when compare to the IEC power curve. Extensive care must be taken in
selection of 30 s data sets; a sensitivity analysis required to be carried is addressed.

3.1 Meteorological Mast and Wind Turbine Configuration

Effects on 30 s Data Sets

The meteorological mast was positioned in predominant wind direction to record

wake free wind speed data. Figure 7 illustrates the meteorological mast position, test
turbine position at the site. From Fig. 7, the wind parcel form western direction would
hit meteorological mast first which is shown in configuration-1. In configuration-2,
wind parcel from southern direction and northern direction would hit the meteo-
rological mast and wind turbine simultaneously. When the data averaging period
A New Technique for Wind Turbine Power Curve Incomplete Bin … 91

Fig. 6 Power curve comparison

CONFIGURATION-1 CONFIGURATION-2
NORTH(0Deg)
NORTH(0Deg)

51Deg
51Deg
TEST TURBINE
TEST TURBINE
METEOROLOGICAL MAST
WIND METEOROLOGICAL MAST EAST(90Deg)
EAST(90Deg)
WEST(270Deg)
WEST(270Deg)

ELIMINATED SECTOR
ELIMINATED SECTOR 129Deg
129Deg

SOUTH(180Deg)
SOUTH(180Deg)

WIND

Fig. 7 Meteorological mast position in site

92 B. Ramdas et al.

is shorter that is thirty seconds, the time lag phenomenon could be present in data
gathered from the western direction (predominant wind direction). The time lag
between wind speed recorded at meteorological mast and wind speed available on
wind turbine rotor is purely dependent on velocity of wind speed and the distance
between meteorological mast to wind turbine. For simplicity purposes, the effects of
wind turbine’s inertia have not been taken into consideration for calculation of wind
speed at the turbine. The correlation coefficients of wind speed and rotor speed in
western direction and southern direction have been compared initially to study the
time lag phenomenon in 30 s data sets.

3.2 Wind Speed and Rotor Speed Correlation in Western

and Southern Directions

The wind speed at meteorological mast (met-mast) and rotor speed correlations have
been compared accurately in western and southern direction by filtering out the pitch
effects on 30 s data sets. Figure 8 illustrates the graph of pitch angle with respect to
wind speed. The pitch angle of the blade is kept constant between 0.2PU to 0.6PU
wind speed. Therefore, data sets at constant pitch angle alone have been considered
to correlate wind speed at met-mast and rotor speed. Figure 9 illustrates the wind
speed at met-mast and rotor speed correlation in western direction. The data sets
from 260° to 280° that is 20° (narrow band) band only have been considered. The
correlation coefficient is calculated as 0.7158.
Figure 10 illustrates the wind speed at met-mast and rotor speed correlation in
southern direction. The data from 170° to 190° that is 20° (narrow band) band only
have been considered. The correlation coefficient is calculated as 0.907. The correla-
tion coefficient holds better in southern direction because the wind speed sensed by
anemometer and rotor speed are recorded simultaneously without time lag. The 30 s
data sets from southern direction are used for analysis because of the existence of a
better correlation between wind speed at met-mast and rotor speed. Further, to make

Fig. 8 Pitch angle as a

function of wind speed

Constant
Pitch
Angle Zone
A New Technique for Wind Turbine Power Curve Incomplete Bin … 93

Fig. 9 Rotor speed function

of wind speed in western
direction

Fig. 10 Rotor speed function of wind speed in southern direction

sure same wind parcel passes through both meteorological mast and wind turbine in
southern direction, the wind speed measured on meteorological mast and the wind
speed available across rotor have been correlated.

3.3 Wind Speed at Met-Mast and Wind Speed Across Rotor

Plane Correlation in Southern Direction

The meteorological mast is 200 m away from the test turbine. The wind parcel
from southern and northern directions passes through the meteorological mast and
wind turbine at same time. In order to make sure that the same wind parcel has
passes through both the meteorological mast and wind turbine positions, the wind
speed recorded at meteorological mast and wind speed available across the wind
turbine rotor have been correlated. The wind speed across the turbine rotor cannot
be measured directly in this case, and hence, it has been derived from the electrical
94 B. Ramdas et al.

Fig. 11 Meteorological
mast wind speed and wind
speed across turbine rotor
correlation

power produced by wind turbine using Eqs. (1) and (2) [5]. The data sets have been
filtered out for 0.2PU to 0.6PU wind speed to eliminate pitch effects during the wind
turbine cut-in to grid and regulation periods.

P(v)
Cp = (1)
0.5 ∗ ρ ∗ A ∗ V 3
 ( 13 )
P(v)
V = (2)
0.5 ∗ ρ ∗ A ∗ Cp

C p —Power co-efficient of wind turbine

A—Swept area of wind turbine (m2 )
P(v)—Power produced by wind turbine (W)
V —Wind speed across rotor (m/s)
ρ—Air density (kg/m3 ).
Figure 11 shows the correlation between wind speed recorded on meteorological
mast and wind speed available across wind turbine rotor. The correlation coefficient
is evaluated as 0.9791 and found good; it is assumed that the same wind parcel
has been passing though meteorological mast and turbine from southern direction.
Therefore, to eliminate time lag phenomena in thirty seconds data sets, the data sets
from southern (most of the data is from southern direction) direction (configuration-
2) alone have been considered for the analysis. Further, in order to demonstrate the
elimination time lag phenomenon from 30 s data base, the IEC power curve (10 min
data sets) and 30 s power (southern data sets) have been compared.

3.4 IEC and 30 s Power Curve’s Correlation in Southern

Direction

The IEC power curve (evaluated from data sets of southern direction) and the 30 s
power curve (evaluated from data sets of southern direction) have been compared.
A New Technique for Wind Turbine Power Curve Incomplete Bin … 95

1.2

1.1

0.9

0.8

0.7
Electrical Power[P.U]

0.6

0.5

0.4

0.3

0.2

0.1

0 Electrical Power P+-s(P)

30sec Power Curve
IEC Power Curve
-0.1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2
Wind Speed[P.U]

Fig. 12 IEC power curve and 30 s power curve comparison in southern direction

Figure 12 illustrates IEC power curve and 30 s power curve comparison made in
southern direction. From Fig. 12, the 30 s power curve is within the uncertainty
band of IEC power curve and no significant differences have been observed between
them. Therefore, it has been concluded that the 30 s data sets from southern direction
eliminate the time lag issues and are appropriate to fill the incomplete bins of IEC
power curve [7].

4 Incomplete Bin Evaluation

4.1 Incomplete Bin Evaluation and Power Curve Comparison

In order to demonstrate the application of valid 30 s data sets throughout the power
curve, an exercise has been carried out using the IEC data base (pitch regulated wind
turbine measurement). A few bins have been manually removed from the IEC data
base (data base 1). The selection of removing bins has been made in such a way that
to see the effectiveness of filling valid thirty second data sets throughout the range
of power curve. Thus, the data sets of 5 m/s bin (near to cut-in wind speed), 11 m/s
bin (at regulation wind speed) and 8 m/s bin (one intermediate wind speed) have
been removed from IEC data base (data base 1). Figure 13 illustrates the IEC power
96 B. Ramdas et al.

Fig. 13 IEC power curve

with missing bin
Incomplete
bins

curve with bins that have been removed. The eliminated data sets have been filled
with valid 30 s data sets (southern direction) and data base 3 has been created (417
data points in 5 m/s bin, 1280 data points in 8 m/s bin, 708 data points in 11 m/s bin),
using data base 3 power curve 3 has been evaluated.
The IEC power curve (power curve 1) and the power curve (power curve 3)
which missing bins are filled with valid thirty second data sets have been compared.
Figure 14 illustrates the power curve comparison. The power curve (power curve 3)

1.2

1.1

0.9

0.8

0.7
Electrical Power[P.U]

0.6

0.5

0.4

0.3

0.2

0.1

-0.1 Electrical Power P+-s(P)

Power Curve which bins are filled with 30sec data
IEC Power Curve
-0.2
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2
Wind Speed[P.U]

Fig. 14 IEC power curve and the power curve which bins are filled with thirty seconds data sets
A New Technique for Wind Turbine Power Curve Incomplete Bin … 97

Table 2 IEC power curve bin and 30 s results comparison in stall regulated wind turbine power
curve
Actual power 30 s results Difference (%)
Wind speed (P.U) Power (P.U) Wind speed (P.U) Power (P.U)
0.96 1.08 0.97 1.07 1.6%

bins filled with thirty second data sets are within the uncertainty band of IEC power
curve (power curve 1), and no significant differences have been noticed between
them [7].
The same technique has been implanted in stall regulated wind turbine by filling
the one bin with 30 s data sets (315 data points filled) near rated wind speed. The
results are tabulated in Table 2. The difference between actual power and the power
evaluated using 30 s data found only 1.6% and results have been improved and
realistic when compare to the interpolation method.
The use of 30 s data sets for incomplete bin evaluation their impact on AEP is
also essential to be studied. The AEP comparison is given.

4.2 Annul Energy Production Comparison

Annual energy production (AEP) has been estimated by applying the power curve
to wind speed frequency distribution. A Rayleigh distribution with shape factor 2
has been used as the reference frequency distribution [1]. AEP estimation has been
calculated for annual average wind speed 4, 5, 6, 7, 8, 9, 10 and 11 m/s. The annual
energy estimation (AEP) of power curve 1 and power curve 3 has been compared
(pitch regulated wind turbine measurement data base). The greatest difference in
AEP of power curve 1 and power curve 3 has been calculated as −0.403% at 4 m/s
annual average wind speed. Therefore, there is no significant difference observed in
annual energy estimation while using 30 s data for incomplete bin evaluation. The
detailed energy estimation comparison has been tabulated in Table 3.
The variation in AEP with use of 30 s data sets in case of stall regulated wind
turbine also studied. The AEP results are given in Table 4. The AEP maximum
difference between actual power curve (IEC Power curve) and the power curve which
bins are filled with 30 s data sets is observed as 0.11% at 11 m/s annual average wind
speed. The AEP difference is reduced by using 30 s data sets when comparing to
linear interpolation technique in stall regulated wind turbine.
98 B. Ramdas et al.

Table 3 Annual energy estimation comparison of pitch regulated wind turbine

Annual average wind Power curve 1 AEP (P.U) Power curve 3 AEP (P.U) Difference (%)
speed (m/s)
4 0.19 0.19 −0.40
5 0.37 0.37 −0.27
6 0.56 0.56 −0.18
7 0.74 0.74 −0.14
8 0.87 0.87 −0.11
9 0.95 0.96 −0.09
10 0.99 1.00 −0.08
11 1.00 1.00 −0.07

Table 4 Annual energy estimation comparison of stall regulated wind turbine

Annual average wind IEC power curve AEP Power curve filled with Difference (%)
speed (m/s) (P.U) 30 s data sets AEP (P.U)
4 0.10 0.10 0.00
5 0.25 0.25 0.00
6 0.43 0.43 0.00
7 0.61 0.61 0.00
8 0.77 0.77 0.00
9 0.89 0.89 0.00
10 0.96 0.96 0.00
11 1.00 0.999 0.11

5 Conclusion

The wind turbine power curve measurement imposes greater challenges to get
adequate data base as specified in IEC 61400-12-1 [1] standard due to the voluminous
installation of wind turbines, grid availability and wind conditions at site. The IEC
warrants estimation of the power curve bins using linear interpolation method for
those bins where data is inadequate. The power curve bins linearly interpolated could
lead to wrong performance assessment due to the nonlinear characteristics of the wind
turbine. In stall regulated wind turbine, the approach over estimates the power about
32%, thereby over estimates the AEP of 1.2%. The paper has demonstrated that the
data averaged over thirty seconds can be used to fill up the incomplete bins of IEC
data base rather than traditional linear interpolation method. The wind speed data
averaged over thirty second could have time lag between wind speed measured at
meteorological mast and wind speed available across turbine rotor, to eliminate time
lag phenomena as discussed the data sets from southern direction alone have been
considered for the indented analysis. The proposed technique has been tested with
A New Technique for Wind Turbine Power Curve Incomplete Bin … 99

pitch and stall regulated wind turbines power curve data base’s and found that the
results are reasonable and realistic. The power curve with missing bins is filled with
thirty second data sets and IEC power curve has been compared and no significant
differences have been noticed. Furthermore, the AEP using power curve 1 and power
curve 3 has been made (pitch regulated wind turbine power curve data base) and the
percentage difference noted was not greater than −0.403%. In stall regulated wind
turbine through proposed technique, the power estimation is improved by 30.4% and
the AEP by 1.1%. Therefore, in conclusion, the data sets averaged over thirty second
can be used to fill up any missing bins of IEC power curve, within the allowable limits
of uncertainty more scientifically based on transient performance of wind turbine.

Acknowledgements The author(s) wishes to acknowledge the efforts of the project team namely
Mr. A. R. Hasan Ali, Mr. Y. Packiyaraj, Mr. M. Karruppuchamy for the Power Curve Measurement
Campaigns of the test turbines.

References

1. IEC 61400-12-1: first edition 2005-12, wind turbines—part 12-1: power performance measure-
ments of electricity producing wind turbines
2. Clifton A, Wagner R (2014) Accounting for the effect of turbulence on wind turbine power
curves. J Phys. https://doi.org/10.1088/1742-6596/524/1/012109
3. Mathew SA, Ramdas B (2014) Wake effects of wind turbines and its impact on power curve
measurements. Int Sci Index 8(10), Part VI:730–735
4. Lakshmanan N, Gomathinayagam S, Harikrishna P, Abraham A, Ganapathi C (2009) Basic wind
speed map of India with long-term hourly wind data. Curr Sci 96(07)
5. IEC 61400-12-2: first edition 2013-03, wind turbines—part 12-2: power performance measure-
ments of electricity producing wind turbines on nacelle anemometry
6. Gottschall J, Peinke J (2008) How to improve the estimation of power curves for wind turbines.
Environ Res Lett 3:015005 (7pp). https://doi.org/10.1088/1748-9326/3/1/015005
7. Elliot D, Infield D (2014) An assessment of the impact of reduced averaging time on small wind
turbine power curves, energy capture prediction and turbulence intensity measurements. Wind
Energy 17:337–342
Grid Integration of Wind Energy
Conversion System with Variable Speed
Analysis Using Simulink Model

Ravneet Kaur, Deepika Bhalla, and Naveen Kumar Sharma

Abstract The conventional energy resources such as wind have a major role in the
production of power since the conventional fossil fuels are getting extinct, while the
requirement for electrical energy is increases. Wind power, as a source of renewable
energy, has proved to be amongst the most reliable and developing source for meeting
the demand of electrical energy. The necessity of grid integrated wind energy conver-
sion system (WECS) is evident. Doubly fed induction generator (DFIG) is universally
accepted for WECS. In this work, the analysis of active and reactive power of 220 kV
grid and the voltage generated at different speeds has been undertaken. Two WECS
operating at different wind speeds have been integrated with grid. The power and
voltage generated has been analyzed when both the WECS operate at same and at
different speed of wind. Wind speed has been analyzed at which the DFIG feeds
power to the grid.

Keywords Wind energy conversion system WECS · Doubly fed induction

generator DFIG · Variable speed

1 Introduction

The industrialization and urbanization have led to a boost in global power demand.
The rise in power demand and depletion of natural energy resource has caused a huge
energy deficiency. It is now broadly received that the conservative source of energy
may not be enough for gathering the power demand. The renewable energy zone in

R. Kaur (B) · D. Bhalla

IKG Punjab Technical University, Main Campus, Kapurthala, Punjab 144603, India
e-mail: [email protected]
D. Bhalla
e-mail: [email protected]
N. K. Sharma
IKG Punjab Technical University, Batala Campus, Batala, Punjab, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 101

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_11
102 R. Kaur et al.

India for wind as a source is among the top five countries in the world. Quite a few
operators are working on renewable power mainly solar and wind for electric power
generation [1]. The growth of wind power in India began in the 1990s in Tamil Nadu
and has extensively risen in the last decade [2]. The entirety amount of reasonably
extractable power available from the wind is considerably more than human power
used from all sources [3]. The electricity generating via wind turbines is a verified
and tested technology now. It gives a safe and sustainable energy supply and now it
can successfully participate with conventional energy production.
Advancements in technology of wind turbine required complicated control
systems to well optimize energy conversion and get better grid integration the power
flow study of grid coupled wind energy conversion system (WECS) in extremely
changeable wind atmosphere has been discussed [4]. As a generally perspective
of energy plants of wind turbine and grid integration, a variety of wind turbine
systems with different generators are being utilized, and different technical features
are compared [5]. Mishra has discussed different characteristic related to turbine and
finish off wind turbines with induction generators (IG) are only used by small wind
farms because they lack reactive power controlling system [6]. Hu observed that
wind turbines generation system (WTGS) based on doubly fed induction generator
(DFIG), with back-to-back voltage source converters rated at about 25–30% of the
generator rating, can be used for high power function [7]. Ritter proposed the suit-
able wind turbine that achieves the highest efficiency and least amount of cost per
MWh [8]. Pena [9] used model reference adaptive system (MRAS) to estimate the
same. Some techniques are proposed by different researchers [10, 11] to estimate
rotor position, rotor currents and torque angle.
The different type of impact of wind turbines on voltage sag and the clarification
and for each kind of impact, a evaluation between wind turbines have been discussed,
along with an measurement for the doubly fed induction generator (DFIG) efficiency
has been obtainable, discussed [12]. In this paper, two cases are discussed. The first
being the analysis of the active and reactive power in grid and the voltage generated by
the WECS for different wind speeds and second is the integration of two WECS with
the grid when observing same wind speed and different wind speeds. In the paper
[13], the electrical a grid component connected changeable speed wind turbine is
considered, which is operational with a permanent magnet synchronous generator
(PMSG).

2 Wind Energy Conversion System

The wind energy conversion system (WECS) changes the wind energy movement into
mechanical energy; the wind turbine generator converts this mechanical energy into
electrical energy. Also, wind mills are used for energy is to do work such as pumping
water, drive machinery, mill grains or generating electricity. A basic WECS is shown
in Fig. 1. The wind turbine is connected with the gearbox which is linked with the
synchronous generator. DFIG is used for generating power. DFIG generates power
Grid Integration of Wind Energy Conversion System with Variable … 103

Fig. 1 Wind energy conversion system (WECS)

Fig. 2 Characteristics of wind power generation

in sub-synchronous speed. When the rotor reaches sub-synchronous speed, the rotor
generates power. Pulse width modulation (PWM) converters are used which converts
AC/DC and DC/AC. It is also back-to-back converter. The characteristics of wind
power generation are shown in Fig. 2. The least speed of wind at which a wind turbine
begins to generate utilizable power is called cut-in speed. The uppermost speed of
wind at which a wind turbine stops generating power is called cut-out speed. The
speed of wind increases, higher than the cut-in speed; the electrical production energy
rank increases speedily.

3 Wind Turbine Power Extraction

Wind is the kinetic energy of airs action. Wind is the association of air diagonally
the surface of the ground, artificial by the low and high pressure [14]. Total energy
of wind flowing from beginning to end an imaginary surface with area A during the
time t is given by

E = 1/2mv 2 (1)
104 R. Kaur et al.

= 1/2(Avtρ)v 2 (2)

= 1/2 Atρv 3 (3)

where ρ is the air density; v is the speed of wind; Avt is the air volume flowing
through A (is measured vertical to the path of the wind); ρ is as a result the mass m
transient through A. The ½ρv2 is the moving air kinetic energy per unit volume [3].
Wind turbines for grid electric power thus require being mainly proficient at greater
wind speeds [15]. Total wind power is given by:

P = E/t = 1/2 Aρv 3 (4)

Equation (4) gives the power P presented in the wind power transferred to the
rotor of wind turbine is condensed by the power coefficient C p :

Pwind turbine
Cp = 1/2 p Av 3
(5)

Pwind turbine = 1/2Cp AV 3 (6)

3.1 Generation of Electricity from Wind Energy by Using

DFIG

The first one is a squirrel cage induction generator, coupled directly to the power grid.
It is used for less range of wind speed turbine. Another one is or secondly, we used an
AC–DC–AC converter, in addition to the induction generator before being connected
to the power grid. The third type is used a wound rotor induction generator, which is
directly connected to the grid where the rotor speed is adjusted using a rheostat [16].
The fourth type is a doubly fed induction generator, which is directly connected to
the grid where speed is adjusted using back-to-back converters [17].

3.2 Active, Reactive Power of the WECS for Different Wind

Speeds

The power is truly consumed by an AC system in power system is known as active

power or real power or true power. The unit of active power is or can be measured
in kilowatt (kW) or MW. The power which flows back and front so as to represent it
moves in bidirectional though the connected circuit and act in response upon itself is
Grid Integration of Wind Energy Conversion System with Variable … 105

called reactive power. The unit of reactive power is kilovolt ampere reactive (kVAR)
or MVAR [18]. The major difficulty of power generated by wind turbine is to manage
the reactive power of the power system. The voltage of a transmission network is
found by the reactive power flow. DFIG has the potential to control the reactive power
flow through the connection network and the network to which they are linked. In
a large wind farm, controlling individual DFIG wind turbine to regulate reactive
power flow is not feasible. It is unable to control the voltage flow in the grid [18].
Generally, the reactive power and voltage control at the grid are achieved by using
compensation of the reactive power. If rapid loss of load, occurrence, the causes of
voltage to rise above 10% of nominal voltage, this could over-excite transmission
transformers and simultaneously cause damage. Extensive work for optimization is
also being carried out [19, 20].

4 Modelling and Simulations

A 9 MW wind farm consisting of six 1.5 MW wind turbines generate 25 kV and supply
power to 120 kV grid. The 25 kV feeder is 30 km, 25 kV feeder, and connected to
resistive load by bus B25. Both the wind turbine and the motor load have a protection
scheme monitoring voltage, current and machine speed. The DC link voltage of DFIG
is monitored. The wind turbines are coupled to DFIG is a wound rotor induction
generator and an AC–DC–AC converter IGBT-based PWM converter. The winding
of stator is linked with 50 Hz grid while the rotor is fed at changeable frequency
during the AC–DC–AC converter.

4.1 Turbine Power Characteristics

The characteristic of wind turbine power is shown in Fig. 3. The mechanical power
of turbine is displayed for wind speeds ranging from 5 to 16.2 m/s. The DFIG is
guarded in instruct to chase the red curve. Turbine speed is optimum between point
B and C on this curve.

4.2 Wind Generator Connected to Line

As seen in Fig. 4, the DFIG is connected with the three-phase line. B575 is the
three-phase measurement component that shows the output at the scope and on
display screen. A, B, C indicate the generated three-phase power and are directly
connected with line. Points ‘m’ is the measurement point of the components that
are connected with demultiplexers. V dc shows the output power (DC) of the turbine
which is transformed to AC with the help of inverter.
106 R. Kaur et al.

Fig. 3 Power characteristics of wind turbine

Fig. 4 Wind turbine model

4.3 120 kV Line Modelling

Figure 5 shows the model of grid side in Simulink tool (MATLAB). The voltage
programmable source is generated at 120 kV which is linked with three-phase mutual
inductance. The three-phase mutual inductance apparatus has a three-phase balanced
inductive and resistive impedance with mutual coupling between phases. B120, three-
phase measurement block is connected between three-phase mutual inductance and
120 kV/25 kV transformers. Scope is connected with the three-phase VI measurement
component. This scope shows the measurement of voltage and current in the wave
Grid Integration of Wind Energy Conversion System with Variable … 107

Fig. 5 Modelling of grid

form in per unit (pu) value. 25 kV line is connected with the grounding transformer
which is further connected with 3.3  grounded resistance.

4.4 Integrated Model of Wind Generator with Grid

In the model, wind turbine and the grid line deliver the power to load. There is
500 MW load connected with the 575 V line which is near the distributed generator
of wind turbine. The feeder voltage of 30 km length is 25 kV. Wind turbines produce
power by a DFIG. It has induction generator (wound rotor induction generator) and
PWM converter (AC–DC–AC IGBT based). The DFIG technology allows extract
the maximum power from the wind for low wind speeds by optimizing the turbine
speed, while minimize the mechanical stresses on the turbine during gusts of wind.
The control system uses a torque manages in organize to continue the speed at 1.2 pu.
The reactive power produced by the wind turbine is synchronized at 0 MVAR. The tip
speed ratio λ (lambda) and the C p (power coefficient) value as function of wind speed.
For a wind speed of 15 m/s, the turbine output power is 1 pu of its rated power, the
pitch angle is 8.70 and the generator speed is 1.2 pu. The system has a back-to-back
converter with compact power rating. The bidirectional power converter consists of
two usual PWM inverters. It is at the present time one of the broadly used converter
topology in wind energy conversion system. The rotor side inverter is controlled so
as to extract the maximum power from the wind turbine and two regulate the reactive
power transferred to the utility grid. The main purpose of the grid side inverter in
the control of DC link voltage regardless of the wind speed. The DFIG is based on
wound rotor type induction machine [17].
108 R. Kaur et al.

4.5 Integration of Two-Wind Power Plants

There are two-wind mill connected with single grid line. The two DFIG generators are
connected with the grid; these two turbines increase the total power and are connected
in parallel. Figure 6 shows the arrangement of two-wind turbine connected with grid.
These Wind Turbine-1 and Turbine-2 are connected in parallel to each other. They
increase the power or matching the load demand which increases day by day. The
two-wind turbines are of same ratio and are connected with line, i.e. 9 MW. There
are two cases consider:
Case-1: The wind speeds of Turbine-1 and Turbine-2 are same at a time.
Case-2: The wind speeds of Turbine-1 and Turbine-2 are different at a time.
To consider these cases, we have check real power at different operating conditions
of both wind turbines. These above cases show the working of wind turbine of the
model. In case-1, the speed of wind is same at both turbine locations and is taken to
be as 8 m/s. In case-2, the wind speeds of these turbines are different. The speed of
wind at Turbine-1 is 8 m/s and the wind speed of Turbine-2 is 12 m/s.

Fig. 6 Two-wind turbine integration

Grid Integration of Wind Energy Conversion System with Variable … 109

5 Results and Discussion

5.1 Case-1: The Wind Speed of Turbine-1 and Turbine-2 Are

Same at a Time

The purpose of this work of simulation and analysis is to observe the relationship
between various wind velocity with that of power generated by the WECS. Consider
Turbine-1 to be operational and five samples are taken and the outputs from simula-
tion are shown of the wind turbine side in Table 1 and grid side parameters in Table
2. Table 1 gives the effects of different wind speeds on different parameters of wind
turbine. The speed has been randomly varied from 2 to 16 m/s. since major issue of
a wind farm is variation of wind speed. It is observed that all the parameters of wind
turbine are affected due to variation of speed. These tables show the measurement
of voltage (V1B525) and current (I575) at wind turbine side, active power (P) and
reactive power (Q), speed of generator and pitch angle at different wind speeds. It is

Table 1 Wind turbine data when Turbine-1 operates

Parameters Different wind speeds
2–4 m/s 4–8 m/s 8–10 m/s 10–12 m/s 12–16 m/s
V 1B575 (pu) 0.9993 1 1 1.001 1.001
I 575 (pu) 0.03514 0.187 0.358 0.6393 0.8796
P (MW) 2125e−6 1.87 3.567 6.373 8.781
Q (MVAR) 0.3511 −0.04157 −0.3224 0.5994 −0.6794
V DC (V) 1200 1200 1200 1200 1200
Speed of generator (pu) 0.654 0.8 0.994 1.199 1.216
Wind speed (m/s) 8 10 12 16
Pitch angle (°) 0 0 0 0 2.896

Table 2 Grid data when Turbine-1 operates

Parameters Different wind speed
2–4 m/s 4–8 m/s 8–10 m/s 10–12 m/s 12–16 m/s
VB120 (pu) 0.9999 0.9999 0.9998 0.9996 0.9993
VB25 (pu) 0.9966 0.9979 0.9984 0.9978 0.995
VB575 (pu) 0.9993 1 1.001 1.001 1.001
PB25 (MW) 2.352 0.4863 −1.297 −3.951 −6.293
QB25 (MVAR) −0.4198 −0.01405 0.3588 0.8S871 1.327
V1 (plant) (pu) 0.9979 0.9993 0.9998 0.9992 0.9973
I1 (plant) (pu) 0.966 0.965 0.9647 0.9651 0.9665
Motor speed (pu) 0.9933 0.9933 0.9933 0.9933 0.9933
110 R. Kaur et al.

observed that V dc is constant and pitch angle is zero at all wind speeds (from 2 to 12
m/s). When the wind speed is 12 m/s, the pitch control is changed from 0° to 2.896°.
The active power (P) is very less at the low wind speed, and it increases as the wind
speed increases.
Table 2 gives the grid side parameters; the e voltage at B120 bus bar, (VB120)
voltage of B25 bus bar (VB25) and voltage of B575 bus bar per unit, active and
reactive power and voltage and current of protection device bus bar, i.e. V1 and I1.
Due to changing the wind speed, the wind turbine parameters are changed and the
effect of variation of wind speed also impacts on the grid side. Table 2 shows that
the voltage of grid side is approximately constant, i.e. 0.99 and near nearly unity.
The bus bar voltage of VB25 is also near approx. unity, i.e. 0.99 per unit. The bus
bar voltage of VB575 is also approximately unity at different wind speeds. But at the
speed of 8 m/s is 1 per unit (pu). The active power at the speed of 4 m/s is 2.35 pu;
when the wind speed is 4–8 m/s, the active power is decreased to 0.4863, hence that
wind plant can feed more load. And when the speed of wind turbine is 10 m/s, the
value of active is negative (−1.297). The negative value shows that the load is totally
feed by the wind turbine.
The optimum value of active power at the speed of 8 m/s is 1.87 pu when both
grid and WECS feed the load, Beyond 8 m/s, the WECS feeds the load as well as
supplies active power to the grid.
Figure 7 shows the scope for wind speed is from 4 to 8 m/s. The IB575 bus

Fig. 7 Parameters of WECS for 4–8 m/s wind speed

Grid Integration of Wind Energy Conversion System with Variable … 111

current is 0.187 per unit. After reaching 5 s, the wind speed increases and the current
decreases. Similarly, when the wind speed is 8 m/s, the generated power has optimum
value while the wind speed is increased at 4–8 m/s that time the power is maximum.
Wind turbine at the speed of 8 m/s final and initial is 4 m/s; this constant value
also shows in Table 1. In Fig. 7, the positive sequence of voltage is approximately
1 per unit and is constant.
B120 shows the bus voltage of grid side, B25 shows the bus voltage at line, and
B575 shows the bus voltage at wind turbine side. PB25 shows the active power of
the line and QB25 shows the reactive power of the line. Vplant shows the voltage of
protection plant and Iplant shows the current of protection system.
These following scopes show the output of wind turbine data which shows in
Tables 1 and 2. These scope analyze the wind speed effects on the different param-
eters, namely voltage, current active reactive power, speed of motor, speed of wind,
pitch angle at wind turbine side and voltages of three different buses. Figure 7 shows
the simulated result of wind turbine measuring parameters. Since DFIG is used;
therefore, it generates the constant voltage as seen the scope.
Figure 8 shows the simulation result of grid side parameters at the wind speed
of 4–8 m/s which is considered as the optimum value. If the velocity of wind is
increased than 8 m/s, the value of active power measured in negative.

Fig. 8 Parameters of grid for 4–8 m/s wind speed grid data
112 R. Kaur et al.

5.2 Case-2: The Wind Speeds of Turbine-1 and Turbine-2

Are Different at a Time

Two-wind mills are connected with a grid line. The use of renewable source of energy
is maximized day by day matching the load demand. The two DFIG generators are
directly connected with the grid. By the connected of two turbine, the real power of
wind turbine is increased in the grid line. Both are connected with parallel to each
other. There are two cases consider:
Case 2-A: Wind speed of Turbine-1 and Turbine-2 are same at that instant.
Case 2-B: Wind speeds of Turbine-1 and Turbine-2 are different at that instant.
Case 2-A: Wind speed of Turbine-1 and Turbine-2 is same at that instant: Consider
the same wind speed at both of turbine data. From the simulation, it is observed that
when the wind speed for both turbines connected with grid is same, then the total
active power of grid side is 3.084 pu and the reactive power is 0.069 as shown in
Fig. 9.
When only one wind turbine was connected with grid and speed of wind was
4–8 m/s, the grid power was however when both turbines operate at the same speed
the grid active power is −3.084.

Fig. 9 Grid side data measuring scope for case 1: wind speed same for both turbines
Grid Integration of Wind Energy Conversion System with Variable … 113

Table 3 Two turbines run at

Turbines parameters Turbine 1 Turbine 2
similar wind speed
Wind speed (m/s) 4–8 4–8
Active power (MW) 1.87 1.83
Speed of generator (pu) 0.8 0.8
Voltage (pu) 1.015 1.015

Table 4 Two turbines run at

Turbines parameters Turbine 1 Turbine 2
two different speeds
Wind speed (m/s) 4–8 8–12
Power (MW) 1.87 6.203
Speed of generator (pu) 0.8 1.193
Voltage (pu) 1.027 1.027

The negative sign of the power is shows that the wind energy conversion system
is feeding power to grid as more. It means that by increasing the number of turbine,
the power of wind energy will also increase. Table 3 shows the data of TURBINE-1
and TURBINE-2, i.e. wind speed in m/s, active power in pu, reactive power in pu,
speed of generator in pu.
Case 2-B: Wind speeds of Turbine-1 and Turbine-2 are different at that instant: On the
simulation of two-wind energy conversion systems connected with grid at different
wind speeds in data shown in Table 4 and also show the digital scope in Fig. 10.
Turbine-1 operates at 4–8 m/s and Turbine-2 operates at 8–12 m/s.
When the single turbine is considered than the total active power of grid on 8 m/s,
wind speed is 0.4863. When two turbines are connected with grid and are operating
at different wind speeds, the total power of grid side assumed as −7.35. The negative
sing shows the direction of active power. The WECS generate more power when
its number of turbines will increase and WECS feedback to grid which is shown in
Fig. 9 and its also shows that due to increasing the wind, the power of active is also
increasing at a particular value.
This shows the total power of wind energy conversion system raise with rising the
number of turbine as well as increasing the effect of wind on wind energy conversion
system. For the turbine operating at lesser speed, the active power of wind energy
conversion system generated was 1.87 and then on higher speed, the generated power
was 6.203. This data shown in Table 4 the higher speed turbine took 76.83% of the
load.
114 R. Kaur et al.

Fig. 10 Grid side data measuring scope for case 2: wind speed different for both turbines

6 Conclusion

On account of increase in power demand, additional sources of power are essentially

needed to meet the need. The production should be such that decrease environmental
pollution; this can be partially met from wind energy plants rather than conventional
fossil plants. In this paper, wind energy system is considered for meeting the load
demand. The configurations for the WECS for achievable results are discussed while
there is integration of wind turbine with grid. The optimum value of wind is discussed
with two-wind turbine integration that operate at same that time in parallel. Typically
generated the electric energy from wind turbine is sent to grid. But due to changeable
nature of wind flow, it is huge challenge to meet the terms with grid code. Work has
been done successfully for checking the operation of two WECS operating at different
wind speeds. This research work is proposed to the current technology linked to wind
electrical system and prospect of future research direction.
Grid Integration of Wind Energy Conversion System with Variable … 115

References

1. Winds of change come to country plagued by power blackouts (2012) Guardian, 31 Dec 2008.
Retrieved 13 Jan 2012
2. Government working to double generation target from solar parks (2016) Economic Times, 18
July 2016. Retrieved 18 July
3. Hurley B (2012) How much wind energy is there?—Brian Hurley—Wind Site Evaluation Ltd.
Claverton Group. Retrieved 8 Apr
4. Subramanian C, Casadei D, Tani A, Rossi C (2014) Modelling and simulation of grid connected
wind energy conversion system based on a doubly fed induction generator (DFIG). IEE Proc
J Electr Energy 2(2)
5. Mishra AK, Ramesh L, Chowdhury SP, Chowdhury S (2011) Review of wind turbine system
and its impact for grid stability. J Electr Eng
6. Zhang Y, Ula S (2008) Comparison and evaluation of three main types of wind turbines. In:
IEEE/PES transmission and distribution conference and exposition, May 2008, pp 1–6
7. Hu J, He Y, Xu L, Williams BW (2009) Improved control of DFIG systems during network
unbalance using PI–R current regulators. IEEE Trans Ind Electron 56(2):439–451
8. Ritter M, Deckert L (2016) Site assessment, turbine selection, and local feed-in tariffs through
the wind energy index. Appl Energy
9. Cardenas R, Pena R, Proboste J, Asher G, Clare J (2005) MRAS observer for sensorless control
of standalone doubly fed induction generators. IEEE Trans Energy Convers 20(4)
10. Xu L, Cheng W (1995) Torque and reactive power control of a doubly fed induction machine
by position sensor-less scheme. IEEE Trans Ind Appl 31(3):636–641
11. Morel L, Godfroid H, Mirzaian A, Kauffmann JM (1998) Doubly fed induction machine:
converter optimization and field orientated control without position sensor. Proc Inst Electr
Eng Electr Power Appl 145(4):360–368
12. Saeed SK, Kamal SA, Alshaer MK (2016) Impact of different types of wind turbines generators
on voltage sag. Int J Adv Res Electr Electron Instrum Eng 5(5)
13. Haniotis AE, Soutis KS, Kladas AG, Tegopoulos JA (2004) Grid connected variable speed wind
turbine modeling, dynamic performance and control. In: IEEE PES power systems conference
and exposition. https://doi.org/10.1109/PSCE.2004.1397569
14. “What is wind?". Renewable UK: education and careers. Renewable UK (2010) Archived from
the original on 4 Mar 2011. Retrieved 9 Apr 2012
15. Harvesting the wind: the physics of wind turbines. Retrieved 10 May 2017
16. Chandrasekaran S, Rossi C, Casadei D, Tani A (2013) Improved control strategy for low voltage
ride through capability of DFIG with grid code requirements. Electr Comput Eng 2
17. Devashish AT, Panigrahi S, Behera RR (2016) A review on wind energy conversion system
and enabling technology. In: International conference on electrical power and energy systems
(ICEPES), pp 527–532
18. Singh SP, Prakash T, Singh VP (2019) Coordinated tuning of controller-parameters using
symbiotic organisms search algorithm for frequency regulation of multi-area wind integrated
power system. Eng Sci Technol
19. Prakash T, Singh VP, Singh SP, Mohanty SR (2018) Economic load dispatch problem: quasi-
oppositional self-learning TLBO algorithm. Energy Syst 9(2):415–438
20. Rathore NS, Singh VP, Phuc BDH (2019) A modified controller design based on symbiotic
organisms search optimization for desalination system. J Water Supply Res Technol, AQUA.
https://doi.org/10.2166/aqua.2019.162
A Comparison of Performance of DPC
Technique for DFIG with Using PI
and PID Controller

Sachhidanand Veer Savarkar and Shashi Bhushan Singh

Abstract This paper presents a comparison of the performance of direct power

control of doubly fed induction generator for wind energy conversion system with PI
and PID controller. Direct power control (DPC) technique provides superior perfor-
mance as compared with the classical field-oriented control method. It provides
power quality improvement, improved transient, and steady-state performance. Space
vector modulation (SVM) scheme provides constant switching frequency which
results in reduced oscillation of power and making system performance robust with
the fast response time. At different wind operating condition, the performance of
DFIG is compared with two different controller PI and PID. It is observed that after
MATLAB simulation, PI controller provides permissible transient, good THD, and
power quality improvement.

Keywords Doubly fed induction generator · Space vector modulation technique ·

Direct power control scheme · PI and PID controller

1 Introduction

Increasing energy demand with population has become vital concern. Fossil fuels
are in finite amount, and it is depleting continuously. To make an alternative solu-
tion of the problem, different country has focused on clean and inexhaustible non-
conventional energy sources. There are different forms of renewable energy sources
like solar energy, wind energy, bioenergy geothermal energy, and tidal energy. Out of
all types of renewable energy, wind energy is the main focus field because it generates
higher amount of power for a single unit as compared to solar energy. The Indian
government is also firmly determined for research and development of renewable

S. V. Savarkar (B)
School of Renewable Energy and Efficiency, NIT Kurukshetra, Kurukshetra, India
e-mail: [email protected]
S. B. Singh
Department of Electrical Engineering, NIT Kurukshetra, Kurukshetra, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 117

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_12
118 S. V. Savarkar and S. B. Singh

energy sources and has set up a separate ministry called “The Ministry of new and
renewable energy” (MNRE) to track the progress and review of different renewable
energy project spread over a different part of the country[1]. India’s current total elec-
tricity generation is around 356,817.6 MW, out of which 78,359 MW comes from
renewable energy sources [2]. 35,815 MW power generates from wind energy of the
total installed capacity of renewable energy sources. As the penetration into the grid
has been increased by the distributed generation, some major concern has come out
like power quality issue, distortion in stator current, oscillation in electromagnetic
torque, and higher THD in the system. Such a major concern can harm the wind
turbine, and it will affect the grid badly [3]. To improve the control performance of
the DFIG, various control strategies are applied to the DFIG. The general and basic
control method is field-oriented control (FOC). In FOC scheme, d-axis and q-axis
components of rotor current are regulated to control the stator active and reactive
power independently. As the report suggested by some researcher, FOC scheme is
highly sensitive for uncertainty in the model [4]. To overcome the drawback of the
vector control method, direct torque control (DTC) method came in existence in the
year 1986 [5]. DTC methods have a feature of less parameter dependency and fast
dynamic response. The problem associated with DTC scheme is that it needs an
external loop for controlling the power. Machine’s performance deteriorates during
starting and low-speed operation [6]. To mitigate the drawback of DTC technique,
the DPC method came in existence by T. Noguchi in the year 1998 [7]. Conventional
DPC method uses a predefined switching table for regulating stator active and reac-
tive power. The main problem of conventional DPC method is that it depends on the
stator flux position which highly relies on stator resistance [8]. The new developed
DPC method avoids the employing of switching table and has a robust controller to
achieve good steady-state performance with power quality improvement. To elim-
inate the problem of variable switching frequency, SVM technique is applied with
DPC. It provides constant switching frequency.
In this paper, two different controllers PI and PID have been used in the system
along with DPC scheme and SVM technique. Performance of these two controllers
with DPC scheme is compared, and PI controller proves to be superior method.
Rest of the content of paper is organized in such way that in Sect. 2, modeling and
direct power control technique of DFIG are explained briefly. Section 3 illustrates
the simulation results. Section 4 highlights the conclusion about the obtained results.

2 DFIG Modeling and Direct Power Control Technique

2.1 DFIG Operation, Modeling, and Simulink Diagram

DFIG is a wound rotor induction generator used in wind farm. Basic analogous
diagram of DFIG system is shown in Fig. 1. The stator is in direct connection with
the grid at three-phase supply voltage and frequency. Rotor circuit is connected
A Comparison of Performance of DPC Technique for DFIG … 119

Fig. 1 Analogous circuit diagram of DFIG for wind power application

through the power converter. The magnetic field in rotor is created by three-phase
rotor current of the machine. The stator magnetic field interacts with the induced
magnetic field of the rotor which results to produce torque [9]. Stator magnetic field
rotates at synchronous speed with constant magnitude. The rotor of DFIG performs
at variable speed in order to optimize tip-speed ratio. The gearbox used in the system
increases the rotational speed. In case of weak grid connection, sometimes voltage
fluctuation takes place then DFIG can produce or absorb an amount of reactive power
to control the voltage from the grid.
Complete MATLAB/Simulink diagram of DFIG is depicted in Fig. 2. There is
an inner control loop present in subsystem which is shown in Fig. 2. It evaluates
the equation of stator power. Inductors are connected at stator side to reduce the
ripple content present in the system, and there is a capacitor which maintains the
DC link voltage constant. Active power flows either to the rotor or from the rotor

Fig. 2 Schematic diagram of DFIG modeling

120 S. V. Savarkar and S. B. Singh

Fig. 3. d-q model of DFIG in synchronous reference frame

which further depends on it is absorbing or injecting active power to the grid. In the
supersynchronous mode of operation, power will flow from stator and converter of
rotor circuit to the grid whereas in subsynchronous mode with positive slip condition,
rotor takes the active power from the grid for their rotor winding excitation. The
converter of rotor side circuit controls the speed and torque of the DFIG and also
regulates the power factor at the stator terminal. Converter of grid side circuit can
absorb or generate reactive power for regulating the variation in voltage.
Analogous circuit of DFIG in synchronous d-q axis reference frame is shown in
Fig. 3. It rotates with synchronous speed ωs with the help of space vector notation.
Differential equation of model of DFIG can be derived. Stator and rotor voltage
equation in a synchronous reference frame after multiplying voltage equation by
e− jθs and e− jθr is expressed in following Eqs. (1)–(4).

dψds
u ds = Rs i ds + − ωs ψqs (1)
dt
dψqs
u qs = Rs i qs + + ωs ψds (2)
dt
dψdr
u dr = Rr i dr + − ωr ψqr (3)
dt
dψqr
u qr = Rr i qr + − ωr ψdr (4)
dt
Similarly, stator and rotor flux in synchronous d-q reference frame is expressed
in Eq. (5)–(8).

ψds = L s i ds + L m i dr (5)

ψqs = L s i qs + L m i qr (6)
A Comparison of Performance of DPC Technique for DFIG … 121

ψdr = L m i ds + L r i dr (7)

ψqr = L m i qs + L r i qr (8)

Electromagnetic torque’s expression is shown in Eq. (9).

3 Lm  
Tem = p ψqs i dr − ψds i qr (9)
2 Ls

2.2 Direct Power Control Scheme

Stator power can be varied and controlled by the converter of rotor side circuit.
Generally, two basic concepts are behind of DPC scheme. First one is that stator power
can be regulated by varying rotor flux vector angular position. The second concept is
that controlling of reactive power takes place due to the varying magnitude of rotor
flux vector [10]. Voltage vector evaluated and modulated at the converter terminal is
the main requirement of direct power control. Reference signal is generated through
PI and PID controller for control of stator power relying upon on the difference
between real power and reference power.
Stator active power along with reactive power is calculated by utilizing current
and voltage equation of DFIG as expressed in equation from (10) to (11).

3  
Ps = Re u s × i s∗ (10)
2
3  
Qs = Im u s × i s∗ (11)
2
Stator power can be expressed in d-q axis synchronous reference frame as
simulated in Fig. 4, and it is expressed in Eqs. (12)–(13).

3 
Ps = u ds i ds + u qs i qs (12)
2
3 
Qs = u qs i ds − u ds i qs (13)
2
Actual active and reactive power which has been evaluated as shown in Eqs. (12)–
(13) and is compared with reference stator real power and reactive power. Zero value
is set for reference reactive power with consideration of unity power factor operation.
Difference whatever comes due to stator real power and reactive power is fed to PI
and PID controller, respectively, and it generates DC rotor voltage and eliminates
122 S. V. Savarkar and S. B. Singh

Fig. 4 Calculation of stator power for DFIG

Fig. 5 Gate signal generation

steady-state error in case of PI controller while the speed of response and stability
improves using PID controller. This process of gate signal generation in the control
loop is shown in Fig. 5.

2.3 PI and PID Controller

PI controller is mostly preferred in industrial application due to its some advan-

tages like that it has a simple structure, easy to tune, and low cost. PI controller
A Comparison of Performance of DPC Technique for DFIG … 123

improves steady-state performance and removes forced oscillation present in the

system. When the speed of response is not a major issue, then using PI controller
will be effective [11]. PID controller is used for controller design and in industry. PID
controller provides some benefits like a fast response for changing in controller input.
Magnifying the control signal with suitable action reduces the error and oscillation.

2.4 Space Vector Modulation

Space vector modulation is real-time modulation method, which is the most preferred
method used in DPC scheme of DFIG. It generates a switching signal for controlling
the voltage source converter. It offers enhanced fundamental voltage amplitude and
provides constant switching frequency. The main focus is to control the stator active
power. Sector position is determined by calculated angle, whereas switching signal
will be generated by appropriate voltage vector selection [12].

3 Simulation Results and Discussion

Analysis of DPC scheme of DFIG is exclusively elaborated in this section. 2KW

DFIG machine is simulated in MATLAB/Simulink, and variation of the different
parameter is shown in Figs. 6, 7, 8, 9, 10, 11, 12, 13, 14, and 15. Performances
of DPC technique of DFIG are compared with controller PI and PID for the variation
in wind speed from (6–8) m/s to (8–10) m/s.

Fig. 6 Rotor speed of the DFIG with PID controller

124 S. V. Savarkar and S. B. Singh

Fig. 7 Rotor speed of DFIG with PI controller

Fig. 8 Electromagnetic torque of DFIG with PID controller

From Fig. 6, rotor speed of DFIG is 158.7 rad/s for 0–2 s. And as the wind speed
varies, speed reaches to 160.9 rad/s up to 4 s. From 4 to 6 s, speed is 164.3 rad/s which
is the stable operating condition of the machine. Figure 7 represents the rotor speed
of DFIG, and it is clear that rotor speed is 158 rad/s at the starting condition which
is generating mode of the machine. When the wind speed changes from 8–10 m/s
for time span 2–4 s, then rotor speed changes from 161 to 165 rad/s, which is the
stable operating speed of rotor. It is observed from the speed waveform that it has no
fluctuation. There is very little but more transient at the starting in PID controller as
compared with PI controller.
A Comparison of Performance of DPC Technique for DFIG … 125

Fig. 9 Electromagnetic torque of DFIG with PI controller

Fig. 10 Rotor current of DFIG with PID controller

From Fig. 8, it is clear that generated torque (orange line) follows closely reference
torque (green line) and it changes from 5.45 to 10.6 Nm (which is negative) as the
wind speed varies. Torque attains stable value quickly as speed changes. From Fig. 9,
it can be observed that electromagnetic torque follows closely reference torque. As
the wind speed changes, the torque also changes from 5.8 to 11 Nm and reaches
the stable operating point with negligible oscillation. Comparing these two torque
126 S. V. Savarkar and S. B. Singh

Fig. 11 Rotor current of the DFIG with PI controller

Fig. 12 Active power delivered to the grid with PID controller

waveform with PI and PID controller, it is found that there are more oscillation and
transient at the time of starting in case of PID controller.
Figure 10 illustrates that rotor current and its value change from 1.9 to 3.7 A for
time span 4 to 6 s as the wind speed changes. Rotor current has less oscillation at
starting as compared to stator current but it quickly comes to the stable operating
point just after 0.1 s then as the wind speed changes, finally it reaches 2 A to 3.5 A
from 2 to 4 s as shown in Fig. 11. With a PID controller, rotor current shows more
oscillation and transient at the starting as compared to the PI controller.
A Comparison of Performance of DPC Technique for DFIG … 127

Fig. 13 Active power delivered to the grid with PI controller

Fig. 14 THD in stator current with PID controller

From Fig. 12, 0.725 kW power is transferred to the grid for the time span 2–4 s
and 1.45 kW power is delivered to the grid for time span 4–6 s as the wind speed
changes. 0.7 kW active power is transferred to the grid from 2 to 4 s, and as the wind
speed changes at 4 s, the machine is generating 1.4 kW power from 4 s onward and
finally obtains stable operating point, which can be clearly understood from Fig. 13.
Starting transient is higher with the PID controller as compared to the PI controller
for the active power.
From Fig. 14, it can be observed that THD in stator current is 27.82% with PID
controller, which is above the permissible limit whereas THD in stator current is only
8.40% with PI controller and it is under the permissible limit as shown in Fig. 15.
128 S. V. Savarkar and S. B. Singh

Fig. 15 THD in stator current with PI controller

4 Conclusion

In this paper, the dynamic performance of DFIG-based wind energy system with
direct power control scheme has been examined. Variation in the different param-
eter of DFIG is shown in results as the wind speed varies, and it is observed that
variation in all parameter is under permissible limit. DPC method with PI controller
provides improved steady-state performance and makes power quality improvement.
It minimizes the transient behavior of the system and reduces the complexity of the
control system and distortion, whereas the DPC method with PID controller increases
transient at starting instant for the different variable of the machine.

Appendix

2 kW DFIG Machine’s Parameters which have been used in this paper are given in
Table 1.
Viscous friction coefficient f = 0.00114, K p = 1.5 and K i = 0.7 for first PI
controller and K p = 1.3, K i = 0.5 for second PI controller whereas K d = 0.2 for

Table 1 DFIG simulation parameter

us f Rs L ls Rr L lr Lm j P
380 V 50 Hz 4.85  0.016 H 3.8  0.016 H 0.285 H 0.031 2
A Comparison of Performance of DPC Technique for DFIG … 129

both PID controller with same K p and K i as of PI controller. Switching frequency

f sw of the device is equal to 1/T s Sampling time Ts = 50e−6.

References

1. The Ministry of New and Renewable Energy, https://mnre.gov.in

2. Central Electricity Authority (2019) Ministry of Power: All India Installed Capacity of Power
Stations. www.cea.nic.in
3. Hu J, Zhu J, Dorrell DG (2015) Predictive direct power control of doubly fed induction gener-
ators under unbalanced grid voltage conditions for power quality improvement. IEEE Trans
Sustain Energy 6(3):943–950
4. Dannehl J, Wessels C, Fuchs FW (2009) Limitations of voltage-oriented PI current control of
grid-connected PWM rectifiers with LCL filters. IEEE Trans Industr Electron 56(2):380–388
5. Takahashi I, Noguchi T (1986) New quick-response and high-efficiency control strategy of an
induction motor. IEEE Trans Ind Appl I(5):820–827
6. Zhi D, Xu L (2007) Direct power control of dfig with constant switching frequency and
improved. IEEE Trans Energy Convers 22(1):110–118
7. Noguchi ITT, Tomiki H, Kondo S (1998) Direct power control of PWM converter without
power source voltage sensors. IEEE Trans Ind Appl 34:473–479
8. Errouissi R, Durra AA, Muyeen SM (2017) Offset-free direct power control of DFIG under
continuous-time model predictive control. IEEE Trans Power Electron 32(3):2265–2277
9. Fletcher J, Yang J (2010) Introduction to doubly fed induction generator for wind power
applications. Rijeka, Croatia
10. Rao YS, Laxmi AJ (2012) A comparison of the performance of vector control and direct power
control of induction generator in wind energy convertion system. AASRI Procedia 2:268–274
11. Rao KS, Mishra R (2014) Comparative study of P, PI and PID controller for speed control of
VSI-fed induction motor. Int J Eng Dev Res 2(2):2740–2744
12. Wu B (2006) Multilevel voltage source converters: high-power convertor and AC drives. The
Institute of Electrical and Electronics Engineers (2006)
A Case Study on 24-h Simulation of V2G
System

Robin Chola and Shashi Bhushan Singh

Abstract As the natural resources are going toward an extinction the world is now
running toward renewable resources like sun, water and wind energy resources. An
electric vehicle can be operated and charged by the energy generated from these
renewable resources also. Therefore, the world is now moving toward the future
that is electrical vehicles. Besides having benefits to the growth, transportation and
road efficiency, electric vehicles can also be used to provide power to the utility
systems and benefits to car owners, as they are able to provide ancillary services by
their cars. Electric vehicle connected to the grid can behave either as a load or as a
power source and this system on whole is called as vehicle-to-grid (V2G) when it is
operated as power source. This paper discusses about the services provided by the
vehicle owners after the implementation of the V2G system to the grid and also a
case study on 24-h model of V2G system is done. The impact of solar shading on the
services provided by the vehicles is studied. This analyzed system consists of actual
power generation from a diesel power generator, base load, peak load, EV aggregator
and the renewable energy resources (solar and wind). For different renewable energy
generation in the grid that is by changing the shading of the solar plant, different
scenarios are analyzed.

Keywords Vehicle-to-grid · EV aggregator · Solar shading · Micro-grid · Shading

factor

1 Introduction

As the natural resources on the earth are diminishing day by day, the world is now
running toward renewable sources like solar, wind, water, etc., and at the same time,

R. Chola (B) · S. B. Singh

Department of Electrical Engineering, National Institute of Technology Kurukshetra,
Kurukshetra, Haryana 136119, India
e-mail: [email protected]; [email protected]
S. B. Singh
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 131

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_13
132 R. Chola and S. B. Singh

due to lack of natural energy resources, the fuels for the vehicles is going on the verge
of extinction. So, now the future of vehicles is in electric domain, we can get as much
of electricity from the solar, wind plants and thus these and never going to diminish
and due to this reason only the energy services from these are in spotlight nowadays,
and hence the energy from these plants can be used to feed electric vehicles. Now
these electric vehicles can take power from any source that is from grid, micro-solar
plants that can be established at any place and can also be used to fed power back
to the grid in case of any kind of shortage due to increase in loads, thus, providing
some kind of stability to the system.
Electrical vehicles have always been advantageous over normal internal combus-
tion engines and there are lots of reasons why they are gaining attention of people
over the IC engine vehicle and some of them are, as the fossil fuels which are required
to run the internal combustion engines are diminishing, the prices of the fuels are
reaching heights, and hence affording and using a conventional vehicle is becoming
difficult for the people, combustion of fuel in the conventional vehicles leads to
emission of harmful substances like carbon monoxide—CO, hydrocarbons—HC,
particulate matter—PM and nitrogen oxides—NOx , which obviously you know is
not at all good for the environment and also as production of noise from the elec-
tric vehicle is very less as compared to the conventional vehicles as the number of
moving parts in the electrical vehicles are very less against the IC engine vehicle,
which makes the efficiency of the electrical vehicle three times as of the conventional
IC engine vehicles. Recently, the market of the electric vehicles in Asian countries
and also the rest of the world have been growing rapidly as the maintenance and the
running cost of these vehicles are low against the normal conventional vehicles and
as these vehicles are electrical they can be charged anywhere where electrical power
is available against the headache of refueling the conventional vehicle from the fuel
station time to time. The countries are also bending their policies to attract people
toward the electrical vehicles and benefiting them by introducing different benefits
like ownership benefits, subsidy on purchasing, free EVs charging at the charging
stations, etc. The electric vehicle in the grid acts as a source and as well as load but
one at a time. The increase in number of EVs reduces the generation requirement in
a grid. The battery of EV charges at the time when the demand is low which is most
likely to be at night. The impact of increase in number of EVs in a grid during peak
demand is studied in [1].
In [2], a solution has been described to provide enough energy to charge the
battery in the discounted parking service. Electric vehicles are often referred to as
“prosumers” because they deliver and consume electricity. It is estimated that electric
vehicles use only 10% of their life, leaving 90% of idle time [3]. In this case, the
energy stored in the battery can be fed back into the grid. The best penetration of
electric vehicles requires not only the battery charging program on the swing market,
but also the occasional discharge in times of high power. This is often referred as
a “network vehicle system” referred to as V2G [4]. First, the term V2G is used to
describe a particular type of system that allows power between the network and an
electric vehicle on both sides. The concept here with V2G is that the power is supplied
to the grid when the electrical vehicles are parked at their location or at the charging
A Case Study on 24-h Simulation of V2G System 133

stations. This collaboration of electrical vehicle in grid helps to shed out the peak
load demands and provide stabilization and increase the efficiency of the grid. Active
and reactive power support has been provided by the electrical vehicles which are
working in V2G mode. A comprehensive summary of diverse V2G systems is studied
in [5, 6], When the fleet of EV provides power to the grid simultaneously for truly
extracting the meaning of electrical vehicle and providing benefits of transportation,
then the V2G operation is said to function properly. The EVs are under an aggregator
agent who takes the regulation charges from the grid for every unit of power sent
to the system at the hour of need. An aggregator has a fleet of EVs under him and
provides the ancillary services with the help of the EV fleet. The term EV aggregator
is used to define the subject of controlling battery charging of the electrical vehicle
and may include many electric vehicles. In the grid scenario, the aggregator acts as
a large controlled weight or generation source that can be used as an add-on service.
The objective of this paper is to evaluate the impact of electric vehicles (EV) to
provide better micro-grid services. The second part briefly describes the upcoming
electrical vehicles technology. In Sect. 3, the impact of solar shading on the power
output of the system is studied and analyzed.

2 EV’s as Supplementary Reserve in the Grid

Large admissions of electrical vehicles in the grid will soon change the future of the
traditional grid. EVs have many applications as they provide transportation services
and energy storage options. In V2G operating mode, the EV acts as a storage unit.
The most commonly reported services are micro-operations, voltage control and peak
shaving [5, 6]. At any instant of time, the operation of grid is based on generation
management and the load at that instant. At any instant, the grid basically main-
tains the balance between the generated power and the demand of the loads which
basically makes the grid to work normally. Every imbalance between demand and
production generates a frequency fluctuation in the system. From these points of
view, the constant speed of the synchronous generator can be compromised, making
the power supply system unstable and adversely affecting the accuracy of the clock.
In order to maintain the balance between load and production, the system manager
must obtain various services from the electricity market from other production units.
The viability of V2G system is to support a variety of services. Spinning reserve and
peak power are the features that V2G can support.
EVs are used to deliver the extra burden on the grid at peak hours and that is called
peak shaving. The energy stored in the battery of electrical vehicles which are parked
somewhere and connected to grid is used in the peak load time that is at the time of
high demand. It helps us to reduce the investment of incorporating new conventional
generators at the time of high demand. Thus, also for the control of voltage and the
congestion management, the battery energy of the electrical vehicles ID fed back to
the grid. And also it can provide frequency regulation services.
134 R. Chola and S. B. Singh

3 Simulation Model Description

The simulation model is an example present in the MATLAB [7] library and it consists
of various power generating plants. It consists of a 15 MW diesel generator, a wind
farm, a solar farm with its shading control, load at the end and an EV aggregator,
which communicates with all the electrical vehicles to provide the ancillary services
to the grid in the time of need. The model is an isolated power system.
The diesel generator, which is capable of generating 15 MW of power, is the
only dependable sources here, which means that it is not affected by any kind of
weather conditions and is totally independent of the wind, rain and all. There are two
renewable energy sources present in the system or you can call the combination as
a micro-grid. Photovoltaic cell-based solar panels which makes a solar plant which
generate energy from solar radiation at the time sun is shining, and a wind farm
which has a wind turbine, that is it generates energy from the wind, the more is the
velocity of the wind the more is the power generated by the wind turbine, also rated
capacity of PV farm is taken as 4.5 MW, and that of wind turbines is 8 MW.
In this simulated model, the number of aggregated EVs is taken to be 200 and
they operate in different profiles that is an amount of electrical vehicles will provide
the ancillary services at a particular period of time and on the basis of these they are
divided into different profiles and different time periods. The required rated power
for charging for the EVs which are 200 in number is 200 × 16 kW [8]. The required
charging power 16 kW for each EV resembles to the level two chargers, this type of
chargers offer charging from 230 V AC plug directly and also it can be used anywhere
to charge where the 230 V AC charging sockets are available. Further, these level
two chargers can be used to charge any type of EV and PHEV. It is assumed that
the charging facilities are available at every household and offices and thus they
are divided into five different scenarios which are listed below. The characteristics
profile of each user is as follows,
Profile 1: People who work at offices from 9 to 5 pm, and use their vehicles for
3 h in this period with availability of charging facilities near offices and homes
represents 17.5% of the total EVs.
Profile 2: People who work at offices from 9 to 5 pm, and drive their vehicles for
4 h in this period with availability of charging facilities near offices and homes
represents 12.5% of the total EVs.
Profile 3: People work at offices from 9 to 5 pm, and drive their vehicles for 3 h
during this period without availability of charging facilities near their offices do
represents 10% of the total EVs.
Profile 4: People who does not even drive their vehicles the whole day and their
cars are connected to the grid all the time represents 30% of total EVs.
Profile 5: People work at offices from 20 pm to 4 am, and drive their vehicles
for 2 h in this period without availability of charging stations near offices at all
represents 17.5% of the total EVs (Fig. 1).
A Case Study on 24-h Simulation of V2G System 135

Fig. 1 24-h simulation of V2G system

3.1 Simulation Description and Different Scenarios

This simulated model is simulated for the whole of 24-h. Solar plant being the most
uncertain one for generation of power as there is no guarantee of when the shading of
solar plant will occur that means the solar plant will not be able to produce a constant
power when needed thus the solar shading (shading factor) is being considered here
and is used to see the performance of the system. At the time of noon, there is
significant amount of shading which will distress the power output from the solar
plant. Shading factor represents the amount of solar plant is shaded in ratio to the
total area of the solar plant. There is an impact on power generation due to the solar
shading of the PV farm and also when different number of EVs used in the system.
Also, the impact of shading is being studied on the different factors of the grid like
the power generation and the frequency of the system. The main here is to maintain
the stability between demand and generation.
Figure 2 represents the 24-h simulation model of a micro-grid along with the
declined output of PV panels which occurs during the noon that is at 12:00 pm and
remains for 10 min when the shading factor is considered to be 0.5 [9]. By changing
the shading factor or the time of shading, the output of the total PV panel varies
and hence the total output varies. Thus, the EV fleet connected to the grid will have
to transfer the excess power, as the output of the wind would be constant and load
being the same; therefore, the EV fleet or electrical vehicle in general can be used to
provide the regulation services or the excess demand.
136 R. Chola and S. B. Singh

Fig. 2 Solar shading and its duration

4 Simulation Results

The power outputs of the different plants that is the solar, wind, diesel generator
are obtained from the simulation and are displayed below at shading factor equal
to 0.5. By changing the shading factor only, the output of the PV farm changes and
hence the total output of the grid changes. Also, the SOC of the electrical vehicles,
regulation and charge parameters have been monitored at the same shading factor.
Also, by changing the shading factor, the behavior of the grid can also be studied,
which basically means that the total power output variation with respect to the shading
factor, as shading factor only changes the power out of the PV farm and rest producing
the same output (Fig. 3).
This plot shows the power produced by all the power-producing elements in the
grid. The first plot is the total power produced by all the components of the grid.
The red-colored one that is the second graph is the power generated from the diesel
generator. The blue graph is the output of the wind farm, whose output totally depends
on the profile of the wind of the area. The last one that is magenta graph is the out of
the solar farm which clearly shows a dip at the time of shading of the PV farm. The
output of the PV farm clearly drops to almost a half when the half of the PV farm is
shaded that is when the shading factor is equal to 0.5.
Figure 4 is the load curve which resembles the output of the grid that is, the graph
of total power generated in the micro-grid and this should also be the case as the
demand should be equal to the supply. The power produced should be equal to the
power needed and at the time of dip in the output of the solar farm or the shading of
the PV Farm, the EV fleet that is the EV aggregator is doing is work that is providing
the regulatory power services to the grid and thus maintaining stability in the system,
which means that the vehicles are charging and discharging the whole time but the
A Case Study on 24-h Simulation of V2G System 137

Fig. 3 Power generated by different plants and total power

Fig. 4 Load curve

vehicles connected to the grid at the time of power dip would be providing the power,
thus, the state of charge of the vehicles should be the thing to monitor here and the
same is shown for the different profile in the EV fleet. Figure 5 shows the state of
charge of the different profiles of the EV fleet under the EV aggregator. The state
of charge varies between −ve 100% and +ve 100%, where negative value of SOC
shows that the EV fleet is doing the vehicle-to-grid operation and otherwise G2V.
Figure 6 shows different parameters like voltage, current and different powers of an
138 R. Chola and S. B. Singh

Fig. 5 SOC of different EV profiles

EV fleet.

5 Conclusion

A successful study of a 24-h simulation of vehicle-to-grid system is done and how

the different parameters of the grid vary by changing the different parameters like
solar shading, number of electrical vehicle under an EV aggregator or in the EV
fleet is analyzed and studied. The more the number of vehicles connected in the grid
under a EV aggregator, more ease it will provide to the grid and the more it will help
to shave out the peak load because we cannot be dependent on the output of wind
plants and the output of the diesel generators does not vary a much.
A Case Study on 24-h Simulation of V2G System 139

Fig. 6 Voltage current and power profile of EV fleet

References

1. Khan AZ, Janjua AK, Jan ST, Ahmed ZN (2017) A comprehensive overview on the impact of
widespread deployment of electric vehicles on power grid. In: IEEE international conference on
smart grid and smart cities (ICSGSC) 195–199
2. Fachechi A, Mainetti L, Palano L, Patrono L, Stefanizzi ML, Vergallo R, Chu P, Gadh R (2015) A
new vehicle-to-grid system for battery charging exploiting IoT protocols. In: IEEE international
conference on industrial technology (ICIT) 17–19
3. Amditis A, Brusaglino G, Spessa E (2015) Optimizing vehicle-to-grid electric energy system
with new technologies. In: International conference on clean electrical power (ICCEP) 353–359
4. Zhou BW, Littler T, Wang HF (2013) The impact of vehicle-to-grid on electric power systems:
a review. In: 2nd IET renewable power generation conference 1–4
5. Mehta R, Trivedi A, Yang J (2017) Hybrid planning method based on cost-benefit analysis
for smart charging of plug-in electric vehicles in distribution systems. IEEE Trans Smart Grid
523–534
140 R. Chola and S. B. Singh

6. Zhou X, Zou L, Ma Y, Gao Z (2017) Research on impacts of the electric vehicles charging and
discharging on power grid. In: 29th Chinese control and decision conference (CCDC) 1398–1402
7. 24-h simulation of a vehicle-to-grid (V2G) system. A MATLAB example 1–1
8. He Y, Venkatesh B, Guan L (2012) Optimal scheduling for charging and discharging of electric
vehicles. IEEE Trans Smart Grid 3(3):1095–1105
9. Vincenzo MCD, Infield D (2013) Detailed PV array model for non-uniform radiance and
validation against experimental data. Sol Energy 97:314–331
A Review on Bidirectional DC-DC
Converters for V2G and G2V
Applications

Robin Chola and Shashi Bhushan Singh

Abstract The interest in DC-DC converters has always been growing for the inter-
facing of electrical vehicles, PV farms and wind farms with the main grid. This paper
reviews all the basic types of DC-DC converters which are divided under the cate-
gories called isolated and non-isolated bidirectional DC-DC converters. Each type of
converter is described and illustrated briefly. This paper intends to be useful for the
future users of DC-DC converters, and also the most efficient and flexible converter
is being singled out from this array of DC-DC converters.

Keywords DC-DC converter · Isolated BD converters · Non-isolated BD

converters

1 Introduction

The extinction of natural resources has made the renewable energy resources more
prominent, and the power generation from the renewable energy resources like solar
and wind which are readily and easily available is becoming popular and imple-
mented. The system based on wind and solar sources is used to supply power
for various applications like lighting, heating, etc. And almost every system here
has storage component in it which are batteries. Only DC power is stored in these
batteries, and these powers can be regulated with the help of bidirectional DC-DC
converters. Thus, looking for the best DC-DC converter is the prime motive, and we
have to do the study for the different types of bidirectional DC-DC converters; and so
this review study is being done. Also, as the electrical vehicles are gaining limelight
and as the power sources of these are only batteries, hence it has a use of DC-DC
converter in them. Now these electric vehicles can take power from any source that

R. Chola (B) · S. B. Singh

Department of Electrical Engineering, National Institute of Technology Kurukshetra,
Kurukshetra, Haryana 136119, India
e-mail: [email protected]; [email protected]
S. B. Singh
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 141

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_14
142 R. Chola and S. B. Singh

is from grid, microsolar plants that can be established at any place and can also be
used to fed power back to the grid in case of any kind of shortage due to increase in
loads, thus providing some kind of stability to the system. Electrical vehicles have
always been advantageous over normal internal combustion engines, and there are
lots of reasons why they are gaining attention of people over the IC engine vehicle,
and some of the reasons of their interest are the fossil fuels which are required to run
the internal combustion engines are diminishing. The prices of the fuels are reaching
heights, and hence affording and using a conventional vehicle is becoming difficult
for the people. Combustion of fuel in the conventional vehicles leads to emission
of harmful substances like carbon monoxide (CO), hydrocarbons (HC), particulate
matter (PM) and nitrogen oxides (NOx ), which obviously you know is not at all good
for the environment. The production of noise from the electric vehicle is very less as
compared to the conventional vehicles as the number of moving parts in the electrical
vehicles is very less against the IC engine vehicle, which makes the efficiency of the
electrical vehicle three times as of the conventional IC engine vehicles. Recently, the
market of the electric vehicles in Asian countries and also the rest of the world has
been growing rapidly as the maintenance and the running cost of these vehicles are
low against the normal conventional vehicles, and as these vehicles are electrical, they
can be charged anywhere where electrical power is available against the headache of
refueling the conventional vehicle from the fuel station time to time. The countries
are also bending their policies to attract people toward the electrical vehicles and
benefiting them by introducing different benefits like ownership benefits, subsidy on
purchasing, free EVs charging at the charging stations, etc. The electric vehicle in
the grid acts as a source and load but one at a time. The increase in number of EVs
reduces the generation requirement in a grid [1]. The battery of EV charges at the
time when the demand is low which is most likely to be at night. The term V2G is
used to describe a particular type of system that allows power between the network
and an electric vehicle on both sides. The concept here with V2G is that the power is
supplied to the grid when the electrical vehicles are parked at their location or at the
charging stations. This collaboration of electrical vehicle in grid helps to shed out
the peak load demands and provide stabilization and increase the efficiency of the
grid. Active and reactive power support has been provided by the electrical vehicles
which are working in V2G mode.
This paper briefs all the types of bidirectional DC-DC converters. The DC-DC
converters are divided under two categories, and those are isolated and non-isolated
bidirectional DC-DC converters (BDCs) [2], and further these isolated [3–6] and non-
isolated converters [7–11] are classified into three types and is explained in brief.
Thus, conclusion is made that which DC-DC converter is best used in application.
A generic diagram (Fig. 1) shows that how DC-DC converter is used in different
applications [7] and also how the converters are implemented in the application
of electrical vehicles in Fig. 2. In electric vehicle application, the DC power from
the battery is regulated with the help of DC-DC converter which is bidirectional in
nature, and then, the regulated power is fed to the DC-AC bidirectional converter
which further feds the AC power from the converter to the motor of the electrical
vehicle and thus the vehicle functions.
A Review on Bidirectional DC-DC Converters for V2G and G2V … 143

Fig. 1 Use of DC-DC converters in different applications

Fig. 2 DC-DC converter in electrical vehicle

Figure 1 shows the implementation of DC-DC converters in different application

and how the converters are implemented with the maximum powerpoint trackers
[12–14] and how the different systems are synchronized with the common DC bus
[5].

2 Bidirectional DC-DC Converters

A DC-DC converter is a device which converts the DC voltage from any input;
be it battery or output of any other circuitry into a different DC voltage level. The
main application is to change the voltage levels of DC. They are commonly used
automobiles, televisions, etc. There are some devices that need a specific DC voltage
to run itself, power access to the required power can destroy the device, or less power
will not make the device to run properly or run at all. Thus, the converter takes the
power from its source and clips or clamps the voltage level to the required value.
The bidirectional DC-DC converter can be categorized into two different types [2],
144 R. Chola and S. B. Singh

Fig. 3 Classification of bidirectional DC-DC converters

i.e., bidirectional isolated DC-DC converters and bidirectional non-isolated DC-DC

converters which are also illustrated in Fig. 3.
The classification of bidirectional DC-DC converters is illustrated in Fig. 3.
The BDCs are divided into two subcategories which are non-isolated bidirectional
DC-DC converter and isolated DC-DC converter on the basis of electrical isolated
provided to them. The non-isolated bidirectional DC-DC converters are categorized
into three subtypes which are single phase with auxiliary ZVS/ZCS circuitry, single-
phase non-isolated BDC and interleaved BDCs [15–17], and further the interleaved
BDC [18, 19] is bifurcated into two categories, and those are uncoupled and coupled.

2.1 Non-isolated BDCs

The non-isolated BDC is the converter which transforms less or lacks in isolation.
In these types of power conversion systems, buck-and-boost DC-DC converters are
used. As the converters in this category are transformerless, it makes the weight of
the overall system lesser and increases the efficiency of the system. These types of
converters can be used in spacecraft systems where weight plays a significant role
and also in high-power applications where the main concern is size. It can be seen
from Fig. 3 that the main types of non-isolated bidirectional converters are classified
below. The basic types of non-isolated BDCs are buck, boost, buck boost and Cuk
converters [20–24], and a good amount of research and studies have already being
done on these converters so the types of non-isolated converters are not illustrated
here. The Cuk converter [25] here is the basic topology for the converters which are
followed in this paper.
A Review on Bidirectional DC-DC Converters for V2G and G2V … 145

Fig. 4 SEPIC/Zeta converter

2.1.1 SEPIC/Zeta

The schematic diagram of a bidirectional Zeta DC-DC converter is shown in Fig. 4.

This converter in the forward power flow behaves as a SEPIC converter; in the reverse
power flow, that is the backward power flow, it behaves and functions like a Zeta
converter, and hence, it is called SEPIC/Zeta converter [26]. This is nothing but a new
arrangement of the Cuk converter. This converter like the Cuk converter is able to
produce both higher and lower output voltage levels compared to the input provided
and in both the directions, but the polarity will be the same. Now during the forward
power flow through the converter, it acts as buck converter. Switch T1 is off during
the charging period, while the switch T2 acts as the power switch. In backward power
flow, the converter behaves as boost converter; in this mode, switch T1 is off and T2
acts as a power switch. The two inductors are to reduce the voltage fluctuations and
the stress on the switches due to the switching.

2.1.2 Topology of BDC with Coupled Inductor

The circuit diagram for this BDC is shown in Fig. 5. A coupled inductor is imple-
mented in this bidirectional DC-DC converter with the same number of turns in both
primary and secondary sides. There are several advantages of this converter over the
conventional converter, and those are as follows [27]: lower switching current for
same electric specs and higher gain ratio. In stepdown mode of this converter, the
switch T1 is controlled by the PWM scheme in which the other two switches, i.e.,
T2 and T3, act as the part of the synchronous V in rectifiers. In stepup mode, both
the switches T2 and T3 are controlled by pulse width modulation, while the switch
T1 acts as a normal switch.
It has three switches and two inductors which are magnetically coupled to each
other; two capacitors are used to filter out the fluctuations and the harmonics in the
input as well as in the output; below is the circuit diagram for this configuration.
146 R. Chola and S. B. Singh

Fig. 5 BDC with coupled inductors

2.1.3 BCD with Tapped Inductor

The use of tapped inductors in the converter helps us achieve high stepup ratio and
avoiding high peak current at the time of switching. Use of tapped inductor also leads
to a lower stepdown ratio [7]. The voltage polarity of the DC bus is opposite w.r.t.
ground which is not wanted in many applications of these BCDs.
There are two switches T1 and T2 one inductor which is being tapped and divided
into two which means it is kind of magnetically coupled and two capacitors to filter
out the voltages. For the buck mode of operation, the switch T1 gets on and switch T2
is off and vice versa for the boost mode of operation. The switches here are controlled
by pulse width modulation.

2.1.4 BDC with Switched Capacitor

The circuit diagram for the bidirectional DC-DC converter implemented using
switched capacitor [28] is shown in Fig. 6. The capacitors basically enable a phase
for charging of another capacitor from the supply. They do not have a good regu-
lation, also the ratio of voltage conversion is totally dependent on the topology of
the system, and also high input ripple currents are there which electromagnetically
disturbs other signal and hence electromagnetic interference, but these disadvantages
can be overcome by a current and voltage control scheme which increases the cost
of the system way too much and also the complexity. As they are free from magnetic
devices, these converters can be realized by IC technology (Fig. 7).
A Review on Bidirectional DC-DC Converters for V2G and G2V … 147

Fig. 6 BDC with tapped inductor

Fig. 7 BDC with switched capacitor

2.1.5 Single-Phase BCD with ZVS/ZCS Circuitry

The basic need from a converter is that it should be able to produce the desired
or the wanted output, it should have high efficiency, lesser complexity and better
reliability, and also its control method should be simple and all these are achieved
with these auxiliary ZVS/ZCS circuitry added BCD [29]. Soft switching techniques
of the components reduce the losses occurring in switching and thus increase the
stability and efficiency of the system. A half bridge single-phase BDC with ZVS and
ZCS is shown in Fig. 8. The SEPIC/Zeta and SC converter can also be implemented
with ZVS and ZCS [30].
148 R. Chola and S. B. Singh

Fig. 8 Single-phase BCD with ZCS

The portion highlighted in the red in Fig. 8 is the auxiliary circuit for the converter;
in this case, it is ZCS but it can be any of ZVS or ZCS.

2.1.6 Interleaved Converters

The converter implementing interleaving has several advantages over the other
converters, and those are better efficiency, good thermal scattering, splitting of
currents (which reduces conduction losses), cancellation of current ripples and also
has high power density. Giving high efficiency in almost every range of input is the
main show off of these types of converters [31, 32]. Interleaving of converters can
be floating also, which further enhances the functioning of the system and a higher
conversion ratio can be achieved with floating interleaving. These systems are more
cost-effective than that of conventional BCDs as they have higher ratio of conversion,
higher efficiency, low ripples in current and also reduced size of inductors and capac-
itors which makes it more compact in size and lighter in weight. A basic diagram for
the floating interleaved BDC is shown in Fig. 9.
Table 1 compares the characteristics of different NBDs like the conversion ratio
in both the modes that is buck and the boost, number of switches required, number
of passive components required and also the ripple current.
A Review on Bidirectional DC-DC Converters for V2G and G2V … 149

Fig. 9 Floating interleaved BDC

2.2 Isolated Bidirectional DC-DC Converter

The isolated BDC has a full-bridge inverter and also a rectifier (full-bridge). The
circuit diagram of the basic isolated bidirectional DC-DC converter [33–37] is shown
in Fig. 5. Basically in DC-DC converters, only two types of operations take place
that is either stepup operation or a stepdown operation which are buck and the boost
operation. In buck mode of operation, the output voltage level is lower than that
of input voltage level, and in boost mode of operation, the output voltage level is
more than that of the input voltage. Transformer is used for the isolation and barrier
between the rectifier and the inverter and also for the transmutation of power from
the higher voltage side to the lower voltage side and also vice versa. Cost reduction
and higher efficiency of the system are possible in non-isolated BDCs as they do
not have transformer in it which reduces the weight, cost of the overall system and
also enhances the efficiency of the system. But in these types of converters, there is
a possibility for leakage of current to flow and also the switching losses occurs as
the voltage of input and the output sides is very dissimilar. Therefore, isolated BDCs
have a main application where the voltage dissimilarity is huge (Fig. 10).
The two elementary circuits in isolated bidirectional DC-DC converters are the
bidirectional inverter and the rectifier. For the isolation, we need a high-frequency
transformer and that is possible only when we are dealing with AC and that is possible
only when the DC supply is converter from the AC to the DC with the help of the
inverter. The output of the inverter is maintained at a particular level and frequency. A
full-bridge topology is implemented here which is also bidirectional that is it can also
convert the AC input into the DC output. Following a high-frequency transformer and
 150

Table 1 Characteristics comparison of NBDCs

Topology Voltage conversion Voltage conversion Output current Number of switches Number of passive Magnetic required
ratio V bat /V in (buck ratio V in /V bat (boost ripple (buck mode) components
mode) mode)
D D Vin D
SEPIC/Zeta (Fig. 4) 1−D 1−D L 1 f SW 2 5 Coupled or two
inductors
D 1+D V0 D
Modified BDC with 2−D 1−D (1+k)L f SW 3 4 Coupled inductors
coupled inductors
(Fig. 5)
D 1−D+n D n 2 Vbat (1−D)
BDC with tapped n= 1−D L f SW 2 3 Single-tapped
D+n(1−D) ,
inductor (Fig. 6) in
inductor
1 + NNbat
Switched capacitor 0.5 2 C f SW (Vin − Vbat ) 4 3 No
(Fig. 7)
1 Vbat (1−2D)
Interleaved (Fig. 9) D 1−D L f SW 2N 2+N N inductors with or
without coupling
R. Chola and S. B. Singh
A Review on Bidirectional DC-DC Converters for V2G and G2V … 151

Fig. 10 Isolated bidirectional DC-DC converter

after which a single-phase full-bridge rectifier is placed which performs the rectifier
action, which is converting the AC input given to it into the DC.

3 Conclusion

Bidirectional DC-DC converters are gaining limelight as they are used in solar,
wind farms and also in the electrical vehicles, and they are future. The different
types of bidirectional DC-DC converters are briefly studied here, and they are being
compared. Advantages and disadvantages of each type of DC-DC Converter are
pointed out. The comparison of different non-isolated DC-DC converters is carried
out in Table 1. Resulting from the review, the half-bridge topology and its interleaving
variations with coupled inductors are the most efficient and promising in the non-
isolated BDC, while if there is a need of isolation we have no choice then going to
the only configuration of isolated BDC.

References

1. Khan AZ, Janjua AK, Jan ST, Ahmed ZN (2017) A comprehensive overview on the impact of
widespread deployment of electric vehicles on power grid. In: IEEE international conference
on smart grid and smart cities (ICSGSC), pp 195–199
2. Alassi A, Al-Aswad A, Gastli A, Brahim LB (2017) Assessment of isolated and non-isolated
DC-DC converters for medium voltage PV applications. In: IEEE GCC conference and
exhibition (GCCCE), pp 1–6
3. Ramteke DS, Gaikwad MB (2018) Isolated DC-DC converter fed DC motor for bidirectional
electrical vehicular applications. In: International conference on smart electric drives and power
system, pp 33–37
4. Ikeda S, Kurokawa F (2017) Isolated and wide input ranged boost full bridge DC-DC converter
with low loss active snubber. In: IEEE energy conversion congress and exposition (ECCE), pp
2213–2218
5. Ge J, Huang X, Xie H (2015) Fast equivalent model of isolated bidirectional DC-DC converters
for microgrid study. In: 41st annual conference of the IEEE industrial electronics society
(IECON), pp 32–35
6. Chen Y, Zhao S, Li Z, Wei X, Kang Y (2017) Modelling and control of the isolated DC-DC
modular multilevel converter for electric ship medium voltage direct current power system.
IEEE J Emerg Select Top Power Electron 5(1):124–139
152 R. Chola and S. B. Singh

7. Tytelmaier K, Husev O, Veligorskyi O, Yershov R (2016) A review of non isolated bidirectional

DC-DC converters for energy storage systems. In: International young scientists forum on
applied physics and engineering. IEEE, pp 1–7
8. Jeong H, Kwon M, Choi S (2017) A high gain non isolated soft switching bi-directional DC-DC
converter with PPS control. In: IEEE energy conversion congress and exposition (ECCE), pp
1723–1727
9. Reddy BM, Samuel P (2016) A comparative analysis of non-isolated DC-DC converters. In:
IEEE 1st conference on power electronics, intelligent control and energy systems (ICPEICES),
pp 1–6
10. Khan MYA, Saeed L, Saleem J, Arif M, Majid A (2018) A high gain multi-port bidirec-
tional non-isolated DC-DC converter for renewable integration. In: International conference
on computing, mathematics and engineering technologies (ICOMET), pp 1–6
11. Shayeghi H, Pourjafar S, Sedaghati F, Eghbali O (2019) A non-isolated based boost DC-
DC converter with VM technique and soft switching capability. In: 10th international power
electronics, drive systems and technologies conference (PEDSTC), pp 92–101
12. Patil SN, Prasad RC (2015) Design and development of MPPT algorithm for high efficient
DC-DC converter for wind energy system connected to grid. In: International conference on
computer, communication and control (IC4), pp 1–7
13. Patil SN, Prasad RC (2015) Design and development of MPPT algorithm for high efficient
DC-DC converter for solar energy system connected to grid. In: International conference on
energy systems and applications, pp 228–233
14. Kumar K, Ramesh N, Prabhu KR (2017) Analysis of integrated boost-Cuk high voltage gain
DC-DC converter with RBFN MPPT for solar PV application. In: Innovations in power and
advanced computing technologies (i-PACT), pp 1–6
15. Xu J, Kang L, Wang J (2011) Application of DC-DC converter in hybrid power system of EV.
In: International conference on electric information and control engineering, pp 4014–4017
16. Veeresh H, Kusagur A (2016) Novel advanced switching technique ZVS/ZCS fed bidirectional
DC-DC topology to Evs. In: International conference on signal processing, communication,
power and embedded system (SCOPES), pp 613–617
17. Lin B-R, Jhong J-Y (2011) Implementation of an interleaved ZVS/ZCS DC/DC converter. In:
IEEE region 10 conference (TENCON), pp 853–857
18. Sedaghati F, Azizkandi ME, Majareh SHL, Shayeghi H (2019) A high-efficiency non-isolated
high-gain interleaved DC-DC converter with reduced voltage stress on devices. In: 10th
international power electronics, drive systems and technologies conference (PEDSTC), pp
729–734
19. Fernanda LAM, Vladimir CDG, Manuel TJ, Héctor BB, Erwin ZT, Felix QM (2015) Compar-
ative analysis of an DC-DC interleaved converter of high power and control techniques. In:
IEEE workshop on power electronics and power quality applications (PEPQA), pp 1–5
20. Wu T-F, Chen Y (1998) Modelling PWM DC/DC converters out of basic converter units. IEEE
Trans Power Electron 13(5) 870–881
21. Bubovich A (2017) The comparison of different types of DC-DC converters in terms of low-
voltage implementation. In: 5th IEEE workshop on advances in information, electronic and
electrical engineering (AIEEE), pp 1–4
22. Sahoo M, Siva Kumar K (2014) High gain step up DC-DC converter for DC microgrid appli-
cation. In: 7th international conference on information and automation for sustainability, pp
1–5
23. Mohammed HE, Fardoun AA (2016) High gain DC-DC converter for PV applications. In: 59th
international midwest symposium on circuits any systems (MWSCAS), pp 1–4
24. Coelho RF, Concer F, Martins DC (2009) A study of the basic DC-DC converters applied in
maximum power point tracking. In: Brazilian power electronics conference (2009), pp 673–678
25. Maroti PK, Padmanaban S, Wheeler P, Blaabjerg F, Rivera M (2017) Modified high voltage
conversion inverting Cuk DC-DC converter for renewable energy applications. In: IEEE
Southern power electronics conference (SPEC), pp 1–5
A Review on Bidirectional DC-DC Converters for V2G and G2V … 153

26. Dhivya M, Shanmugapriya P (2016) A high step-up three port SEPIC & Zeta converter for
stand alone photovoltaic systems. In: International conference on computation of power, energy
information and communication (ICCPEIC), pp 419–424
27. Dhanasenkaran R, Radha Lakshmi K (2016) Design and analysis of coupled inductors in inter-
leaved three-level DC-DC converter. In: International conference on advanced communication
control and computing technologies (ICACCCT), pp 827–831
28. Axelrod B, Berkovich Y, Loinovici A (2008) Switched capacitor/switched inductor structures
for getting transformerless hybrid DC-DC PWM converters. IEEE Trans Circ Syst 55(2):687–
696
29. Sunrahmanya Kumara Bhajan VV, Rama Reddy S (2009) A novel ZVS-ZCS bidirectional
DC-DC converter for fuel cell and battery application. In: International conference on power
electronics and drive systems (PEDS), pp 12–17
30. Fathy K, Morimoto K, Doi T, Lee HW, Nakaoka M (2006) An asymmetrical SC and lossless
inductor quasi-resonant snubber-assisted ZCS-PWM DC-DC converter with high frequency
link. In: CES/IEEE 5th international power electronics and motion control conference, vol 2,
pp 124–131
31. Habumugisha D, Choedhury S, Chowdhury SP (2013) A DC-DC interleaved forward converter
to step up DC voltage for DC microgrid applications. In: IEEE power and energy society general
meeting, pp 1–5
32. Thiyagarajan A, Praveen Kumar SG, Nandini A (2014) Analysis and comparison of conven-
tional and interleaved DC-DC converter. In: Second international conference on current trends
in engineering and technology (ICCTET), pp 198–205
33. Mehdipour A, Farhangi S (2011) Comparison of three isolated bi-directional DC-DC converter
topologies for a backup photovoltaic application. In: 2nd international conference on electric
power and energy conversion systems (EPECS), pp 1–5
34. Matsumori H, Kosaka T, Sekido K, Kim K, Egawa T (2019) Isolated DC-DC converter utilizing
GaN power device for automotive application. In: IEEE applied power electronics conference
and exposition (APEC), pp 1704–1709
35. Jing P, Wang C (2012) Analysis of isolated three level half bridge bidirectional DC-DC
converter. In: IEEE international symposium on industrial electronics, pp 194–199
36. Modepalli K, Suryadevara R, Parsa L (2016) High frequency isolated DC-DC converter for
offshore wind energy systems. ECCE, pp 1–6
37. Ramteke DS, Gaikwad MB (2018) Isolated DC-DC converter fed DC motor for bidirectional
electric vehicular application. In: International conference on smart electric drives and power
system (ICSEDPS), pp 33–37
Grid Integrated with Wind Turbine
System on Fault Analysis Studies

Ravneet Kaur, Deepika Bhalla, and Naveen Kumar Sharma

Abstract Wind is a renewable and non-polluting energy resource. This resource of

energy has gained fame in the previous few years and is essentially used for supplying
power for distribution systems. Time tested protective systems are available to protect
the conventional power generation equipment. The need of the hour is to address the
issues that arise when the wind turbine system is connected to the grid. The wind
speed and velocity are variable parameters, and there are times when the wind speed
is too high and the wind turbine system needs to be disconnected from the grid.
Also, faults are unavoidable in power systems. In this paper, work has been carried
out to study the protection of wind turbine generator against three faults namely:
over voltage, under voltage, and high speed. Study been carried out for the generated
power by a wind turbine generated through simulations. Distributed wind turbine
systems that are integrated with grid have been considered. It is found that in the
proposed protection disconnects the wind turbine system from the grid when the over
voltage, under voltage, and high-speed wind turbine rotating speed faults occur.

Keywords Wind turbine system · Over voltage fault · Under voltage fault · Over
speed of wind turbine

R. Kaur (B) · D. Bhalla

Department of Electrical Engineering, IKG Punjab Technical University, Main Campus,
Jalandhar, Kapurthala, Punjab 144 603, India
e-mail: [email protected]
D. Bhalla
e-mail: [email protected]
N. K. Sharma
IKG Punjab Technical University, Batala Campus, Batala, Punjab, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 155

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_15
156 R. Kaur et al.

1 Introduction

The global population and their energy load demand are rapidly increasing, conven-
tionally means to satisfy the rapidly increasing energy demands require revalua-
tion. Wind is clean and eco-friendly source of energy. The major reason of global
warming is the carbon dioxide, resultant the greenhouse gas emission. Wind energy
has attained considerable global attention for generation of electric power due to
the global environmental concerns associated with fossil energy resources [1]. For
the industries of power generation, wind energy is capable and hopeful renewable
energy. In various nations, the renewable resources are better option to meet the
load demand by consumers [2]. On the energy-saving, a lot of energy policies and
research projects work have been done in the last two decades [3]. To fulfil the load
demand by the consumer, the alternative sources are helping us. The generation of
electricity in India with the help of renewable sources like solar, wind, biogas, tidal,
etc., have attained a considerable contribution. But the main difficulty is the system
stability: to disconnect the renewable wind turbine system from the grid when a fault
occurs. In this paper, the protection disconnects the wind turbine from the grid when
a fault occurs. The various types of wind turbines are being used practically, and
they contribute to various power quality issues in the system [4]. The wind turbine
uses an induction generator for power generation. If the number of incoming induc-
tion generators are increasing into the grid. The power quality problems are mainly
current harmonics, power factor, and reactive power; these are increasing due to the
increasing number of wind energy conversion systems. These issues are harmful for
weak grids [5–8]. To reduce the power quality at the consumer’s end, power losses
increases, and malfunction in communication system are cause of the harmonics
effect in the power system [9]. The small signal performance of a grid connected
doubly fed induction generator wind turbine generator has been reported elsewhere,
with focus on the electrical generator [10–14]. To model of wind, mechanical and
electrical systems different tools such as TurbSim, FAST, and MATLAB/Simulink
have been used, respectively, [15, 16] while the active and passive structural control
of floating offshore wind turbine has been investigated by using the tool ‘FAST’ [17].

2 Overview of Faults in Power System

Faults are unpredictable events produced in power system which leads to malfunc-
tioning of power system equipment and over voltages of buses and lines. Due to this,
power loss increases resulting violation of the voltage limit. This causes outages and
at times black-out of the power system. If there is continuous power loss in power
system operation, more power needs to be generated to meet the demand making it
uneconomical. In this work, the appended faults have been considered: AC under
voltage (positive-sequence), AC over voltage (positive-sequence), under speed, over
speed. In case of these four faults, the isolation of faulted area is possible. In electrical
Grid Integrated with Wind Turbine System … 157

system, the electrical disturbance causes fluctuation in the electromechanical torque

and speed of doubly fed induction generator (DFIG) of a wind farm which in turn
may affect the structural dynamics of the wind turbine [18]. The fluctuations have a
harmful impact on power quality and system stability in electrical system [3]. Several
damage levels have been recorded resulting occasionally from the drawbacks of the
power system protection [19].
As a result, simple and non-integrated protection methods of wind farm providers
are still being utilized [20, 21]. Extensive work has been done regarding integration
of distributed generation with the grid. Work has also been carried out for economic
procurement of power from distributed wind generators.

2.1 Over Voltage

The occurrence of high voltage for a very small time period in a power system is
called the over voltage or surge voltage or transient voltage. By the occurrence of over
voltage, the stress of voltage can harm the lines and components which are connected
with the power system. High frequency due to load switching and lightning, medium
frequency caused due to capacitor energizing, or low frequency can be generated by
the transient over voltage [16].
The other causes of over voltage are direct lightning stroke. These could be due
to the presence of charged clouds nearby, thus causing the electro-statically induced
voltage. There could be changes in atmosphere along the length of the line, which
induce the over voltage. There could be a lightning discharge taking place near the
line: called side stroke. The over voltage tends to stress the insulation of electrical
equipment and it is expected that these harms them when it regularly occurs. Over
voltage caused by surges can result in flash over and spark over between phase and
ground at the weakest point of the network, breakdown of gaseous or liquid or solid
insulation, or connection breakdown.

2.2 Under Voltage

The dip of voltage is a very ordinary and grave disturbance of power quality. Due
to dip voltage, sensitive equipment and industrial processes are affected [1]. Voltage
dip could arise when there is a bulky load such as energizing of the transformer, ener-
gizing of capacitor, start-up of motor, electric load switching, temporary overload,
or a fault in the connected system. The wind generator disconnection could have
a harmful impact on the system stability due to generation loss can causes of the
under voltage [22]. For at least one minute in the drop of apply voltage to approxi-
mately 90% of its rated voltage, or a lesser amount that condition is called the under
voltage. When a capacity of voltage asks for added power than the line can deliver,
158 R. Kaur et al.

low-voltage situation happen. The capacity addition of fresh apparatus and improved
the load beyond what the line could power send.
The other causes of under voltage are when the power is shorted to ground, a
bulky load is connected without notifying the power company, or there are loose
cable connections. The effects of under voltage are that the resistive loads will run
less effectively, and loads with controllers to do a job—i.e., thermostats to maintain
A/C, refrigerators, ovens, etc., will run longer and harder and end up using the same
power until the voltage is way too low and the units is running.

2.3 Over and Under Speed of Wind Turbine Rotor

The reactive power flow from the transmission line is drawn by the DFIG; this can
raise the transmission system liability in the incident of a collapse. The wound rotor
is requiring in DFIG configuration. A variable speed wind turbine with a permanent
magnet synchronous generator and alternating voltage is produced by the induction
generator [22]. The rotor speed of wind turbine is generally affected by the speed of
wind. The AC voltage generated frequency by the wind turbine is a purpose of the
rotor speed contained by the generator:

120 f
N= (1)
p

where N is speed of the rotor in rpm, p is the number of poles in the generator, and
f is the output voltage frequency in hertz. In the power system, if the speed of rotor
varies, speed of wind varies, and the voltage frequency varies than the electricity
generated power line cannot be directly connected with the transmission system.
Power converters are used, which results in the de-coupling of the wind turbine from
the transmission system. As extra wind turbines are integrated in a country power
system, the inertia is reduced. This means that the transmission system frequency is
powerfully affected by the failure of a particular generating unit.

3 Faults Occurring in Grid Connected System

To cater the requirements of increasing demand of power and to preserve generation

and load equilibrium, electric utilities are opting for interconnections of existing
networks and integration of renewable energy into the system. Though these devel-
opments may fulfil the power demand, but at the same time, they may affect the
secure and reliable operation of entire system [23]. Worldwide the generation by
wind power has expanded and is steadily increasing the contribution towards the
power demand of a country. The prime importance is the re-establishing of stable
grid conditions after a transient fault. This needs the recovering of the grid voltage
Grid Integrated with Wind Turbine System … 159

and damping power system oscillations caused by the fault. Wind turbines had been
considered as embedded generator, which did not contribute to the stability of the
power system. In the system, the wind turbines were essential not to aggressively
effort to control the power and the voltage. When the abnormal condition occurred,
the wind turbine needs to be cut-off from the grid line due to the operation of protec-
tion system. Faults are inherent unavoidable issues in the integrated power system.
Due to many reasons, faults occur in power system.
A fault is any abnormal electric current. The faults affect the performance for
many equipment or appliances which are connected with line. When the wind mills
are interconnected with the grid there is voltage sag or swell in the line. Other main
problem in wind power is when the speed of wind suddenly increases or decreases
due to atmospheric impact, it is important to disconnect the wind turbine from the
grid.

4 Modelling

The wind turbine and grid line supply the power to load are modelled using
MATLAB/SIMULINK. A 500 MW load is connected with a 575 V line that is
nearest to the distributed generator of wind turbine. Six units, each of 1.5 MW wind
turbine combine to form the 9 MW wind farm which is interfaced to a 25-kV distri-
bution power system to supply power to 120 kV grid by using a 30 km feeder.
The feeder voltage is 25 kV. Wind turbines are coupled to a doubly fed induction
generator (DFIG). The wind energy system has an induction generator (i.e., wound
rotor induction generator) and PWM converter based on AC to DC and DC to AC
insulated gate bipolar transistor (IGBT). The stator winding is directly connected
to the 50 Hz grid, whereas the rotor is fed at variable frequency via AC to DC and
DC to AC converter. The DFIG technology permits to detect maximum energy from
the wind for low speeds of wind through optimizing the turbine speed and reducing
mechanical stresses from the turbine while gusts of wind.

4.1 Integration of Two Wind Power Plants

There are two wind mill are connected with single grid line. The two DFIG generators
are connected with the grid, these two turbines increase the total power. Figure 1
shows the arrangement of two wind turbine connected with grid. These Wind Turbine-
1 and Turbine-2 are connected in parallel to each other. They increase the power or
matching the load demand which is increasing day by day. The two wind turbines
are of same ratio and are connected with line.
There are two cases consider:
Case-1 The wind speed of Turbine-1 and Turbine-2 are same at a time.
160 R. Kaur et al.

Fig. 1 Two wind turbine integration

Case-2 The wind speeds of Turbine-1 and Turbine-2 are different at a time.

5 Discussion and Results

The three type of faults that commonly occur in power system have been analysed;
these are over voltage fault, under voltage fault, and over speed of wind fault. When
the over voltage or under voltage fault or high-speed wind occurs, there must be
a protection system which turn off the wind turbine connected with grid to protect
all the connected equipment. Therefore, the protection system has been installed
between the grid and wind farm line so as to cut-off the turbine section during a
fault.

5.1 Case 1: Over Voltage Fault

Suppose due to some reason at the grid side the amplitude of voltage is increase +
0.30 V per unit, the tripping system gets activate and disconnect the wind farm with
grid. The simulation scope as shown in Fig. 2 is for the over voltage fault for duration
0–30 s time period. From 1 to 5 s, the system operates smoothly and after 5 s the
fault occurs and the tripping system is activated; the fault status of over voltage and
under voltage status is 1. That shows which type of faults are occurrd in system.
Grid Integrated with Wind Turbine System … 161

Fig. 2 Over voltage fault simulation

5.2 Case 2: Under Voltage

As seen in Fig. 3, the amplitude of voltage has decreased due to fault and the protection
systems protect the wind turbine by tripping the wind energy conversion system from
the grid. The amplitude of voltage decreased to −30 at the grid side before the fault
occur the system is operating properly but after 5 s the under voltage faults occurs
and the protection disconnected the wind energy conversion system from the grid
and the status of under voltage observed is 1.

5.3 Case 3: Over Wind Speed

The effects on the wind turbine by the increase of the speed wind is shown in Fig. 4.
162 R. Kaur et al.

Fig. 3 Under voltage fault simulation

The wind speed is 30 m/s the voltage level is approximately 0.99 p.u. After 5 s, the
protection system activates for operation. When the power reaches at 0.9 p.u., it is
constant from 1 to 22 s. At 22 s, the power drops from 0.9 to 0 due to the operation
of protective system. The speed of generator is gradually increased. The maximum
value is 2.3 p.u. its shows that when the speed of induction generator has increased to
2 p.u. the protection system protects the system from the hyper-synchronous speed
of wind turbine. The pitch angle is gradually increased after 13 s on the simulation
of two turbines with different wind speed. The wind speed of Turbine-1 is 8 m/s and
Turbine 2 is 30 m/s which is over speed condition for Turbine-2. At the rated speed
of wind, i.e., 8 m/s the active power of grid is 1.87 p.u. and at the 30 m/s speed of
wind. The tripping circuit is activated and helps to cut out the WECS from grid as
seen in Fig. 4. The rated cut out speed of the turbine is 25 m/s.
Its shows that when the speed of wind is increased beyond the optimum level,
over speed fault occurs and the Turbine-2 trips and is disconnected from the grid
which as shown in Fig. 4. Table 1 gives the turbine parameters when Turbine-1 is
operating at normal speed of wind and Turbine-2 at over speed.
Grid Integrated with Wind Turbine System … 163

Fig. 4 High wind speed fault simulation

Table 1 Turbine parameters

Turbines Turbine 1 Turbine 2 (over speed)
when Turbine-2 operates at
over speed Wind speed (m/s) 8 30
Power (MW) 1.87 0 (trip)
Reactive power (MVAR) 6806e−08 0 (trip)
Speed of generator (p.u.) 0.8 1.35
Voltage (p.u.) 1.007 1.007

6 Conclusion

The electrical faults that commonly occur in power systems and the proper operation
of protection system is extremely essential. The challenges related to the fundamental
protection of wind turbines systems that are integrated with the grid are discussed
and their respective solutions evaluated. The problem of unnecessary separation of
all wind turbines is not acceptable, only the disconnection of faulty unit is accept-
able; as it effects the continuity and stability of the complete system. The study
accentuated the need for the protection schemes for wind farms to appreciate power
system performance as well as decrease the damages resultant from the abnormal
events occurrence. When the two wind turbine systems are connected with grid if
164 R. Kaur et al.

the faults occur in one turbine system then the faulty turbine system should only be
disconnected from the grid rather than both.

References

1. Shafiullah GM, Shawkat Ali A, Wolfs P (2013) Potential challenges of integrating large-scale
wind energy into the power grid—a review. Renew Sustain Energy Rev 20:306–321
2. Ackermann T (2012) Wind power in power systems, 2nd edn. Wiley-Blackwell, Hoboken
3. Davitian H (1978) Wind power and electric utilities: a review of the problems and prospects.
Wind Eng 2:1–28
4. Marko S, Darul’a I (2007) Large scale integration of renewable electricity production into the
grids. J Electr Eng 58:58–60
5. Alegrıaa IM, Andreua J, Martına JL, Ibanezb P, Villateb JL, Camblongc H (2007) Connection
requirements for wind farms: a survey on technical requirements and regulation. Renew Sustain
Energy Rev 11:1858–1872
6. Smith JC, Milligan MR, DeMeo EA, Parsons B (2007) Utility wind integration and operating
impact state of the art. IEEE Trans Power Syst 22:900–908
7. Dragomir D (2011) The aspects related to the grid connection of the wind farms into national
power system. UPB Sci Bull 73:273–279
8. Khadem SK, Basu M, Conlon MF (2010) Power quality in grid connected renewable energy
systems: role of custom power devices. In: International conference on renewable energies and
power quality, vol 3. Spain, pp 23–25
9. Sharad W, Mohan M, Aware V (2011) Power quality issues and its improvement in wind energy
generation interface to grid system. MIT Int J Electr Instrum 1:116–122
10. Ghosh S, Senroy N (2015) Electromechanical dynamics of controlled variable-speed wind
turbines. IEEE Syst J 9:639–646
11. Mei F, Pal B (2007) modal analysis of grid-connected doubly fed induction generators. IEEE
Trans Energy Convers 22:728–736
12. Mishra Y, Mishra S, Li F, Dong ZY, Bansal RC (2009) Small-signal stability analysis of
a DFIG-based wind power system under different modes of operation. IEEE Trans Energy
Convers 24:972–982
13. Wu F, Zhang X-P, Godfrey K, Ju P (2007) Small signal stability analysis and optimal control
of a wind turbine with doubly fed induction generator. IET Gener Trans Distrib 5:751–760
14. Mehta B, Bhatt P, Pandya V (2015) Small signal stability enhancement of DFIG based wind
power system using optimized controllers parameters. Int J Elect Power Energy Syst 70:70–82
15. Fadaeinedjad R, Moallem M, Moschopoulos G (2008) Simulation of a wind turbine with doubly
fed induction generator by FAST and simulink. IEEE Trans Energy Convers 23:690–700
16. Ganesh P, Senroy PN, Kar IN (2017) Wind turbine structural modeling consideration for
dynamic studies of DFIG based system. IEEE Trans Sustain Energy. 8:1463–1472
17. Lackner MA, Rotea M (2011) Structural control of floating wind turbines. Mechatronics
21:704–719
18. Anaya-Lara O, Jenkins N, Ekanayake J, Kartwrite P, Hughes M (2009) Wind energy generation:
modelling and control. Wiley, New York
19. Bauschke S, Obkircher C, Achleitner G, Fickert L, Sakulin M (2006) Improved protection
system for electrical components in wind energy plants. In: International conference on power
system protection. Bled-Slovenia
20. Ramakumar R, Butler N, Rodriguez A, Venkata S (1993) Economic aspects of advanced energy
technologies. Proc IEEE 81:318–332
21. Hornak D, Chau N (2004) Green power—wind generated protection and control considerations.
Protective relay engineers. In: 57th annual conference, pp 110–131
Performance Improvement of Wind
Turbine Induction Generator Using
Neural Network Controller

Sanjay Dewangan and Shelly Vadhera

Abstract Self-excited induction generators (SEIGs) are best suited for power gener-
ation with wind turbine system due to their advantages like robustness, high efficiency
and low maintenance. In this paper, the performance of standalone self-excited induc-
tion generator using indirect vector control (IVC) is investigated. The generator-side
proportional-integral (PI) controller is replaced with neural network controller (NNC)
to minimize the DC voltage transient and improve the performance of the proposed
system. The MATLAB/Simulink model of the proposed isolated SEIG is developed
and its performance is compared at variable speed conditions for both PI controller
and NNC.

Keywords Induction generator · PI controller · NNC · WTIG system

1 Introduction

The ever rising requirement of power and depletion of fossil fuels reserves as well as
the global exposure on environmental concerns leads to renewable energy sources,
for example, solar, wind, geothermal energy and hydroelectric power. Wind power
has attracted much attention among all non-conventional resources and is currently
one of the top-growing technologies. Different types of electric generators are being
used such as permanent magnet synchronous generator, doubly fed induction gener-
ator and squirrel-cage induction generator (SCIG). Among these, SCIG is almost
mostly employed. Self-excited induction generators (SEIGs) are chosen in standalone
systems over alternator due its benefits like low cost, comfort of protection [1]. Wind
turbine induction generator with static synchronous compensator (STATCOM) was
investigated in [2], which produces constant frequency and constant stator voltage.

S. Dewangan (B) · S. Vadhera

Department of Electrical Engineering, National Institute of Technology Kurukshetra,
Kurukshetra, Haryana 136119, India
e-mail: [email protected]
S. Vadhera
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 165

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_16
166 S. Dewangan and S. Vadhera

The power converters in wind turbine induction generator (WTIG) system are
controlled with the aim of extracting the highest power from kinetic energy of wind.
By maintaining the speed of the generator, the first job is done by the generator-side
converter, whereas second one is completed successfully by the grid-side inverter by
controlling the DC voltage, and ensuring small harmonic deformation in compliance
with the grid codes. Vector control technique is employed to regulate the active and
reactive power independently. Implementation of direct torque control for doubly fed
induction generator-based wind energy system was discussed in [3]. Boost converter
is used for advance controlling of SEIG at variable speed and isolated wind energy
system is used for pumping system [4, 5]. In [6], the control of the wind-driven
SEIG using fuzzy logic controller at variable speed is proposed, which uses indirect
vector control technique. A decouple control is designed to decrease flux ripples and
to develop the dynamic performance. In [7], the voltage improvement process of an
induction generator using stator flux-oriented vector control was tested. An improved
solution to control voltage of the asynchronous generator was described using rotor
field-oriented control in [8].
To overcome the variation of the SEIG voltage during step change of wind speed,
an indirect control (IVC) technique using back to back converters for isolated SEIG
had been implemented and it is also used to minimize the torque ripple. NNC is
implemented in IVC in place of PI controller to improve the performance of system.

2 Wind Turbine Modeling

1
Pmax = Cp (λ)ρπ R 2 Vwind
3
(1)
2
The coefficient of power = Cp (λ, β), presents the aerodynamic efficiency of the wind
turbine, β = blades pitch angle, so the ratio of tip speed = λ which can be presented
as follows λ = ωVt R ; Cp is defined as:
  C
C2 5
Cp (λ, β) = C1 − C3 β − C4 e λi + C6 λ (2)
λi

With:

C1 = 0.5176, C2 = 116, C3 = 0.4, C4 = 5, C5 = 21, C6 = 0.0068 (3)

1 1 0.035
= − 3
λi λ + 0.008β β +1

Betz limit is defined by Cp = 59.3%.

Performance Improvement of Wind Turbine Induction … 167

3 SEIG Modeling Using IVC

Indirect vector control technique is employed to regulate the torque and flux inde-
pendently. The dynamic equations of SEIG model are described by the following
equations

dψqs
Vqs = Rs i qs + + ωe ψds (4)
dt
dψds
Vds = Rs i ds + − ωe ψqs (5)
dt
dψqr
0 = Rr i qr + + (ωe − ωr )ψdr (6)
dt
dψdr
0 = Rr i dr + − (ωe − ωr )ψqr (7)
dt
The stator and rotor flux linkages are given as:

ψqs = L s i qs + L m i qr , ψqr = L r i qr + L m i qs , (8)

ψds = L s i ds + L m i dr , ψdr = L r i dr + L m i ds , (9)

From the above equations, the machine voltage and current equations in matrix
form are written as:
⎡ ⎤ ⎡ ⎤⎡ ⎤
Vds Rs + s L s −ωe L s s Lm −ωe L m i ds
⎢ Vqs ⎥ ⎢ ωe L s Rs + s L s ωe L m s Lm ⎥ ⎢ ⎥
⎢ ⎥ ⎢ ⎥⎢ i qs ⎥
⎣ 0 ⎦=⎣ s Lm −(ωe − ωr )L m Rr + s L r −(ωe − ωr )L m ⎦ ⎣ i dr ⎦
0 (ωe − ωr )L m s Lm (ωe − ωr )L r Rr + s L r i qr
(10)

Torque equation

3
p  Lm
Te = i qs ψdr (11)
2 2 Lr

where Vsd and Vqs are stator voltage, i sd and i sq are stator currents, ir d and irq rotor
currents, Rs and Rr are rotor resistances; L s and L r are stator and rotor inductances,
respectively. L m is magnetizing inductance, rotor speed is ωr , Laplace operator is
‘s ’, Vr d and Vrq are rotor voltage.
168 S. Dewangan and S. Vadhera

Fig. 1 Neural network architecture

4 Neural Network Controller

Input signals are speed error and change of speed error for neural network
controller. Reference torque is obtained by neural network controller. Figure 1 shows
neural network controller which has been implemented in MATLAB/Simulink for
wind energy conversion system. Simulation data is used to train the NNC using
backpropagation algorithm.

5 Results and Discussion

5.1 Proposed System

Figure 2 describes the basic arrangement of the isolated wind turbine induction gener-
ator system. Indirect vector control technique is implemented to regulate the varia-
tion of stator voltage, torque and flux independently. The proposed system consists
of 4 kW SEIG, wind turbine and two levels power converters. The back to back
converters are employed to regulate the variation real and reactive power. The fixed
capacitor bank is employed to provide magnetizing power to the induction machine
to build up voltage.
Performance Improvement of Wind Turbine Induction … 169

Fig. 2 SEIG-based wind energy system

5.2 Simulation Results

The wind speed has been changed from 9 to 13 m/s, at the time 2 s. The DC
voltage, stator voltage, speed responses and torque are described during variable
speed. Figures 3 and 4 describe the variation of stator voltage and DC link voltage
at variable speed using PI controller and NNC-based wind turbine system. NNC-
based standalone SEIG provides the good response, fewer transient compared to
PI controller. At maximum speed, stator peak voltage of 95% is obtained using PI

Fig. 3 SEIG voltage (V)

170 S. Dewangan and S. Vadhera

Fig. 4 DC link voltage (V)

controller, but NNC provides the stator voltage of 97% which is near to rated voltage.
Reference DC voltage is 400 V. PI controller provides DC link voltage overshoot of
6.5%, but NNC provides DC link voltage overshoot of 3% at the time t = 2 s. Hence,
NNC is superior to PI controller.
Figures 5 and 6 describe the variations of SEIG speed and electromagnetic
torque at variable speed using PI controller and NNC-based wind turbine system.
PI controller gives speed overshoot of 2.44%, but NNC provides speed overshoot
of 0.48%. PI controller provides 5% torque ripples with settling time of 0.75 s, but
NNC provides torque ripples of 3% with 0.65 s during starting. At step change of

Fig. 5 SEIG speed (pu)

Performance Improvement of Wind Turbine Induction … 171

Fig.6 Torque (pu)

Table 1 Comparisons for

Operating system Parameters PI controller/NNC
torque response using PI
controller and NNC During starting Torque ripple (%) 5/3
Settling time (s) 0.75/0.65
Step change in speed Torque ripple (%) 5/4
Settling time (s) 0.76/0.61

speed, PI controller provides 5% torque ripples with more settling time of 0.76 s, but
NNC provides torque ripples of 4% with settling time of 0.61 s. Table 1 provides the
comparisons data for torque.

6 Conclusion

A neural network controller-based indirect vector control method for a wind turbine
induction generator is discussed here to adjust the SEIG speed, torque and DC voltage
at step change of wind speed. The performance of the proposed system is improved
by replacing torque PI controller with NNC with lower torque ripples, high dynamic
response and good transient response of DC link voltage. Hence, NNC is better than
PI controller.

Acknowledgements The current work is supported by the Ministry of Human Resource Develop-
ment, Government of India through a Ph.D. scholarship grant.
172 S. Dewangan and S. Vadhera

Appendix

Induction machine parameters

P = 4000 W, V = 400 V, f = 50 Hz, Rs = 0.035 pu, N = 1430 rpm,

Lls = 0.045 pu.

Converter parameters

Ron = 0.001 , L on = 1 mH, V f = 0.8 V.

IGBT inverter parameters

Ron = 1 m, Rs = 105 .

References

1. Dewangan S, Kumar N, Dyanamina G (2018) Performance improvement with fuzzy logic

controller for generator control in wind energy system. In: 2018 IEEE 8th power India
international conference. Kurukshetra
2. Bharti S, Dewangan S, Ram S (2016) Performance of wind farm distribution system under
balanced and unbalanced condition with SVC and STATCOM. In: International conference on
electrical power and energy system, pp 182–187
3. Jaladi KK, Sandhu KS (2019) Real-time simulator based hybrid control of DFIG-WES. ISA
Trans J 93:325–340
4. Ahmed T, Nakaoka M, Tanaka T (2006) Advanced control of a boost AC–DC PWM rectifier
for variable-speed induction generator. In: Twenty-first annual IEEE applied power electronics
conference and exposition, pp 956–962
5. Miranda MS, Lyra RC, Eeufmg C, Silva SR (1999) An alternative isolated wind electric pumping
system using induction machines. IEEE Trans Energy Convers 14(4):1611–1616
6. Dewangan S, Dyanamina G, Kumar N (2019) Performance improvement of wind driven self
excited induction generator using fuzzy logic controller. Int Trans Electr Energ Syst 29:1–20
7. Seyoum D, Rahman MF, Grantham C (2003) Terminal voltage control of a wind turbine driven
isolated induction generator using stator oriented field control. In: Eighteenth annual IEEE
applied power electronics conferences and exposition, pp 846–852
8. Leidhold R, Garcia G, Valla MI (2002) Field-oriented controlled induction generator with loss
minimization. IEEE Trans Ind Electron 49(1):147–156
Solar Resource Assessment and Potential
in Indian Context

Vinod Nandal, Raj Kumar, and S. K. Singh

Abstract The demand for electricity is increasing day by day causing environ-
mental problems due to the widespread burning of fossil fuels. Therefore, research
has been promoted in solar energy to mitigate environmental pollution. The land of
India receives a good amount of solar energy. This paper describes the assessment of
solar radiation resources in India. All solar resource stations have advance measuring
instruments of the same type to collect, store and send radiation and meteorological
parameters at central station National Institute of Wind Energy (NIWE) Chennai.
One-year solar data from January to December 2015 of DCRUST, Murthal center
is assessed, which assist solar project developers for feasibility analysis and devel-
opment of new solar projects in this area. Moreover, state-wise solar potential and
installed capacity are also described. This study is significant for design, modeling
and performance analysis of solar plants. In addition to this solar resources give
impetus for the deployment of solar energy along with futuristic solar technologies
in India.

Keywords Solar resource · Solar energy · Radiation · Solar potential

1 Introduction

The role of energy is critical for industrial and socio-economic development of any
country [1]. India’s energy need, along with growing population is mainly lead by

V. Nandal (B) · R. Kumar

Department of Mechanical Engineering, Deenbandhu Chhotu Ram University of Science and
Technology, Murthal, Haryana, India
e-mail: [email protected]
R. Kumar
e-mail: [email protected]
S. K. Singh
Department of Physics, Deenbandhu Chhotu Ram University of Science and Technology,
Murthal, Haryana, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 173

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_17
174 V. Nandal et al.

the conventional fuel-based sources which are depleting day by day [2]. Being an
emergent nation, India’s entire installed capacity of power from diverse resources
is 1.362 GW in 1947 to 344.002 GW at the end of the financial year, 2018 [3].
India’s power-producing capacity is mainly commanded by coal since independence
[4]. The conventional sources affect the environment as well as the health of human
beings. These problems will make unsustainable situations and create irreversible
threat toward human societies [5]. India’s strategy toward sustainable renewable
energy is an important pillar to fulfill the ever-rising demand for electricity while
addressing energy security, climate change, greenhouse gas emissions and water
shortage objectives. Moreover, renewable energy emerged as a sustainable alternative
source of energy and offer an accessible solution to the growing challenges [6]. In
addition to this deployment of reliable, affordable and modern renewable energy
technologies offer an excellent opportunity as a green energy solutions for mankind.
In order to meet up electric demand of India, sustainable renewable sources
along with modern energy technologies will provide a prominent solution for the
overall development of India. During the Paris agreement, the Government of India
committed to raising the percentage share of sustainable renewable energy from 30%
in 2015 to 40% by 2030 [7]. This incorporates an ambitious target of implementing
175 GW from renewable sources by 2022, and from this 100 GW is only planned
through solar energy [8]. To fulfill the increasing energy demand of India in a sustain-
able way, the emphasis is being given for the development and dissemination of solar
energy. Moreover, in India, solar energy plays a critical role in future energy mix [8].
Such an initiative requires the identification of suitable areas for development and
dissemination of solar electric power. A detailed potential evaluation for the whole
nation and identification of niche area shall be of great help for the researchers,
policymakers and solar developers.

1.1 Current Power Scenario in India

The power sector is the growth engine for economic, industrial and societal devel-
opment of any country. Access of affordable, quality and sustainable power is the
basic need of Indian communities. It is evaluated that the consumption of energy in
India may increase by 132% where the electricity demand may increase three times
by 2035 [9]. Figure 1 illustrates the source-wise all India installed capacity of power
as on March 31, 2019.
Being a developing nation, India’s overall installed capacity in terms of power
from various sources is 1.362 GW in 1947 to 356.1 GW as on March 31, 2019 [3,
10]. The share of thermal power is 63.54%, the contribution of renewable energy is
21.80%, and hydropower holds 12.75% while nuclear power share is 1.90%. More-
over, a major source of power supply in India is from coal-based thermal power
with 56.36% share [10]. Therefore, coal acts as a prime source of power since
independence and also the main source of environmental pollution.
Solar Resource Assessment and Potential in Indian Context 175

Renewable, 21.80

Hydro, 12.75

Coal , 56.36
Nuclear, 1.90

Diesel, 0.18

Natural Gas, 7.00

Fig. 1 Source-wise all India installed capacity of power in percentage (Source [10], p. 1)

Now, it is time to look ahead to augment the share of renewable power in the Indian
power sector. The blend of renewable energy will meet the ever-growing demand for
power without breaching environmental laws [11]. The solar energy plays a critical
role to achieve the 100 GW ambitious targets of solar power in India by 2022 [8].
And this is the driving force toward to assess the solar radiation resources and solar
potential. Therefore, solar energy has the immense potential to lead the Indian power
sector in a sustainable manner. India’s total installed capacity of sun-based power is
28.18 GW as on March 2019. Figure 2 described the growth story of solar power in
India [3, 10].
The growth of solar power started at 2.12 GW in the year 2008–2009 and observed
the marvelous growth with 28.18 GW by the year 2018–2019 [3, 7]. Moreover, for
escalating solar energy utilization, the Indian government has implemented various
policies and schemes in a different time frame.

Solar Power Installed Capacity, 2008-09 to 2018-19

30000
Installed capacity (MW)

28180.71
25000
21651.48
20000

15000
12288.83
10000 6762.85
3743.97
5000 2631.93
1686.44
2.12 6 32.39 941.31
0

Year

Fig. 2 Growth of solar power in India (Source [10], p. 1, [3], p. 26])

176 V. Nandal et al.

1.2 Harnessing Solar Energy in India

Solar energy is abundant in nature and largely untapped resource which has the
potential to resolve the ever-growing demand of power in a sustainable manner.
Solar radiation emits from the sun at the rate of 3.8 × 1023 kW and out of this nearly
1.8 × 1014 kW is received on the earth [12]. Solar radiation can be harnessed in
the form of heat and light. The dominant part of the radiation is lost in the form of
reflection, absorption and scattering by clouds. Though, solar energy is sufficient to
fulfill the entire electricity demand of the world [13].
India has tremendous scope of harnessing solar energy to meet the overall demand
of the electricity. India’s demographic position as a tropical nation having landmass
of 2.9 million km2 with excellent solar irradiance has huge potential to harness solar
energy. India is blessed with 250–300 sunshine days in a year, annual solar irradiance
of 4–7 kWh/m2 /day to generate 5000 trillion kWh of energy [14, 15].
The potential use of solar energy in India mainly depends upon the assessment
of solar radiation resources and availability of land. Therefore, to harness the solar
energy it is necessary to assess the solar radiation resources of the country which have
a significant impact toward the design of the solar plant, solar resource forecasting
and solar technology selection. Moreover, based on the availability of land in the
country, it also helps to assess the solar energy potential in the country. So to explore
the potential use of solar energy, it is necessary to study the state of solar energy in
Indian context.

2 State of the Solar Energy

The segment of solar power in India is increasing at a rapid rate with the goal of
100 GW solar power mission till 2022 [16]. Production of solar power is directly
influenced by the amount of incident radiation at a specific site. The yield of this
environmentally friendly power relies on the radiation quality, sunshine hour, ambient
temperature and use of solar technology [17, 18]. Therefore, precise, high quality,
reliable and accessibility of solar radiation data and meteorological data is imperative
for solar energy development [19].
Estimation of ground-based solar data needs precise measurements. Pyranometer
a solar radiation measuring apparatus is utilized for direct horizontal irradiance (DHI)
and global horizontal irradiance (GHI) while pyrheliometer is utilized for direct
normal irradiance (DNI). The quality of solar data can be kept up at a specific
level by removing uncertainty, operational mistakes and instrumental errors [20].
There are various forms of solar radiation data are available but for site-specific
analysis, satellite-based solar data is used. The success story of any site-specific solar
project requires at least one-year ground-based solar data [21]. In addition to this,
the recognition of any solar project is profoundly dependent upon the measurement
of solar resource quality and its appraisal [22].
Solar Resource Assessment and Potential in Indian Context 177

Solar data of any region plays a critical role for assessing the solar potential and
for selection of solar applications in that zone [23–25]. The knowledge of solar radi-
ation evaluation is essential for solar system designing, modeling, simulation and
analysis as well as for solar appliances and development of futuristic solar technolo-
gies [26–28]. Evaluation of global solar radiation assumes an essential job for solar
technologies and applications selection along with feasibility evaluation of solar
project [29, 30]. Karakoti et al. [31] analyzed the diffused solar radiation data of 23
solar stations in India and suggested the empirical models which demonstrate the rela-
tionships with daylight hours, temperature and relative humidity. Kandasamy et al.
[32] assessed the solar energy potential using PVSYST software in the Tamil Nadu
state of India. Abubakar and Muthuchamy [33] assessed the solar power potential in
three states, i.e., Tamil Nadu, Kerala and Karnataka using photovoltaic geographical
information system (PVGIS). Moreno-Tejera et al. [34] explored the parameters and
solar radiation data which influence the evaluation of solar power of the specific local
site. Zafarani et al. [35] evaluated the utility of meteorological data for estimating
solar power.
Few studies have been depicted about the potential estimation and solar radi-
ation appraisal in the Indian context. Ramachandra et al. [36] assessed the solar
power potential of Himachal Pradesh utilizing the satellite-driven GHI data collec-
tions. Schwandt et al. [37] examined the 51 Indian solar centers data of solar radia-
tion. There are some gaps observed in solar radiation data because of instrumental
errors, misalignment, failures of power supply, stations operational errors, main-
tenance problems, etc. Additionally, the technique of gap filling is depicted and
further checked the quality of solar radiation data. Harinarayana and Kashyap [38]
described the factors which affect the generation of sun-based energy in India and
the total yearly energy production varies from 510,000 to 800,000 kW/acre of land.
Gupta et al. [39] depicted the solar and wind asset appraisal process for India. Sharma
et al. [40] analyzed the solar and climatologically parameters which have a signifi-
cant effect to evaluate the solar energy potential. Moreover, authors also explored the
solar potential based upon the availability of wasteland of the country. Mahima et al.
[41] studied the solar radiation pattern for a specific area which is imperative for the
determination of solar applications based on their performance. Raghuwanshi and
Arya [42] investigated the solar energy potential in the Indian context and promoted
clean and green energy technologies to fight against environmental pollution.
It is very important to assess solar radiation precisely for the growth of clean
and green energy in India. This requires quality and reliable solar database. Also,
India still needs more solar resource centers for the solar database so as to fulfill the
accurate assessment of solar-based resources. This article summarizes solar centers
network design, usage and quality affirmation of solar data. This investigation also
incorporates the examination of one-year solar radiation parameters information of a
specific station. In addition to this, present investigation contributes significant effect
toward an assessment of solar potential, technological selection, forecasting of solar
resources and solar-based modeling of power plants in India as well as globally.
178 V. Nandal et al.

3 India’s Solar Radiation Resources

Deployment of solar energy in India and for designing solar power plant needs
quality and reliable solar radiation data. Precise, the evaluation of ground-based
solar data and other meteorological parameters are the basic essential elements for
solar radiation resources assessment and potential estimations, which can accomplish
the following objectives:
• Provides database for solar project developers, solar manufactures, solar
researchers and policymakers.
• Support the development of analysis tools, solar technologies and predict solar
resources.
• Facilitate meteorological research in the country.
• Support in potential estimations in India as well as at a particular area.
Solar radiation is the imperative parameter to appraise the potential and power
yield of the solar plants; however, it does not reach on earth in a full amount. Some
portion of the radiation is absorbed, some reflected and some dispersed because of
complex interactions of the Earth’s climate with the occurrence of solar radiation.
So there is a need to think about the basic of radiation, which is used in solar-
powered conversion technologies [22]. Generation of electricity from solar tech-
nologies requires solar irradiance as input. The basic knowledge of solar irradiance
as well as measuring instruments is essential to evaluate the solar data and further
to evaluate the solar potential. There is a huge potential for solar energy in India. So
the assessment of solar resource is very important to harvest environment-friendly
solar power.

3.1 Network of Solar Radiation Resource

National Institute of Wind Energy (NIWE) under the aegis of Ministry of New and
Renewable Energy (MNRE), Indian government started a pilot project in the year
2011 in phase manner of erection and commissioning of SRRA stations all over the
India in technically collaboration from the Deutsche Gesellschaft für Internationale
Zusammenarbeit (GIZ GmbH). In the first stage, 51 SRRA stations were established
by NIWE in 11 states and 1 in Union Territory. While in the second stage, 60 SRRA
stations in 28 states and 3 Union Territories alongside 4 Advanced Measurements
Stations (AMS) were also set up by June 2014. NIWE has additionally commis-
sioned 6 SRRA stations for Maharashtra Energy Development Agency (MEDA) in
Maharashtra, under consultancy mode by May 2015 [43, 44]. So at present in India,
a total of 121 SRRA stations was established by NIWE.
The network of Solar Radiation Resource Assessment (SRRA) stations in India is
one of the most advanced setups for assessment and measurement of solar irradiance,
processing and handling of data, quality assurance, modeling and development of
Solar Resource Assessment and Potential in Indian Context 179

Fig. 3 Solar resource data flow (own elaboration)

solar atlas of the country. Moreover, data collection and assessment of meteorological
parameters are also done by SRRA stations for the dissemination of solar power in
the country [39, 43]. Figure 3 illustrates the solar resource data flow in India.

3.2 Instruments in Indian Solar Resource Stations

Radiation is an important factor for design, performance and site selection of solar
plant. There are various ways and technologies for the measurement of solar irra-
diance. During measurements, there are other meteorological parameters such as
ambient temperature, atmospheric pressure, wind speed, wind direction, relative
humidity and precipitation which are very important to measure because these param-
eters affect solar irradiance. Figure 4 illustrates the instruments installed at one of
the SRRA stations in India.
In India, all SRRA stations are identical in design and have the same model and
quantity of instruments. Table 1 illustrates the instruments and sensors mounted in
SRRA stations with their model and manufacturer details [46].
To decrease the fossil fuel-based power percentage, the solar-generated power will
play a critical role in the Indian power sector. So the estimation of solar irradiance is
180 V. Nandal et al.

Fig. 4 One of SRRA station instruments in India (Source [45], p.3)

Table 1 Details of instruments used in SRRA station (Source [46], p. 3)

Instrument types Parameters measured Manufacturer model
First pyranometer GHI (W/m2 ) Eppley Lab USA-PSP
Second pyranometer DHI (W/m2 ) Eppley Lab USA-PSP
Pyrheliometer DNI (W/m2 ) Eppley Lab USA-NIP
Solar tracker Mounted with shaded Geonica Spain- SMT-3
Temp. and relative humidity pyranometer and pyrheliometer RM Young, USA-41382VC
sensor Ambient Temp. (ºC), RM Young, USA-85000
Ultrasonic wind sensor Humidity (%) RM Young, USA-52203
Rain gauge Wind speed (m/s), wind direction RM Young, USA-61302L
barometer (o ) Geonica, Spain
Data logger and modem Rain accumulation (mm) Garmin, USA
GPS Atmospheric pressure (hPa) Garmin, USA
GPRS Gathering data from instruments Moserbaer, India
Solar PV panel and send it to central station at Exide, India
External battery CWET, Chennai
Synchronize sun tracker with sun
movement
Transfer data through mobile
SIM cards to central stations
CWET, Chennai
Supply power for charging
station batteries
Electrical storage
Solar Resource Assessment and Potential in Indian Context 181

Table 2 Average monthly values of solar irradiance and meteorological parameters of the year
2015 ( Source SRRA Murthal Station-2389)
Month GHI DNI DHI AT ( RH WS WD ( PR
(kWh/m2 /d) (kWh/m2 /d) (kWh/m2 /d) ºC) (%) (m/s) º) (hpa)
January 2.4 1.4 1.7 11.4 88 2.3 299 989
February 3.9 2.9 2.0 17.1 37 2.7 327 986
March 4.9 3.7 2.3 20.2 15 3.0 293 985
April 5.7 3.5 3.0 26.8 44 3.0 84 980
May 5.2 2.6 3.3 32.7 42 2.8 270 975
June 6.0 2.6 3.8 30.1 52 2.9 265 972
July 4.9 1.6 3.6 29.8 25 2.8 125 971
August 4.6 1.3 3.3 29.0 34 2.2 240 974
September 5.2 1.9 3.7 28.8 87 2.6 299 978
October 4.4 1.8 3.1 26.0 68 2.1 312 983
November 2.9 1.5 1.9 21.0 65 1.9 324 986
December 2.1 0.4 1.8 14.8 83 2.0 291 989

essential to deploy the environmental friendly clean and green power in the country.
The database of solar irradiance and meteorological parameters are measured at
SRRA stations, after that send and stored at CWET-Chennai (NIWE) central station.

3.3 Solar Data of Murthal Station

Solar energy availability is free and in great amount, due to its sustainable and envi-
ronmentally friendly nature which has immense potential to fulfill the energy require-
ment and mitigate the greenhouse gas emissions. The Indian government has set up
the world largest SRRA network. One such station is established in D.C.R.U.S.T,
Murthal (Station ID-2389) in the year 2013. The latitude, longitude and elevation of
the station are 29o 1 40.9 N, 77o 3 28 E and 213 m amsl, respectively. In this study,
we explore the solar irradiance data and other meteorological data of the year 2015
of D.C.R.U.S.T, Murthal station. Table 2 illustrates the average monthly values of
solar irradiance and meteorological parameters of the year 2015.

4 Solar Energy Potential in India

The GOI has released various reforms to increase the solar energy mix in the Indian
power sector so that to sustain the development of the country. The Indian govern-
ment has implemented several policies, schemes and projects, especially since the
182 V. Nandal et al.

commencement of the Jawaharlal Nehru national solar mission (JNNSM). Due to

the various reforms, prominent growth in the solar sector has been observed with an
installed capacity of 2 MW in the 2008–2009 to more than 28,000 MW in 2018–
2019 [10]. Solar power is an emerging industry in the Indian power sector. The Indian
government has given emphasis to expand its solar plans, targeting 100 GW addition
of solar capacity by 2022 [8].
In India, the availability of solar energy is around 5000 trillion kWh/year. The
ability of solar power production in India is about 0.20 kWh/m2 [47, 48]. In terms of
solar capacity, India stands on the tenth position in the world, but in the near future,
India could be one of the highest generators of sun-based power. The present Indian
government reforms were going in the right direction to bring solar revolution by
2022. Moreover, India could emerge as a world leader in solar market to achieve
sustainable manufacturing of solar products along with economical energy. This can
be achieved under the “Make in India” mission so that to decrease the rate of solar
power [49]. The production of solar power is influenced by the various parameters
such as radiation, climate and geographic position and also has a significant effect on
the solar potential estimation [38, 40]. The assessments of the potential exploration
rely on the wasteland selected, solar irradiance and wind speed for solar thermal and
solar photovoltaic power generation.
For setting up of utility-scale solar plants in India, there is plenty of wastelands
available in the country. It is estimated that the 3% wasteland of the nation is sufficient
to generate 748.98 GW of power by harnessing solar energy [8, 50]. Table 3 illustrated
the state-wise estimated solar potential and solar installed capacity as on March
31, 2019 [51, 52]. It is also evaluated that approximately 46.7 million hectares of
wasteland is accessible in India for solar energy [50]. Table 3 shows that the total
estimated solar potential is 748.97 GW while so far India’s total solar installed
capacity as on March 31, 2019 is 28.181 GW. It is observed that there is a huge gap
exists between estimated solar potential and solar installed capacity so far. Therefore,
there is a great opportunity to harness the solar energy and fulfill the entire demand
for electricity of the nation.
Solar energy is mainly dominated in Rajasthan, Gujarat, Maharashtra and Madhya
Pradesh. Figure 5 illustrates the average annual solar radiation potential capacity in
India [53]. Rajasthan and Gujarat region in India has been the highest potential of
radiation around 2150 kWh/m2 . Rajasthan is the highest developed state in terms of
solar power because its estimated solar potential is 142.31 GW and as on March 31,
2019, the total solar installed capacity is 3.227 GW. Also, a 4 GW solar project is
being started in this state.
In northern India, lowest solar potential states are Jammu and Kashmir and
Himachal Pradesh. The western part of India has emerged as a solar power leader due
to various features such the high potential of solar power, easily availability of waste-
land, strong transmission and distribution network, good connectivity and efficient
utilities. At present, installed capacity of solar power in Gujarat state is 2.44 GW
while the solar potential is 35.77 GW there is huge potential to tap solar power in this
state. Maharashtra, MP and southern region of India have the second-highest region
for radiation with 2000 kWh/m2 . A 750 MW solar power project was proposed by
Solar Resource Assessment and Potential in Indian Context 183

Table 3 State-wise estimated

Sr. No States/UTs Solar potential Solar installed
solar potential and solar
(GW) capacity (GW)
installed capacity as on
March 31, 2019 (Source [51], 1 Andhra 38.44 3.085
pp. 2–3, [52]) Pradesh
2 Arunachal 8.65 0.005
Pradesh
3 Assam 13.76 0.022
4 Bihar 11.2 0.142
5 Chhatisgarh 18.27 0.231
6 Delhi 2.05 0.126
7 Goa 0.88 0.003
8 Gujarat 35.77 2.440
9 Haryana 4.56 0.224
10 Himachal 33.84 0.022
Pradesh
11 Jammu & 111.05 0.014
Kashmir
12 Jharkhand 18.18 0.034
13 Karnataka 24.7 6.096
14 Kerala 6.11 0.138
15 Madhya 61.66 1.840
Pradesh
16 Maharashtra 64.32 1.634
17 Manipur 10.63 0.003
18 Meghalaya 5.86 0.001
19 Mizoram 9.09 0.005
20 Nagaland 7.27 0.001
21 Odisha 25.78 0.395
22 Punjab 2.81 0.906
23 Rajasthan 142.31 3.227
24 Sikkim 4.94 0.000
25 Tamil Nadu 17.67 2.575
26 Telangana 20.41 3.592
27 Tripura 2.08 0.005
28 Uttar Pradesh 22.83 0.960
29 Uttarakhand 16.8 0.306
30 West Bengal 6.26 0.075
31 UT 0.79 0.070
Total (GW) 748.97 28.181
184 V. Nandal et al.

Fig. 5 Average annual solar irradiation (kWh/m2 ) potential in India (Source [53], p.13)

MP state. Also, a large utility solar power project of 648 MW in Kamuthi of Tamil
Nadu state was formally committed to the nation. Eastern region of India shows
the lowest solar radiation pattern with less than 1550 kWh/m2 . The lowest installed
capacity of solar power is 0.01 MW in Sikkim due to the low solar intensity pattern.

5 Discussions

The highest value of GHI is 6 kWh/m2 /day and DHI is 3.8 kWh/m2 /day with corre-
sponding ambient temperature is 30.1 °C in the month of June 2015 and found a
Solar Resource Assessment and Potential in Indian Context 185

good relationship between these parameters. In the month of December 2015, the
lowest average value of GHI is observed with the corresponding ambient tempera-
ture of 14.85 °C. The association of high average value of GHI and DHI with a high
average value of ambient temperature is also observed by the Abbas et al. [54] in
his study. The DNI is used for CPV and CSP power plants because this parameter is
helpful for the design of the solar plant and work over a longer duration of time. The
annual cycle of these parameters varies due to variation of seasons in India. From
February to June, this area receives a good amount of solar resources and the low
value of relative humidity is observed, which shows a good relationship among these
parameters. Moreover, relative humidity is the factor that has a great impact on the
solar radiations. It is also observed that when ambient temperature decreases the rela-
tive humidity increases as found in the month of January and October to December.
Abbas et al. [54] and Ettah et al. [55] analyzed the effect of relative humidity on solar
plant generation and found that when relative humidity decreases, the current output
as well as the efficiency of solar power generation system increases. Therefore, a
good correlation is observed among DNI and relative humidity.
In India, there is plenty of wastelands is available and it is sufficient to generate
clean and green power which can fulfill the entire demand of the nation. Moreover, the
estimated potential of solar power is 748.98 GW when considering the 3% wasteland
of the country. It is also evaluated that there is a huge difference among potential
estimation and solar installed capacity. So there is ample opportunity to harness solar
power in India.
In conclusion, this study shows the effect of variation of solar radiation resources
and variation of meteorological parameters in the area of Murthal, SRRA station.
The GHI radiation parameter is explored to study the impact of solar irradiance on
solar photovoltaic as well as on the solar thermal system. GHI is evaluated using the
sum of DNI and DHI with respect to the incident radiation. These meteorological
parameters also have an effect on solar irradiance, which further affects the output
performance of the solar power generation system. In addition to these solar resources
of the country also helps to assess the solar potential, performance and selection of
futuristic solar technologies.

6 Conclusions

Availability of clean, reliable and sustainable energy plays a critical role for industrial
as well as the socio-economic development of any country. It is essential to assess
solar radiation resources and solar potential for harnessing solar energy in India.
Solar and meteorological data plays a critical role for accurate estimation of solar
potential, solar technology selection, research and development in the field of solar,
policymakers and project developers. The conclusions of this study are as follows:
186 V. Nandal et al.

• There are 121 solar resource stations were commissioned in India with a central
receiving station at NIWE-Chennai, which is responsible to disseminate solar data
for the development of solar power in India.
• This study analyzed the one-year solar radiation data from January 2015 to
December 2015 of DCRUST, Murthal station. The maximum monthly average
GHI is 6.0 kWh/m2 /day in the month of June, while the corresponding ambient
temperature is 30.1 °C. The minimum monthly average GHI is 2.1 kWh/m2 /day
in the month of December and the corresponding ambient temperature is 14.8 °C.
• A good relation was found among GHI and ambient temperature. This area
received a good amount of solar irradiance from February to October. More-
over, the association of high GHI with the high value of ambient temperature
behavior is also observed by other authors in a different part of the country.
• The relative humidity is observed at the higher side when the ambient temperature
is at the lower side as seen in the month of January and October to December. In
addition to this relative humidity decreases the DNI part of solar irradiance. Due
to this a good relationship is observed among relative humidity and DNI.
• The maximum landmass of India received greater than 1700 kWh/m2 average
annual solar radiation and this offers huge potential to harness solar power.
• In India, the estimated solar potential is 748.97 GW, while the total solar installed
capacity as on March 31, 2019 is 28.181 GW. Due to the huge difference between
estimated solar potential and solar installed capacity, there is great opportunity to
harness the solar energy and fulfill the entire electricity demand of the nation.
The utilization of advanced instruments for solar resource estimation in India helps
to develop new solar power projects and advance technology. This analysis of solar
resources will support to develop solar maps, evaluate the feasibility of solar projects
and will act as a base for the development of futuristic solar technologies. Moreover,
this study helps to deploy solar energy not only in India but also in neighboring
countries.

References

1. Shukla KN, Sudhakar K, Rangnekar S (2015) Estimation and validation of solar radiation
incident on horizontal and tilted surface at Bhopal, Madhya Pradesh, India. Am-Euras J Agric
Environ Sci 15:129–39
2. Edenhofer O, Pichs-Madruga R, Sokona Y, Seyboth K, Kadner S, Zwickel T, Eickemeier P,
Hansen G, Schlömer S, Von Stechow C, Matschoss P (2011) Renewable energy sources and
climate change mitigation: special report of the intergovernmental panel on climate change.
Cambridge University Press
3. Central Electricity Authority.: Growth of electricity sector in India from 1947–2018. 1–83
(2018a) https://www.cea.nic.in/reports/others/planning/pdm/growth_2018.pdf
4. Central Electricity Authority.: All India installed capacity of power stations. 1–7 (2018b) https://
www.cea.nic.in/reports/monthly/installedcapacity/2018/installed_capacity-02.pdf
5. United Nations FCC (2015) Adoption of the Paris agreement. https://unfccc.int/resource/docs/
2015/cop21/eng/l09.pdf
6. Tiwari GN, Mishra RK (2012) Advanced renewable energy sources. R Soc Chem
Solar Resource Assessment and Potential in Indian Context 187

7. Birol F, Gould T (2015) India energy outlook. World energy outlook special report
8. NITI Aayog and IEEJ (2017) Energizing India, a joint project report of NITI Aayog and IEEJ.1–
144 (2017) https://niti.gov.in/writereaddata/files/document publication/Energy%20Booklet.
pdf
9. Central Electricity Authority (2015) Ministry of Power Monthly all India installed generation
capacity report Government of India
10. Central Electricity Authourity (2019) All India installed capacity of power stations, 1–7. https://
cea.nic.in/reports/monthly/installedcapacity/2019/installed_capacity-03.pdf
11. Sharma T, Balachandra P (2018) Will the integration of renewable energy enable sustainable
transition of Indian electricity system? Energy Strat Rev 31(21):137–148
12. Panwar N, Kaushik S, Kothari S (2011) Role of renewable energy sources in environmental
protection: are view. Renew Sustain Energy Rev 15(3):1513–1524
13. Lewis NS (2007) Toward cost-effective solar energy use. Science 315(5813):798–801
14. Rohankar N, Jain AK, Nangia OP, Dwivedi P (2016) A study of existing solar power policy
framework in India for viability of the solar projects perspective. Renew Sustain Energy Rev
1(56):510–518
15. Pandey S, Singh VS, Gangwar NP, Vijayvergia MM, Prakash C, Pandey DN (2012) Determi-
nants of success for promoting solar energy in Rajasthan. India. Renew Sustain Energy Rev
16(6):3593–3598
16. Nathan HS (2014) Solar energy for rural electricity in India: a misplaced emphasis. Econ Polit
Weekly 4:49(50), 60–67
17. Zell E, Gasim S, Wilcox S, Katamoura S, Stoffel T, Shibli H, Engel-Cox J, Al Subie M (2015)
Assessment of solar radiation resources in Saudi Arabia. Solar Energy 119:422–438
18. Makade RG, Jamil B (2018) Statistical analysis of sunshine based global solar radiation (GSR)
models for tropical wet and dry climatic Region in Nagpur, India: a case study. Renew Sustain
Energy Rev 87:22–43
19. Katiyar AK, Pandey CK (2010) Simple correlation for estimating the global solar radiation
on horizontal surfaces in India. Energy. 35(12):5043–5048. https://doi.org/10.1016/j.energy.
2010.08.014
20. Younes S, Claywell R, Muneer T (2005) Quality control of solar radiation data: present status
and proposed new approaches. Energy. 30(9):1533–1549
21. World Meteorological Organization CIM (2006) Guide to meteorological instruments and
methods of observation. Preliminary Seventh Edition. WMO-No. 8
22. Stoffel T, Renné D, Myers D, Wilcox S, Sengupta M, GeorgeR, Turchi C (2010) Concentrating
solar power: best practices handbook for the collection and use of solar resource data (CSP)
(No. NREL/TP-550-47465). National Renewable Energy Lab.(NREL), Golden, CO (United
States)
23. Nehrir MH, Wang C, Strunz K, Aki H, Ramakumar R, Bing J, Miao Z, Salameh Z (2011)
A review of hybrid renewable/alternative energy systems for electric power generation:
configurations, control, and applications. IEEE Trans Sustain Energy 2(4):392–403
24. Capizzi G, Napoli C, Bonanno F (2012) Innovative second-generation wavelets construction
with recurrent neural networks for solar radiation forecasting. IEEE Trans Neural Netw Learn
Syst 23(11):1805–1815
25. Vindel JM, Polo J, Zarzalejo LF (2015) Modeling monthly mean variation of the solar global
irradiation. J Atmos Solar Terr Phys 122:108–118
26. Celik AN, Muneer T (2013) Neural network based method for conversion of solar radiation
data. Energy Convers Manage 67:117–124
27. Dahmani K, Dizene R, Notton G, Paoli C, Voyant C, Nivet ML (2014) Estimation of 5-min
time-step data of tilted solar global irradiation using ANN (Artificial neural network) model.
Energy 70:374–381
28. Vinod Kumar R, Singh SK (2018) Solar photovoltaic modeling and simulation: as a renewable
energy solution. Energy Rep 4:701–712
29. Boland J, Huang J, Ridley B (2013) Decomposing global solar radiation into its direct and
diffuse components. Renew Sustain Energy Rev 28:749–756
188 V. Nandal et al.

30. Despotovic M, Nedic V, Despotovic D, Cvetanovic S (2016) Evaluation of empirical models for
predicting monthly mean horizontal diffuse solar radiation. Renew Sustain Energy Rev 56:246–
260 10.1016/j.rser.2015.11.058
31. Karakoti I, Das PK, Singh SK (2012) Predicting monthly mean daily diffuse radiation for India.
Appl Energy 91(1):412–425. https://doi:10.1016/j.apenergy.2011.10.012.
32. Kandasamy CP, Prabu P, Niruba K (2013) Solar potential assessment using PVSYST software.
In: 2013 international conference on green computing, communication and conservation of
energy (ICGCE). IEEE, pp 667–672
33. Abubakar GA, Muthuchamy M (2017) Solar energy for electricity generation—a comparative
study of three states in India. In: 1st international conference on large-scale grid integration of
renewable energy in India, At New Delhi India, 1–6
34. Moreno-Tejera S, Ramírez-Santigosa L, Silva-Pérez MA (2015) A proposed methodology for
quick assessment of timestamp and quality control results of solar radiation data. Renew Energy
78:531–537
35. Zafarani R, Eftekharnejad S, Patel U (2018) Assessing the utility of weather data for
photovoltaic power prediction. arXiv preprint arXiv:1802.03913
36. Ramachandra TV, Krishnadas G, Jain R (2012) Solar potential in the Himalayan landscape.
ISRN Renewable Energy. https://doi:10.5402/2012/203149
37. Schwandt M, Chhatbar K, Meyer R, Fross K, Mitra I, Vashistha R, Giridhar G, Gomathi-
nayagam S, Kumar A (2014) Development and test of gap filling procedures for solar radiation
data of the Indian SRRA measurement network. Energy Procedia 57:1100–1109
38. Harinarayana T, Kashyap KJ (2014) Solar energy generation potential estimation in India and
Gujarat, Andhra, Telangana States. Smart Grid Renew Energy 5:275–289. https://doi:10.4236/
sgre.2014.511025
39. Gupta A, Spears D (2017) Health externalities of India’s expansion of coal plants: evidence
from a national panel of 40,000 households. J Environ Econ Manage 86:262–276. https://doi.
org/10.1016/j.jeem.2017.04.007
40. Sharma C, Sharma AK, Mullick SC, Kandpal TC (2017) Solar thermal power generation in
India: effect of potential incentives on unit cost of electricity. Int J Sustain Energy 36(8):722–
737. https://doi:10.1080/14786451.2015.1088016
41. Mahima, Karakoti I, Nandan H, Pathak PP (2018) An analytical study on global radiation and
meteorological parameters for India. Int J Amb Energy 1–11. https://doi:10.1080/01430750.
2018.1507929
42. Raghuwanshi SS, Arya R (2019) Renewable energy potential in India and future agenda of
research. Int J Sustain Eng 1–12. https://doi:10.1080/19397038.2019.1602174
43. Mitra I, Chhatbar K, Kumar A, Giridhar G, Vashistha R, Meyer R, Schwandt M (2014) Solmap:
project in India’s solar resource assessment. Int J Renew Energy Dev 3(3):207–216. https://dx.
doi.org/10.14710/ijred.3.3.207-216
44. Müller J, Mitra I, Mieslinger T, Meyer R, Chhatbar K, Gomathinayagam S, Giridhar G (2017)
Towards building solar in India-A combined mapping and monitoring approach for creating a
new solar atlas. Energy Sustain Dev 40:31–40
45. Joshi A, Patel RJ (2015) Installation and commissioning of an automatic solar radiation moni-
toring system (ASRMS). Refueling Technology Division. https://www.barc.gov.in/publicati
ons/nl/2015/2015030405.pdf
46. Kumar A, Gomathinayagam S, Giridhar G, Mitra I, Vashistha R, Meyer R, Schwandt M,
Chhatbar K (2014) Field experiences with the operation of solar radiation resource assessment
stations in India. Energy Procedia 49:2351–2361
47. Sarkar T, Dan AK, Ghosh S, Das Bhattacharya K, Saha H (2018) Interfacing solar PV
power plant with rural distribution grid: challenges and possible solutions. Int J Sustain
Energy 37(10):999–1018. https://doi:10.1080/14786451.2017.1414051
48. Kar SK, Sharma A, Roy B (2016) Solar energy market developments in India. Renew Sustain
Energy Rev 62:121–133. https://doi:10.1016/j.rser.2016.04.043.
49. Sharma AK, Thakur NS (2016) Analyze the factors effecting the development of hydro power
projects in hydro rich regions of India. Perspect Sci 8:406–408. https://doi:10.1016/j.pisc.2016.
04.090
Solar Resource Assessment and Potential in Indian Context 189

50. Department of Land Resources. Wasteland atlas of India (2011) https://www.dolr.nic.in/Was

telandsAtlas2011/Wastelands_Atlas_2011.pdf.
51. Ministry of New and Renewable Energy (2014) State-wise Estimated Solar Power Potential in
the Country 1–5. https://mnre.gov.in/file-manager/UserFiles/Statewise-Solar-Potential-NISE.
pdf
52. State wise installed capacity of grid interactive renewable power as on 31.03.2019, MNRE.
https://mnre.gov.in> State wise installed capacity as on 31.03.2019.xlsx.
53. Harinarayana T, Kashyap KJ (2014) Solar energy generation potential estimation in India and
Gugarat, Andhra, Telangana States. Smart Grid Renew Energy 5:275–289. https://dx.doi.org/
10.4236/sgre.2014.511025
54. Abbas Z, Harijan K, Shaikh P, Walasai G, Ali F (2017) Effect of ambient temperature and
relative humidity on solar PV system performance: a case study of Quaid-e-Azam Solar Park,
Pakistan”. Sindh Univ Res J-SURJ (Sci Ser) 49(4):721–726
55. Ettah EB, Ushie PO, Obiefuna JN, Nwachukwu NC (2015) Comparative study of the effects
of relative humidity on solar electricity generation in UYO and Port- Harcourt, Nigeria. Int J
Math Phys Sci Res 3(2):66–70
Implementation of Hill Climb Search
Algorithm Based Maximum Power Point
Tracking in Wind Energy Conversion
Systems

Biaklian Tonsing, Shelly Vadhera, and Atma Ram Gupta

Abstract This paper presents cutting edge Hill Climb Search (HCS) based
Maximum Power Point Tracking (MPPT) algorithm for Wind Energy Conversion
Systems (WECS). Because of the erratic attribute of wind, it is difficult to generate
power constantly. So, it is required to extract maximum power from the wind when
it becomes available. This can be achieved by Hill Climb Search Algorithm which
offers the generator to track the Maximum Power Point of the wind turbine system
under fluctuating and inconsistent wind conditions. The proposed Wind Energy
Conversion System model is based on Permanent Magnet Synchronous Generator
(PMSG) and implemented in MATLAB/SIMULINK. The proposed algorithm is
feasible for small-scale wind energy conversion systems.

Keywords HCS · MPPT · WECS · PMSG

1 Introduction

With the fast exhaustion, staggering expense, and negative effects of conventional
sources of energy, there arises the need for non-conventional sources of energy as
an alternative or even a replacement in the near future. Wind energy has become a
popular and viable emerging non-conventional source of energy [1, 2].
Unlike solar energy which can only be captured during daytime, wind energy
system offers significant advantages over its solar counterpart as it can operate during
any time of the day or night subject to the availability of wind energy. However,
the major setback of harnessing wind energy is that it is not consistently available
throughout the day or night. Because of this inconsistent nature, it is necessary to
capture maximum wind energy at maximum power point from the varying wind
speed. In order to achieve this, Maximum Power Point Tracking (MPPT) based on
Hill Climb Search (HCS) is implemented.

B. Tonsing (B) · S. Vadhera · A. R. Gupta

National Institute of Technology Kurukshetra, Kurukshetra, Haryana 136119, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 191

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_18
192 B. Tonsing et al.

2 Wind Turbine and MPPT Concept

The wind turbine characteristics in a Wind Energy Conversion System (WECS) are
discussed below.
Power generated by the air flow is given by [3]

1
Pair = ρ Av 3 (1)
2
Power produced by the wind turbine rotor is given by

Pw = Cp Pair (2)

The efficiency of the wind turbine is described by the power co-efficient function,
Cp (λ, β),

Pm
Cp = (3)
Pw

It shows to the amount of actual power extracted by the wind turbine over the
amount of theoretical power accessible [4]. The two factors that impact the efficiency
are the pitch angle, β and the tip speed ratio (TSR), λ.
The TSR of the wind turbine [3] is given by

ωr R
λ= (4)
Vw

where; Pw = Rotor’s Power (Watts); ρ = Air Density (Kg/m3 ); C p = Power Co-

efficient; A = Rotor Blades’ Area (m2 ); V w = Wind Speed (m/s2 ); ωr = Shaft Speed
of Rotor (rad/s); R = Rotor Swept Area’s Radius (m).
The power characteristics of a typical wind turbine is shown in Fig. 1 for wind
speeds.

3 Proposed HCS MPPT Algorithm

The hill climb search method of control is an efficient optimization method which
uses the principle of searching for the local optimum point of a given function [5].
The MPPT process in the proposed system is based on directly adjusting the dc/dc
boost converter duty cycle. In a fixed step size based HCS method, in order to reduce
oscillation around the peak operating, we can change the value of duty cycle of the
converter by introducing the step size, i.e., D at each sample based on the working
condition. The maximum power point of operation is obtained mathematically when
the following condition is satisfied.
Implementation of Hill Climb Search Algorithm … 193

Fig. 1 Turbine output power (kW) versus turbine speed (m/s)

dPdc
=0 (5)
dVdc

where P is the dc link power and V is the dc link voltage. Like the power versus
speed graph, the function dPdc (Vdc ) has also a single operating point where maximum
power can be achieved.
From Fig. 2, it is clear that as the duty-cycle adjustment is guided by the direction
of slope of the function, i.e., dP dc
dVdc
, the duty-cycle value is increased in the high-speed
side of the WG characteristic. Hence, WG rotor-speed decreases and power increases
till the controller reaches the Maximum Power Point is reached. In the same way,

Fig. 2 Power versus voltage

of DC [6]
Popt

dω
Power

dP

ω opt
Generator Speed ω r
194 B. Tonsing et al.

when the starting point is in the low-speed side, following the direction of slope of
the function, i.e., dPdc
dVdc
which results in a decrease in the value of duty cycle. Hence,
the WG rotor speed is gradually and the controlling variable subsequently converges
at the MPP [7–9].
Flow chart for MPPT algorithm describes the step by step working of a MPPT
controller based on the described algorithm. As shown in Fig. 3, the inputs taken are
the dc link voltage and current and output is the required duty cycle. In the second
step, the controller is computing the dc link power by multiplying mathematically
the input dc link voltage and current. The next step explains (n − 1)th value of dc link
power which is obtained by using a delay function in Simulink. If the difference in

START

Read
V dc (n),
I dc (n)

Compute P dc (n)

Delay P dc (n),
V dc (n), I dc (n)

YES NO
P dc (n)-P dc (n-1)>0

V dc (n)- V dc (n-1)>0 V dc (n)- V dc (n-1)<0

YES NO YES NO

d=d+ ∆d d=d- ∆d d=d+ ∆d d=d- ∆d

TO SWITCH

Fig. 3 Flowchart of HCS MPPT

Implementation of Hill Climb Search Algorithm … 195

nth and (n − 1)th value of dc link power is greater than zero then the algorithm goes
forward and computes the difference in dc link voltage. Accordingly, the difference
in dc link voltage generates new duty cycle by adjusting delta D which is added or
subtracted as per the conditions applied.

4 Simulation of PMSG Based WECS

The simulation circuit of the proposed system is implemented in

SIMULINK/MATLAB as shown in Fig. 4. The model comprises of boost converter
controlled by a hill climb search to achieve a highly efficient output.
The subsystem model of the wind turbine is in Fig. 4 is represented in Fig. 5.
Here, the inputs are the wind speed and generator speed and the output is the torque
applied to the generator shaft. The torque of the generator is based on the generated
power and speed. The wind turbine setup is designed using MATLAB. The rotor
shaft is driven by the wind turbine who produce the mechanical torque according to

Fig. 4 Diagram of simulated wind energy system

196 B. Tonsing et al.

Fig. 5 Sub-circuit of wind generation system

generator and wind speed values. The electrical power output of the generator (stator
winding) is connector to the grid.

5 Results and Discussion

When the base wind input speed is given at 12 m/s, the rotor speed (rad/s) of the
turbine for a duration of 10 s is shown in Fig. 6. The torque generated by the turbine
in Watts is shown is Fig. 7. The Pitch angle β is set to zero as shown in Fig. 8.
The generated Grid voltage V ab and current I ab are represented in Figs. 9 and 10
respectively. The output power generated by the system is shown in Fig. 11.
Table 1 shows the comparison of power generated at various wind speeds of 0, 4,
6, 8, 10, and 12 m/s with HCS based MPPT algorithm and without MPPT inferred
from the results carried out in the work.

Fig. 6 Turbine rotor speed w.r.t. time

Implementation of Hill Climb Search Algorithm … 197

Fig. 7 Turbine torque generated w.r.t. time

Fig. 8 Pitch angle w.r.t. time

Fig. 9 Voltage generated w.r.t. time

Fig. 10 Current generated w.r.t. time

198 B. Tonsing et al.

Fig. 11 Simulated output power of proposed MPPT w.r.t. time

Table 1 Comparison of turbine output with and without HCS-based MPPT

Wind speed (m/s) Turbine power without Turbine power with MPPT Difference (W)
MPPT (W) (W)
0 0 0 0
4 885.55 886.1 0.55
6 1295 1299 4
8 2373 2382 9
10 3944 3968 24
12 6220 6250 30
As the wind speed increases, the difference in the turbine power with and without MPPT increases
as shown in the table

6 Conclusions

Hill Climb Search based MPPT technique is implemented to provide high power
output. From all the results and comparisons made above, it can be concluded that
by using HSC technique in WECS, the electrical loses can be mitigated, thereby,
increasing the reliability and efficiency compared to conventional systems. The
proposed algorithm was tested under different wind conditions like constant wind
speed, randomly varying wind speed and abruptly changing wind speed. In all cases,
the extracted power is at its peak with respect to the wind curve for the turbine. The
reduced ripple power and increased efficiency are the biggest advantages of HCS
algorithm.

References

1. Abdullah MA, Yatim AHM, Tan CW (2011) A study of maximum power point tracking algo-
rithms for wind energy system. In: 2011 IEEE, first conference on clean energy and technology
(CET), pp 321–326.
Implementation of Hill Climb Search Algorithm … 199

2. Saidur R, Islam MR, Rahim NA, Solangi KH (2010) A review on global wind energy policy.
Renew Sustain Energy Rev 14:1744–1762
3. Freris LL (1990) Wind energy conversion system. Prentice Hall, London
4. Muyeen SM, Takahashi R, Murata T, Tamura J (2010) A variable speed wind turbine control
strategy to meet wind farm grid code requirements. IEEE Trans Power Syst 25(I):331–340
5. Hyeung-Gyun Kim, Dong-Choon Lee, Jul-Ki Seok, G-Myoung Lee (2003) Stand-alone wind
power generation system using vector-controlled cage-type induction generators. In: Proceed-
ings of international conference on electrical machines and systems, vol 1, pp 289–292, Nov
2003
6. Grabic S, Celanovic N, Katic VA (2008) Permanent magnet synchronous generator cascade for
wind turbine application. IEEE Trans Power Electron 23(3):1136–1142
7. Chinchilla M, Arnaltes S, Burgos JC (2006) Control of permanent magnet generators applied
to variable-speed wind-energy systems connected to the grid. IEEE Trans Energy Conver
21(1):130–135
8. Nakamura T, Morimoto S, Sanada M, Takeda Y (2002) Optimum control of PMSG for wind
generation system. In: Proceedings of the power conversion conference, vol 3. PCC Osaka, pp
1435–1440
9. Tanaka T, Toumiya T, Suzuki T (1997) Output control by Hill Climbing method for a small scale
wind power generating system. Renew Energy 12:387–400
Analytical Review of Solar Cell
as Globalized Approach

Srishtee Chaudhary and Rajesh Mehra

Abstract Sustainable development is emerging as an interdisciplinary issue to

deal with environment prospects considering the future of the planet. The glob-
ally changing scenario is witnessing the extinction of resources and its adverse effect
on environment. The limitation regarding current world scenario has led researchers
to explore the perspectives related to renewable energy for sustainable development
of the planet. Investigating the possibilities with renewable resources of energy as
these resources are being available in abundance. Among all renewable resources of
energy solar energy is utmost available clean and abundant. Solar energy cells based
upon organic–inorganic materials are researched and tested globally for better effi-
ciency. In this paper, the evolution mechanism of solar cells and their impact on the
environment are discussed. This paper presents the review of various technologies
and the synthesis methods of the solar cell based on the performance characteristics.
The challenges and prospects to develop better solar cells are also reviewed and
discussed.

Keywords Solar cell · Perovskite solar cell (PSC) · Power conversion efficiency ·
Hole transport layer · Electron transport layer

1 Introduction

As centuries passed, it is observed that many revolutions resulted in degrading

impacts to the environment. The increased standard of living is on the cost of
destroyed nature hood, pollution is the aftermath of industrialization. Plastics and
plastic industries are becoming a threat to the planet and living due to its inability to
decompose. The planet is at the periphery of extinction of non-renewable resources
of energy so there is an urgent requirement to search and research for alternatives
through renewable resources. Awareness for sustainability influenced the researchers
to propose and test work related to renewable energy. Various common sources of

S. Chaudhary (B) · R. Mehra

National Institute of Technical Teachers Training and Research, Chandigarh, UT, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 201

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_19
202 S. Chaudhary and R. Mehra

renewable energy are solar (sun) energy, water, geothermal, and wind [1]. The main
advantage of these resources is that they are replenished, are available in abun-
dance, are environment-friendly, and lead to clean technologies. The haul from non-
renewable resources to renewable resources is mainly due to the fact that they are
infinite. Also, a lot of research is going to have production of energy at a massive
scale using renewable energy, as their implementation is also expensive.
Globally, the work for sustainable development is mainly focussed upon
decreasing the degradation of resources and proposing perspectives for renewable
energy. Solar energy is the utmost available source of energy. Solar energy harvested
mostly by photovoltaic solar cells so far. But research in perovskite solar cells has
shown increased solar cell efficiency from 3.8% in 2009 to 22.7% recently in single
junction architectures and 26.7% in silicon-based tandem cells [2]. Most common
applications of solar cells include solar lighting, solar cars, power plants, homes,
ventilation system, satellites. A solar cell working principle is to convert the light
energy absorbed from the sun into electricity using the phenomena of the photovoltaic
effect. The energy in the form of photons from Sun absorbed through a material in
the solar cell. After this electron is raised to a higher energy state, then the flow of
this high-energy electron passes through to an external circuit.
Solar cells typically get their name from semiconducting materials used in and for
their fabrication and materials must absorb sunlight. Solar cells have evolved from
first-generation, second-generation, third-generation cells to currently perovskites
as fourth generation. The traditional cells made of crystalline silicon fall under
first generation (photovoltaic with polysilicon and monocrystalline silicon mate-
rials). Thin-film solar cells constitute second-generation cells that include amor-
phous silicon, cadmium telluride (CdTe), and copper indium gallium selenide (CIGS)
cells. The number of emerging photovoltaic thin-film technologies as copper zinc tin
sulfide (CZTS), dye-sensitized solar cell, organic solar cell and Quantum dot solar
cell included for the third generation and are emerging technologies with a lot of
research work being carried for them. Monocrystalline and polycrystalline Si have
high efficiency but result in higher manufacturing costs due to the supply shortage of
silicon. On the contrary, CIGS and CdTe are low-priced. The following table summa-
rizes the different technology-based solar cells for respective efficiency documented
[3].

2 Opto-Electronic Properties of Perovskite Solar Cells

The oxide perovskites are mostly studied due to electric properties of ferroelec-
tricity and superconductivity.CH3 MH3 PbI3 perovskite has been reported with 10
times greater absorption coefficient than conventional ruthenium-based modular dye.
A non-sensitization-type PSC with mixed haide perovskite CH3 NH3 PbI3−x Clx that
was coated with Al2 O3 film observed PCE of 10.9%. Even though dye-synthesized
reported PCE of greater than 12%, there were concerns of leakage due to elec-
trolyte usage. A HTM is required for filling the pores of dye-sensitized mesoporous
Analytical Review of Solar Cell as Globalized Approach 203

TiO2 film so that hetro-junction can be induced. A molecular-type HTM 2,2 ,7,7 -
Tetrakis(N,N-p-dimethoxy-phenylamino)-9,90-spirobifluorene (spiro-MeOTAD) is
preferred for pore filling. The thickness of TiO2 film is inversely proportional to
absorption coefficient, higher absorption coefficient thinner the film, and vice versa.
Spiro-OTAD limits the TiO2 film thickness to 2 um [4]. In case of perovskite (oxide),
an perfect cubic perovskite is envisioned when tolerance factor t is 1 and if t < 1
octahedral distortion is there affecting electrical properties [5].
It is known that ions losing one or more electrons acquiring positive charge form
cations. But perovskite structure is only cations with the radius between ≈1.6 and
≈2.5 Å. Methylammonium with its iconic radius ≈1.8 Å is appropriate for lead
halide perovskites. PSC can work without electron injecting layer. The solid solution
CH3 NH3 PbI3−x Brx perovskite reported stability in humidity soaking test. CuI is
considered efficient inorganic HTM due to its high hole conductivity [6]. But both
CuI and CuSCN suffer from low Voc that is associated with their recombination issue
because of free iodine inducing valence band trap states. In solar cell applications,
narrow bandgap is favorable. The stability issues of organolead halide perovskites
do not propose them for liquid electrolyte-based sensitized solar cells. Substituting
solid hole conductor instead of liquid electrolyte for basic dye-sensitized solar cells
solves leakage concern. In HTM, conductivity and hole diffusion length need to be
examined concurrently as they affect the thickness of TiO2 film. To trap maximum
of incoming light, thickness of 10 um TiO2 film is needed. The inclusion of bromide
in CH3 NH3 PbI3 enhances the stability. Controlling the luminescent property of PSC
can improve Voc [7]. The substitution element is required instead of Pb element to
overcome toxicity issue for environment.
The analysis of CH3 NH3 PbX3 organic lead halide perovskites signified that the
cation A is organic (methyl-ammonium CH3 NH3 ]+ ), M to be Pb, and X can be I, Br,
Cl. The Pb and X relate to the optical absorption. The bromides compounds display
higher bandgap and luminescence at shorter wavelengths whereas compounds with
X=I are vice versa [8].
For sharper absorption peaks, Cl and Br can be employed. Variation in absorption
coefficient affects the optical design SC. PbX6 octahedra impact the optical proper-
ties. The methylammonium cations indirectly confer the band edges for both VB and
CB. PDOS (projected density of states) is suggested for electronic structures rather
than for photoexcitation properties [9].

3 Structure, Fabrication and Mechanisms in PSC

Solar cells commercialization incorporated crystalline, multicrystalline, and CdTe

solar cells. PSC is high performance with low CO2 emissions but encounters stability
issues, film growth and deposition, lifetime and toxicity [10]. The good film unifor-
mity is must for fabrication of large-area high-quality perovskites. The spin-coating,
doctor blading, slot die coating, printing, spray-deposition, soft-cover deposition,
204 S. Chaudhary and R. Mehra

dip coating, and vapor-based deposition are various thin-film deposition technolo-
gies [11, 12]. For thin films, acquiring small area spin-coating is preferred. The drying
time of film, film and substrate temperature, and solvent unstability are parameters
that are taken into account for spray coating process. Doctor blading is realistic under
ambient conditions. The cost of any PSC is subject to the material used and fabrica-
tion process involved but output is determined by the efficiency and lifetime of solar
modules.
TCO can be considered in case of large area devices. But TCO encounters
decreased FF due to its sheet resistance. This issue is resolved through wired trans-
parent electrodes [13]. Spin-coating fabrication technique is not suitable for large and
high throughput thin films. Also during the spin-coating fabrication process, a large
amount of solution is wasted, thus increases costs. In case of the doctor blading fabri-
cation process, there is no solution waste. The fabrication of solar cells with printing
technique is easy but the main challenge is infiltration of perovskite precursor solu-
tion resulting in lower efficiency. In vaccum deposition method, the need of high
vacuum leads to increase fabrication cost. The slot die is used for organic solar cells
[14]. The ETL TiO2 deposition can be through spray coating and screen printing.
The HTL spiro-MeOTAD can be deposited via vacuum deposition. Spiro-MeOTAD
as HTM avoided corrosive liquid electrolyte in perovskite DSSC.
Si based SC features being economical and stability at high temperatures whereas
organometallic halide perovskites features wider spectrum solar absorption and
low nonradioactive recombination losses. Solid hole conductor improved stability
compared to liquid HC [15]. Br with low ionic radius improved efficiency and stability
of mixed halide perovskites. PCE of CH3 NH3 PbBr3 is less than CH3 NH3 PbI3 [16].
Organoled halide perovskites have greater absorbance. An insulating Al2 O3 showed
quicker charge collection compared to TiO2 mesoporous-based SC.
Lanthanum (La)-doped BaSnO3 (LBSO) with good electron mobility keeps 93%
of its initial performance even after 1000 h whereas TiO2 degrades after 500 h.
Methylammonium lead iodide (MAPbI3 ) with bandgap 1.55 eV project considerable
optical and electrical properties [17]. Commercialization of PSC work is proposed for
humidity-resistance, stability and encapsulation materials. Using planar or meso n-
i-p PSC makes it difficult to deposit uniform pin-hole free films at lower temperature
(<100 °C). Therefore, these architectures are fabricated at high temperature because
TiO2 is used as electron-transporting layer (ETL) [18]. Meso-superstructured SC
with p-i-n configuration has the potential of low temperature processing. The n-i-p
MSSCs shows higher efficiency using SnO2 as ETL with improved Voc. The limited
sheet resistance of ITO results in decreased fill factor which ultimately provides
decreased overall efficiency for large area devices. Utilizing 100 nm mesoporous
Al2 O3 for small area devices improved the photovoltaic performance compared to
the planar counterpart due to reduced surface trap states [19]. For better stability
possibilities, perovskite composition can be varied and the use of inorganic selective
contacts could be considered. Varying device area also affects the PCE. Addition of
Al2 O3 layer hinders the degradation process for better stability. Doping treatments
can be performed for desired better results.
Analytical Review of Solar Cell as Globalized Approach 205

4 Unconventional PSCs

Incorporation of bromine in perovskies exhibits stability, whereas Cl exhibits high

charge generation due to greater diffusion lengths, whereas spiro-MeOTD lacks dura-
bility. HTMs PEDOT: PSS lack chemical stability and good lifespan. Some efficient
HTM are namely copper iodide, Cu2 O, NiOx and spiro-MeOTAD. Similarly, some
of the ETMs are ZnO, TiO2 , SnO2 . But TiO2 features complications as degradation
under UV illumination and surface traps due to oxygen vacancy [20, 21]. The utiliza-
tion CuI exhibited PCE 25.68% compared to 24.83% of spiro-OMeTAD with 3.42%
improvement. PSC can be proposed with non-oxide materials as ETL. Inorganic p-
type copper-based HTMs provide high current density, chemical stability and higher
absorption coefficient. The doping concentration of ETM and HTM can intensify the
performance of PSC.
The selection of optimum values for electron affinities of HTM and ETM mini-
mizes recombination at interfaces. In a PSC, donor density of ZnO as ETL is less than
acceptor density of CuSCN as HTL. Carbon nanotubes films used for PSC exhibit
high efficiency by doping the CNT as doping increases the conductivity [22]. Imple-
mentation of high mobility oxide is the best way to increase efficiency, for this copper
oxide and nickel oxide can be considered. Increasing mobility of HTL enhances effi-
ciency but decreased mobility results in an increase in the resistance of HTL. This
increased resistance further result in drop in FF and Jsc. The reduction in acceptor
doping concentration in tin-based PSC in active layer improved performance. By
increasing the work function of the contact, efficiency can be enhanced. Tin-based
halides can be optimized for better efficiency as lead-based is toxic in nature. So
for efficient PSC the materials with high mobility as HTM can be explored as they
ensure less series resistance, low diffusion length and efficient carrier transport.
For HTM and ETM, various organic and inorganic materials can be used. Organic
HTMs so far employed are Spiro-MeOTAD, PEDOT: PSS, P3HT, and PTAA but
possess high materials cost and fabrication cost and lack of chemical stability with
poor durability. For these reasons, inorganic HTMs are being implemented such as
CuSCN, Cu2 O, CuI as they provide s higher stability, cost-effectiveness, and high
absorption coefficient. TiO2 and ZnO as ETMs are available in abundance and are
most commonly employed due to electron affinity and large surface area. The ETMs
thickness can be varied from 30 to 60 nm. But for higher thickness, performance
decreases for both TiO2 and ZnO due to partial absorption of incident light [23, 24].
Maximum quantum efficiency achieved is in range of 300–800 nm. The increase
in defect levels hampers the performance by causing the recombination of charge
carriers before reaching the transport materials due to uneven surfaces. ZnO devices
are challenging choice for practical applications as they are more sensitive to the
applied defects. TiO2 lacked in electron mobility. Highly conductive ETM can stabi-
lize the device. Sb2 Se3 (antimony selenide) compound features high absorption
coefficient with 1–1.3 eV bandgap energy and also features non-toxic, low-cost,
and natural abundance. Sb2 Se3 thin films can be deposited using vapor transporting
deposition method. Lower efficiency cells have higher non-Ohmic current. Research
206 S. Chaudhary and R. Mehra

was carried for Sb2 Se3 film samples at three different positions (d = 21, 22 and
23) and analyzed with area of unit cell 0.3 × 0.3 cm2 . A solar simulator (Newport
Xe-lamp) with a source meter (Keithley 2400) at 100 mW/cm2 , and AM1.5G illumi-
nation was used for current–voltage (J-V) measurement of Sb2 Se3 thin-film SC. It is
observed that for better efficiency absorbent films need to be more crystalline. Also,
carrier recombination is dominated by interface recombination. Tunneling enhanced
recombination can also be proposed. On the contrary, higher recombination rate
provides smaller Voc. Also, the saturation current and carrier recombination affect
the efficiency [25].
One of the exciting research is perovskites can drink coffee at perfect temperature
to serve 85 °C. The concept of molecular lock is studied. This molecular lock is
the outcome of strong interaction between caffeine and Pb2+ ions. The interaction
increases the activation energy during film crystallization, provides improvement
in electronic properties, and enhances thermal stability. The black phase of cesium
(Cs)- and formamidinium (FA)-based PVSKs is thermodynamically unfavorable at
room temperature but the tetragonal black phase of MA PVSKs are stable at low
temperatures. Aluminum-doped zinc oxide showed thermal stability of 500 h at
85 °C in air. Hole-conductor free architectures made by infiltrating porous layers of
carbon and ZrO2 also observed better stability. Photoluminescence of caffeine PSC
is 6 times enhanced. The fast decay phase lifetime decreased from 57.8 to 41.9%,
and the lifetime increased from 7.4 to 21.98 ns with the concept of molecular lock.
But it is difficult to slow down the crystal growth without sacrificing its electronic
properties [26].

5 Conclusion

The stability problems of PSC are due to the environmental factors (oxygen, humidity,
heat, and illumination) and layers adjoining to the perovskite. Intrinsic factors also
degrade the performance as material or material composition and working envi-
ronment. PSC is composed of a transparent conductive electrode, electron trans-
port layer, hole transport layer, a perovskite layer, and a back/front-contact metal
electrode. Transparent conductive electrodes ITO or FTO are the most expensive
element of PSC. The gold or silver is commonly used back-contact metal electrode.
Gold is expensive while silver has degradation issues. Copper back-contact elec-
trode provides high performance and stability. Carbon electrodes can be used for
low temperatures. Gold electrode causes degradation in performance as Au atoms
diffuse into the spiro-MeOTAD and perovskite layers under high temperature. Low-
cost Ag electrodes readily react with perovskite degrading its performance even in an
encapsulated device. Mixed-cation and mixed-anion PSC can be explored for better
stability. Degradation can be overcome by suitable encapsulation techniques and UV
filters utilization. Also, encapsulation counterparts the toxicity of toxic element as
Pb. But the use of UV filters affects light absorption capability because photons are
lost. Solution to this is using photocurable fluoropolymer coating outside solar cells
Analytical Review of Solar Cell as Globalized Approach 207

forming barrier against moisture. This re-emits visible light and absorbs UV light
with no loss.

SOLAR CELLS

TH
ST 4 GEN
1 GEN

ND RD
2 GEN 3 GEN PEROVSKITE
MONOCRYSTALLINE

POLYCRYSTALLINE
AMORPHOUS MULTIJUNCTION
SILICON

CADMIUM
DSSC
TELLURIDE

CIGS ORGANIC

QUANTUM DOT

Fig. 1 Solar cells generation overview

Table 1 Solar cells

S. No. Efficiency documented Solar cell technology
technology-based efficiencies
type
[3]
1 11.9 Dye-sensitized solar cell
2 25.6 Si crystalline cells
3 21.3 Si multicrystalline cells
4 11.2 Organic solar cell
5 21.0 Cadmium telluride (Cd
Te)
6 21.0 copper indium gallium
selenide (CIGS)
7 19.7 Perovskite
8 38.8 Multi-junction
208 S. Chaudhary and R. Mehra

Table 2 AMX3 structure

A signifies large organic/inorganic
Cation
M signifies smaller metal cation
X3 signifies An anion from halide series

References

1. Khan NMA, Alrikabi MA (2014) Renewable energy types. J Clean Energy Technol 2(1):61–64
2. Yang WS, Park B, Jung EH, Jeon NJ, Kim YC, Lee D, Shin SS, Seo J, Kim EK, Noh JH,
Seok S (2017) Iodide management in formamidinium-lead-halide-based perovskite layers for
efficient solar cells. Science 356(6345):1376–1379. https://doi.org/10.1126/science.aan2301
3. Ranabhat K, Patrikeev L, Revina AA, Andrianov K, Lapshinsky V, Sofronova E (2016) An
introduction to solar cell technology. Paper number: 14(2016)4, 405, 481–491. https://doi.org/
10.5937/jaes14-10879May
4. Kim H-S, Lee CR, Jang I-H, KangW, Park N-G (2012) Effect of overlayer thickness of hole
transport material on photovoltaic performance in solid-sate dye-sensitized solar cell. Bull
Korean Chem Soc 33(2):670–674
5. Rinni M, Tobey R, Dean N, Itatani J, Tomioka Y, Tokura Y, Schoenlein RW, Cavaller A (2007)
Control of the electronic phase of a manganite by mode-selective vibrational excitation. Nat
Int J Sci 449:2–74
6. Christians JA, Fung RCM, Kamat PV (2014) An Inorganic hole conductor for organo-lead
halide perovskites solar cells. Improved hole conductivity with copper iodide. J Am Chem
136(2):758–764
7. Ogomi Y, Morita A, Tsukamoto S, Saitho T, Fujikawa N, Shen Q, Toyoda T, Kenji Y, Pandey
SS, Ma T, Hayase S (2014) CH3 NH3 Snx Pb(1−x) I3 perovskite solar cells covering upto 1060 nm.
J Phys Chem Lett 5(6):1004–1011
8. Mosconi E, Amat A, Nazeeruddin MK, Gratzel M, Angelis F (2013) First-principles modeling
of mixed halide organometal perovskites for photovoltaic application. J Phys Chem C
117:13902–13913. https://doi.org/10.1021/jp4048659
9. Kitazawa N, Watanabe Y, Nakamura Y (2002) Optical properties of CH3 NH3 PbX3 (X =
halogen) and their mixed-halide crystals. J Mater Sci 37:3585–3587. https://doi.org/10.1023/
A:1016584519829
10. Chu S, Cui Y, Liu N (2017) The path towards sustainable energy. J Nat Mater 16:16–22
11. Cai M, Wu Y, Chen H, Yang X, Qiang Y, Han L (2016) Cost-performance analysis of perovskite
solar modules. J Adv Sci 4:1–6
12. Espinosa N, Serrano-Lujan L, Urbina A, Krebs FC (2015) Solution and vapour deposited lead
perovskite solar cells: ecotoxicity from a life cycle assessment perspective. J Solar Energy
Mater Solar Cells 137:303–310
13. Galagan Y, Coenen EWC, Verhees WJH, Andriessen R (2016) Towards the scaling up of
perovskite solar cells and modules. J Mater Chem A 4:5700–5705
14. Cotella G, Baker J, Worsley D, Rossi FD, Pleydell-Pearce C, Carnie M, Watson T (2017) One-
step deposition by slot-die coating of mixed lead halide perovskite for photovoltaic applications.
J Solar Energy Mater Solar Cells 159:362–369
15. Heo JH, Song DH, Han HJ, Kim SY, Kim JH, Kim D et al (2015) Planar CH3 NH3 PbI3
perovskite solar cells with constant 17.2% average power conversion efficiency irrespective of
the scan rate. J Adv Mater 27(22):3424–3430
16. Zhang M et al (2014) Composition-dependent photoluminescence intensity and prolonged
recombination lifetime of perovskite CH3 NH3 PbBr3−x Clx films. J Chem Commun
50(79):11727–11730
17. Shin SS et al (2017) Colloidally prepared La-doped BaSnO3 electrodes for efficient, photostable
perovskite solar cells. J Sci 356(6334):167–171
Analytical Review of Solar Cell as Globalized Approach 209

18. Ramirez D, Escobar MAM, Montoya JF, Jaramillo F (2016) Understanding the role of the
mesoporous layer in the thermal crystallization of a meso-superstructuredperovskite solar cell.
J Phys Chem C 120:8559–8567. https://doi.org/10.1021/acs.jpcc.6b02808
19. Ramirez D, Schutt K, Montoya JF, Mesa S, Lim J, Snaith HJ et al (2018) Meso-superstructured
perovskite solar cells: revealing the role of the mesoporous layer. J Phys Chem C 122:21239–
21247 (2018). https://doi.org/10.1021/acs.jpcc.8b07124
20. Yue Y, Salim N, Wu Y, Yang X, Islam A (2016) Enhanced stability of perovskite solar
cells through corrosion-free pyridine derivatives in hole-transporting materials. J Adv Mater
28(48):10738–10743
21. Leijtens T, Eperon GE, Pathak S, Abate A, Lee MM, Snaith HJ (2013) Overcoming ultraviolet
light instability of sensitized TiO2 with meso-superstructured organometal tri-halide perovskite
solar cells. J Nat Commun 4:2885
22. Ou QD, Li C, Wang QK, Li YQ, Tang JX (2017) Recent advances in energetics of metal halide
perovskite interfaces. J Adv Mater Interfaces 4:1600694
23. Casas GA, Coppelletti MA, Cedola AP, Soucase BM, Peltzer y Blanca EL (2017) Analysis of the
power conversion efficiency of perovskite solar cells with different materials as hole-transport
layer by numerical simulations. J Superlattices Microstruct 107:136–143
24. Yu Z, Sun L (2017) Inorganic hole-transport materials for perovskite solar cells. Small Methods
2(2):1700280, 1–16
25. Tao J, Hu X, Xue J, Wang Y, Weng G, Zhu SCZ, Chu J (2019) Investigation of electronic
transport mechanisms in Sb2 Se3 thin-film solar cells. J Solar Energy Mater Solar Cells 197:1–6
26. Wang R, Xue J, Meng L, Yuan Y, Huang Y, Yang Y (2019) Caffeine improves the performance
and thermal stability of perovskite solar cells. Joule 3:1–14 (2019). Elsevier Inc. https://doi.
org/10.1016/j.joule.2019.04.005
Active Battery-Balancing Using Ćuk
Converter

Nikhil Sarode and M. Divya

Abstract One of the major contributors to environmental pollution is the automo-

bile industry, which uses convention fuels causing air pollution and various other
environmental issues. Electric vehicles (EVs) use electrochemical cells as fuel and
are gaining popularity due to higher fuel economy and lower carbon emissions. In
this paper Ćuk Converter based active battery balancing is discussed. Circuit analysis
using MATLAB/SIMULINK and its hardware implementation along with the results
obtained are reported.

Keywords Electric vehicle · Li-Ion battery · Battery balancing · Ćuk converter ·

DCVM

1 Introduction

The Li-Ion battery pack used in electric vehicle consists of cells connected in series–
parallel combinations to meet the required voltage, current, and energy so as to drive
the electric motor for a specific distance. Each cell in the battery pack is different from
the other in terms of capacity, internal resistance, self-discharge rate, and leakage
current [1, 2]. Such a battery pack when charges or discharges at high current results
in voltage imbalance after some charge–discharge cycles. This may result in over-
charging and over discharging of some cells. Overcharge causes cells to heat and
further increase the imbalance. In the worst case, thermal runaway can occur and
battery pack may catch fire, causing danger to the life of passengers in the vehicle
[3, 4]. To prevent this situation battery balancing is critical.
Battery balancing can be done in two ways viz. Passive method and Active method.
Passive balancing circuit dissipates the excess energy from a highly charged cell

N. Sarode (B) · M. Divya

Department of Electrical Engineering, Fr. C. Rodrigues Institute of Technology, Vashi, India
e-mail: [email protected]
M. Divya
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 211

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_20
212 N. Sarode and M. Divya

in the form of heat through resisters. On the other hand, active balancing circuit
transfers the excess energy from a cell with a higher state of charge to the cell
with a lower state of charge. Active balancing topologies are further classified as
Capacitor based, Inductor based and DC-DC Converter based topologies. Capacitor
based balancing circuits are very efficient but balancing speed is very slow. Inductor
based balancing circuits balances the cells quickly but are not as efficient as that of
capacitor-based balancers. DC-DC converter based balances combine the advantages
of both capacitor and inductor based balancing circuits [5–7]. In this paper, Ćuk
converter topology is analyzed and a hardware prototype for a 4S battery pack is
implemented using a simple control scheme.
Ćuk converter can be operated in three modes viz. Continuous Inductor Current
Mode (DICM), Continuous Capacitor Voltage Mode (CCVM) and Discontinuous
Capacitor Voltage Mode (DCVM) [8, 9]. In DCVM, switches are operated when
switch voltage is zero, which reduces the current stress on the switches [8, 10].
Section 2 describes the Ćuk Converter topology and circuit operation in DCVM.
Experimental setup and hardware results are explained Sect. 3 and conclusions are
summarized in Sect. 4.

2 Scheme for Battery Balancing Using Ćuk Converter

Figure 1 shows the 4S battery pack with a Ćuk converter battery balancing circuit,
charger, and load. Cell voltages are represented as V B1 , V B2 , V B3, and V B4 respec-
tively. Converter 1, shown in Fig. 2, will perform balancing operation for V B1 and
V B2 , converter 2 will perform balancing operation for V B2 and V B3 and converter
3 will perform balancing operation for V B3 and V B4 . All the converter circuits are
identical. Converter 1 consists of two inductors, L 1 and L 2 , one energy transfer

Fig. 1 4S battery pack with

Ćuk converter balancing
circuit
Active Battery-Balancing Using Ćuk Converter 213

Fig. 2 Converter module 1

capacitor, C 1 and two power MOSFETs Q1 and Q2 with body diodes D1 and D2 .
Inductors L 1 and L 2 are large enough to have continuous current and capacitor C 1
is small enough that it completely discharges during switching period T S so that the
converter operated in DCVM.
The analysis is done for converter 1 when V B1 > V B2 . In this case, switch Q1
is controlled and switch Q2 is kept ideal. Assuming the capacitor C 1 is charged
at maximum value initially before switch Q1 is turned ON. Initially, switch Q1 is
ON and diode D2 gets reverse biased as shown in Fig. 3a. V B1 charges inductor L 1
and capacitor C 1 discharges through inductor L 2 and charges cell V B2 . Energy from
capacitor C 1 is completely transferred to cell V B2 and capacitor voltage V C1 becomes
zero.
Since switch Q1 is still ON and capacitor voltage V C1 = 0, diode D2 becomes
forward biased as shown in Fig. 3b. Thus, energy stored in inductor L 2 is transferred
to cell V B2 . V B1 continues to charge L 1 . Switch Q1 is turned OFF and diode D2
continues to remain in forward bias from the previous interval as shown in Fig. 3c.
In this mode capacitor voltage V C1 starts to charge through L 1 and cell V B1 from
zero and reaches the peak.
To operate Ćuk converter in DCVM, following conditions should be satisfied [10],

VB2
D≥ 2 f s C1 (1)
I L2

VB2 (1 − D)2
L1 ≥ (2)
2D f s I L2
VB2 (1 − D)
L2 ≥ (3)
2 f s I L2

3 Simulation and Experimental Results

3.1 Simulation Results

A MATLAB/SIMULINK model was created for four 4S battery pack with three Ćuk
converter balancing circuit to validate the working of the circuit. Battery model was
taken from electric drives library from SIMULINK [11]. Cell parameterization was
214 N. Sarode and M. Divya

Fig. 3 a Equivalent circuit for Q1 ON. b Equivalent circuit for Q1 and D2 ON. c Equivalent circuit
for D2 ON

done according to the Li-Ion cell manufacturer datasheet used in hardware prototype
[12]. State of charge of the cells was initialized as 50, 51, 52 and 53% for cell 1,
cell 2, cell 3 and cell 4 respectively, hence the terminal voltages are also different
initially. Inductors L 1 = L 2 = L 3 = L 4 = L 5 = L 6 = 150 µF, capacitors C 1 = C 2
= C 3 = 0.47 µH are chosen for 50 kHz switching frequency. In order to operate the
circuit in DCVM, duty ratio of D = 55%.
The control algorithm for controlling the switches in converter 1 is, if V B1 > V B2 ,
then switch Q1 will be operated with 55% duty ratio whereas switch Q2 will be
ideal. Else, switch Q2 will be operated with 55% duty ratio whereas switch Q1 will
Active Battery-Balancing Using Ćuk Converter 215

Fig. 4 Stresses on circuit components

be ideal. Similarly, converter 2, if V B2 > V B3 , then switch Q3 will be operated with

55% duty ratio whereas switch Q4 will be ideal. Else, switch Q4 will be operated
with 55% duty ratio whereas switch Q3 will be ideal. Converter 3 is also operated
in the same algorithm. Simulation results are taken for the converter 1 with V B2 >
V B1 . Figure 4 shows the stresses on the components and Fig. 5 shows the simulation
results of battery balancing of entire pack while charging at 1C rate, discharging at
1C rate and at ideal condition respectively.

3.2 Hardware Results

A hardware prototype of Ćuk Converter based active battery balancing circuit was
implemented for balancing battery pack consisting of SAMSUNG ICR 18650-26f
Li-Ion cells in 4S configuration. Cell 1 is discharged to have relative lower SOC
compared to other cells. Control algorithm explained in above section is programmed
in LAUNCHXL F28069-M Microcontroller. The component details are given in
Table 1.
Switches are operated according to the control algorithm described in Sect. 3.1
at switching frequency of 50 kHz and duty ratio of 55%. An experiment is carried
out while charging the battery pack with 1C charging rate using the CCCV charging
method as recommended by the cell manufacturer.
Figure 6a shows the voltage across the energy transfer capacitor and Fig. 6b shows
the inductor current. Voltage V 1 , V 2, V 3 and V 4 in Fig. 7 are the cell voltages V B1 ,
V B2 , V B3, and V B4 respectively. The terminal voltage of cells while balancing is as
shown in Fig. 8. It can be observed from Figs. 7 and 9 that voltages of the cells after
balanced charging are equalized to voltage around 3.5 V. Final voltages are measured
after resting of 2–3 min.
216 N. Sarode and M. Divya

Fig. 5 a Balancing along with charging at 1C rate. b Balancing along with discharging at 1C rate.
c Balancing at ideal condition
Active Battery-Balancing Using Ćuk Converter 217

Table 1 Details of components

Component Part no. Make Rating
Li-Ion cells ICR 18650-26f Samsung SDI 2600 mAHr
MOSFETs NTP60N06 ON semiconductors V DS = 60 V, I D = 60 A
and RDS(ON) = 14 m
Inductors SPRH127-151M Shenzhen Codaca 150 µH
Electronic Co., Ltd.
Capacitors ECW-FD2J474K Panasonic 0.47 µF
Processor LAUNCHXL-F28069M Texas instruments NA

Fig. 6 a Voltage across energy transfer capacitor. b Inductor current

Fig. 7 Cell voltages before balancing

218 N. Sarode and M. Divya

Fig. 8 Cell terminal voltages of cell 1 and cell 2 during balancing

Fig. 9 Cell voltages after balancing

Active Battery-Balancing Using Ćuk Converter 219

4 Conclusion

A DC-DC converter based active cell balancing technique using Ćuk Converter is
studied in this paper. Both simulation, as well as hardware prototype for 4S battery
pack balancing, is implemented. The converter is operated in DCVM so that zero
voltage switching (ZVS) can be achieved to reduce the stresses on switches, even-
tually improving the efficiency of the circuit. Also, a high operating frequency of
50 kHz is used so as to reduce the size of passive components in the circuit which
helps to reduce the overall footprint of the circuit. It is observed from the simulation
and hardware results that battery balancing can be achieved using this topology.

References

1. Plett GL (2015) Battery management systems, volume I: battery modeling, 1st edn. Artech
House, Boston
2. Wıllıamson SS (2013) Energy management strategies for electric and plug-in hybrid electric
vehicles, 1st edn. Wylly
3. Wu C, Sun J, Zhu C, Ge Y, Zhao Y (2015) Research on overcharge and overdischarge effect
on Lithium-Ion batteries. In: 2015 IEEE vehicle power propulsion, pp 1–6
4. Cassani PA, Williamson SS (2009) Significance of battery cell equalization and monitoring
for practical commercialization of plug-in hybrid electric vehicles. In: IEEE applied power
electronics conference and exposition, pp 465–471
5. Hannan MA, Hoque MM, Hussain A, Yusof Y, Ker PJ (2018) State-of-the-art and energy
management system of Lithium-Ion batteries in electric vehicle applications: issues and
recommendations. IEEE Access 6:19362–19378
6. Hoque MM, Hannan MA, Mohamed A, Ayob A (2017) Battery charge equalization controller
in electric vehicle applications: a review. Renew Sustain Energy Rev 75:1363–1385
7. Williamson SS, Rathore AK, Musavi F (2015) Industrial electronics for electric transportation:
current state-of-the-art and future challenges. IEEE Trans Ind Electron 62(5):3021–3032
8. Poorali B, Adib E, Farzanehfard H (2017) Soft-switching DC-DC Ćuk converter operating in
discontinuous-capacitor-voltage mode. In: IET Power Electron 10(13):1679–1686
9. Maksimovic D, Ćuk S (1991) A unified analysis of PWM converters in discontinuous modes.
IEEE Trans Power Electron 6(3):476–490
10. Lee YS, Cheng MW, Yang SC, Hsu CL (2006) Individual cell equalization for series connected
Lithium-Ion batteries. IEICE Trans Commun E89-B(9):2596–2607
11. MathWorks, https://in.mathworks.com/
12. Samsung SDI, https://www.samsungsdi.com/
Analysis of Series-Connected PV Cells
Using gEDA and ngSPICE

Ritesh Kumar and Shelly Vadhera

Abstract The mathematical analysis of photovoltaic cell establishes a relationship

between the voltage and current of the cell through a coupled nonlinear equation,
which is difficult to solve analytically. This paper presents a method for simula-
tion of series-connected photovoltaic cells using open-source software (gEDA and
ngSPICE) which are able to evince the I–V characteristic and P–V characteristics of
photovoltaic module in case of identical cells, shading and shading with protection.

Keywords PV analysis · Shading effect · Open-source circuit simulator

1 Introduction

When photovoltaic (PV) modules are deployed, they are expected to perform for
up to 20 years. Environmental factors, temperature and incident irradiation, have
an affect on the performance of PV cell. Apart from these environmental factors,
shading also affects the performance of photovoltaic modules. This paper discusses
the characteristics of photovoltaic module in case of identical cells connected in series
and parallel along with the affect of shading on the series- and parallel-connected
modules. To study the affects, gEDA and ngSPICE software which are open-source
circuit simulator. ngSPICE is the result of combining existing SPICE features with
some extra analyses , modelling methods and device simulation features. It provides

R. Kumar (B)
School of Renewable Energy and Efficiency, National Institute of Technology Kurukshetra,
Kurukshetra, Haryana 136119, India
e-mail: [email protected]
S. Vadhera
Electrical Engineering Department, National Institute of Technology Kurukshetra, Kurukshetra,
Haryana 136119, India

© Springer Nature Singapore Pte Ltd. 2021 221

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_21
222 R. Kumar and S. Vadhera

Fig. 1 Equivalent circuit of

identical cell in series

a simulation engine, a command-line driven front-end and the capability to plot

simulation results graphically for gEDA distribution [1]. This paper highlights the
use of open software (ngSPICE, gEDA) for the analysis of PV cell.

2 Identical Cells in Series

2.1 Equivalent Circuit of Identical Cells in Series

When two identical cells are connected in series as shown in Fig. 1, the current
through each cell remains the same but the voltage across each cell adds up to give
the terminal voltage [2].

i T1 = i T2 = i T (1)

vT = vT 1 + vT 2 (2)

2.2 Circuit Used for Simulation

In Fig. 2, the gEDA circuit of identical cells in series used for simulation has been
shown.
Analysis of Series-Connected PV Cells Using gEDA and ngSPICE 223

Fig. 2 gEDA circuit of identical cells in series

2.3 Command in Linux

* simulation of series connection of pv cells

.tran 0.01ms 5ms

.include series.net
.control
set color0 = white ; set background as white
set color1 = black ; set foreground as
black run
plot i(vo)*10 vs nvti(vo)*10 vs nvt2 i(vo)*10 vs nvt-nvt2
plot i(vo)*10 vs nvti(vo)*nvt vs nvti(vo)*(nvt-nvt2) vs nvti(vo)*nvt2 vs nvt
.endc
224 R. Kumar and S. Vadhera

2.4 Plot

See Fig. 3.
PVc = I–V characteristics of series-connected PV cell 1 and PV cell 2.
V oc1 = V oc2 = Open-circuit voltage of PV cell 1 and PV cell 2.
V oc1 + V oc2 = V oc = Open-circuit voltage of combination of series-connected
identical PV cell 1 and PV cell 2.

See Fig. 4.
P1 = Power across PV cell 1.
P2 = Power across PV cell 2.
Pc = Power across combination of series-connected PV cell 1 and PV cell 2.

3 Shading Effect in Case of Series-Connected Modules

In practice, the PV cells or solar cells are never identical. Even the cells from same
batch of manufacturing will not have same I–V characteristics.
When PV modules are mounted and exposed to the solar radiation, there may
be neighbouring buildings and trees, the shadow of which may be falling on some

Fig. 3 I–V characteristics of identical cell in series

Analysis of Series-Connected PV Cells Using gEDA and ngSPICE 225

Fig. 4 P–V characteristics of identical cell in series

portion of the module or a part of the string of PV-cells; therefore, some of the PV
cells or modules will be under shade and some of them will be under bright sun light.
Hence, Isc of the PV cells, which are under bright sun light will be much higher
as compared to the PV cells which are under shade. In case of series connection, the
shadowed portion of module or some part of string operates in second quadrant with
negative voltage and forward current, whereas the illuminated portion of module of
string operates in first quadrant with positive voltage, that has been shown in Fig. 6,
therefore, some of the cell will act as source and some of them will act as sink, as
a consequence the cell which is sinking power becomes hot and deteriorates fast [3,
4]. The number of shadowed cells will hence limits the voltage that the shadowed
module or string can produce.

3.1 Circuit Diagram in gEDA to Simulate Shading Effect

in Series-Connected Cells

Solar cells having photon current I p23 = I p24 = I p25 = 0.7 A, which belongs to PV
module 2, are under shading condition, as shown in Fig. 5. Since these cells of PV
module 2 are partially shaded, therefore, it can be seen in the I–V characteristics of
PV module 2 (Fig. 6) that the voltage across PV module 2 is negative and during a
portion of I–V characteristics of combined cell, one of the PV modules (PV module
2) is showing negative power, indicating that PV module 2 is acting like sink (Fig. 7).
226 R. Kumar and S. Vadhera

Fig. 5 Circuit diagram in gEDA of series-connected non-identical cells

3.2 Command in Linux

* simulation of series connection of non identicalpv cells

.tran 0.01ms 5ms
.include seriesN.net
.control
set color0 = white ; set background as
white set color1 = black ; set foreground as
black
run
plot i(vo)*10 vs nvƟ(vo)*10 vs nvt2 i(vo)*10 vs nvt-nvt2

plot i(vo)*10 vs nvƟ(vo)*nvt vs nvƟ(vo)*(nvt-nvt2) vs nvƟ(vo)*nvt2 vs nvt

.endc
Analysis of Series-Connected PV Cells Using gEDA and ngSPICE 227

3.3 Plot

See Fig. 6.
PVc = I–V characteristics of series-connected PV module 1 and PV module 2.
PV1 = I–V characteristics of PV module 1.
PV2 = I–V characteristics of PV module 2 (Shaded Module) V oc1 = Open-circuit
voltage of PV module 1.
V oc2 = Open-circuit voltage of PV module 2.
V oc1 + V oc2 = V oc = Open-circuit voltage of combination series-connected PV
module 1 and PV module 2.

See Fig. 7.
P1 = Power across PV module 1.
P2 = Power across PV module 2.
Pc = Power across combination of series-connected PV module 1 and PV module
2.

Fig. 6 I–V characteristics of non-identical cells connected in series

228 R. Kumar and S. Vadhera

Fig. 7 P–V characteristics of non-identical cells connected in series

4 Series-Connected Cells with Protection in Case

of Shading

Due to shading, PV module 2 is behaving like sink; hence, it is dissipating power

and becoming hot therefore it may get damaged. This is avoided by connecting an
anti-parallel diode (Bypass Diode) across the modules [5].
The moment PV module 2 (shaded module) reverses polarity, i.e. goes from
sourcing to sinking, bypass diode gets forward biased and the voltage across it gets
clamped to the voltage equal to the forward voltage of the bypass diode as shown in
I–V characteristics of PV modules (Fig. 9) and bypasses PV module 2; hence, there
will not be power dissipation with in it and cell will not get hot and deteriorate. This
point onwards since PV module 2 is bypassed by the bypass diode therefore the I–V
characteristics of combination of PV modules connected in series will be same as
the I–V characteristics of PV module 1 as shown in Fig. 9.
The shaded module or string produces multiple maxima in the P–V characteristics
therefore the shading effect may severely limit the power produced because the MPPT
algorithms may tend to operate PV module or string on these multiple maxima, which
has been shown in Fig. 10.
Analysis of Series-Connected PV Cells Using gEDA and ngSPICE 229

4.1 Circuit Diagram in gEDA to Simulate Series-Connected

Cells with Protection in Case of Shading

See Fig. 8.

Fig. 8 Circuit diagram in gEDA of series-connected cells with protection in case of shading
230 R. Kumar and S. Vadhera

4.2 Command in Linux

* simulaƟon of series connecƟon of non idenƟcalpv cells

.tran 0.01ms 5ms

.include seriesNP.net
.control

set color0 = white ; set background as

white set color1 = black ; set foreground
as black
run

plot i(vo)*10 vs nvƟ(vo)*10 vs nvt2 i(vo)*10 vs nvt-nvt2

plot i(vo)*10 vs nvƟ(vo)*nvt vs nvƟ(vo)*(nvt-nvt2) vs nvƟ(vo)*nvt2 vs nvt

.endc

4.3 Plot

See Fig. 9.
PV1 = I–V Characteristics of PV module 1.
PV2 = I–V Characteristics of PV module 2.
PVc = I–V Characteristics of combination of series-connected PV module 1 and
PV module 2 with protection.

See Fig. 10.

PVc = I–V characteristics of combination of series-connected PV module 1 and
PV module 2 with protection.
P1 = Power across PV module 1.
P2 = Power across PV module 2.
Pc = Power across combination of series-connected PV module 1 and PV module
2 with protection.
Analysis of Series-Connected PV Cells Using gEDA and ngSPICE 231

Fig. 9 I–V characteristics of non-identical cells connected in series with protection

Fig. 10 P–V characteristics of non-identical cells connected in series with protection

232 R. Kumar and S. Vadhera

References

1. Tools for circuit simulation with ngSPICE (2019)

2. Al Tarabsheh A, Akmal M, Ghazal M (2017) Series connected photovoltaic cells—modelling
and analysis
3. Laamami S, Benhamed M, Sbita L (2017) Analysis of shading effects on a photovoltaic array
4. Patel H, Agarwal V (2008) MATLAB-based modeling to study the effects of partial shading on
PV array characteristics. IEEE Trans Energy Conver 23(1):302–310
5. Silvestre S, Boronat A, Chouder A (2009) Study of bypass diodes configuration on PV modules.
Appl Energy 86(9):1632–1640
An Experimental Study on a Portable
SPV-Integrated Forced Convective Solar
Dryer

Debajit Misra

Abstract In this work, a portable roof-ventilated solar dryer operated under 12 DC

fan, powered by 21 Wp solar photovoltaic (SPV) module and 12 V battery backup
has been developed to investigate drying performance of different kinds of agricul-
ture produces. An even span roof-type dryer has been constructed with a floor area
of 1 × 1 m2 and for air circulation, two continuous openings with 1 × 0.075 m2 of
each have also been provided at base. The dryer is made out of wooden structure,
clear polycarbonate sheets having 1.3 mm thickness, and blackened aluminium sheet
on its base to absorb the incidence of solar radiation. To investigate drying perfor-
mance, grapes and tomatoes have been taken for the different experimental runs. One
peak summer month has been chosen to identify thermal behaviour and drying char-
acteristics of grapes and tomatoes. The experimental observation depicts that after
26–28 h of drying time the grape fruits, whose initial moisture content was 85%
(wb), had reached to a moisture content of 15% (wb). However, tomatoes took more
drying time, about 32–36 h, starting with initial moisture content of 92% (wb) and
reached to 10% (wb). It was clearly found that drying zone temperature remained
38–59 °C whenever ambient temperature remained 30–41 °C, and solar radiation
intensity varied from 400 to 900 W/m2 during those days. Dried products have been
found to be of satisfactory quality level in terms of taste, colour and nutritious value.

Keywords Solar dryer · Moisture content · Drying time

1 Introduction

In India, a lot of agricultural products (fruits and vegetables) are wasted due to
unsafe harvesting, high relative humidity effect, sudden rainfall, etc. Again, very
little amount can be preserved and stored for off-seasonal consumption due to high
refrigeration cost. Open sun-drying technique is a traditional method by which some

D. Misra (B)
Department of Mechanical Engineering, Saroj Mohan Institute of Technology, Guptipara,
Hooghly 712512, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 233

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_22
234 D. Misra

kinds of fruits and vegetables can be preserved for off-seasonal consumption as

abundant solar energy is available in the most parts of the country. Though it is a very
simple and effective technique, it requires lots of labour cost. Besides this, more open
space is required for spreading the products. Again, it requires proper supervision
as sun-dried products are usually prone to overheating, cracking, contamination of
vermin, bacteria, dust, etc., rewetting during sudden rains, plundering by animals
and humans. All these lead to significant loss of product and impaired quality. Thus,
in order to avert the aforementioned difficulties, solar dryer is a viable option to get
desired quality produces. It can be considered as an effective sun drying system,
where solar heat is trapped to increase the inside air temperature of the dryer. As
dryer air temperature rises, it takes moisture from the wet products. Not only heat is
required to dry wet products, but proper air circulation is also an influential parameter
for effective drying. Sodha et al. [1] reported that the drying rate differs when the kind
of agriculture products differ and it depends on various external factors like solar
intensity level, ambient temperature, free wind velocity and relative humidity and
also on some internal factors like initial moisture content, products’ absorptivity,
mass of product per unit exposed area, etc. Sharma et al. [2] performed various
experimental analyses on three types of solar dryers; the first one was cabinet-type
natural convection dryer, the second one was multistack natural convection dryer,
and the third one was multishelf forced convection dryer. They reported that less
drying time was required in case of the forced convection dryer than the cabinet
or the multistacked natural convection dryer particularly on cloudy days. Finally,
they concluded that the cabinet-type dryer was suitable for small quality drying in
household scale, the multistack natural convection dryer was suitable for drying
a limited crop volume on frames, and the multishelf forced convection dryer was
suitable for large scale in industries. Barnwal and Tiwari [3] conducted experimental
study on a hybrid photovoltaic-thermal (PV/T) greenhouse dryer using Thompson
seedless grapes under forced ventilation mode. During experiments, they collected
hourly microclimatic data of the dryer to evaluate heat and mass transfer for the
proposed system. They evaluated the convective heat transfer coefficients for grapes
in greenhouse and also for open conditions. Janjai et al. [4] conducted an experimental
study on PV-ventilated solar greenhouse dryer. They considered banana and peeled
longan for their experiment and observed that the drying time of peeled longan in
solar greenhouse dryer was 3 days, whereas in natural sun drying, it required 5–
6 days under similar types of weather conditions. Sethi et al. [5] experimented on
a greenhouse solar dryer using reflection of inclined north wall. They saw that the
inside air and crop temperature were enhanced by 1 °C to 6.7 °C and 1 °C to 4 °C,
respectively, and the drying time was decreased by 16.67%.
Many experimental investigations and studies have been done for drying of agri-
cultural products by using solar dryers. A few number of studies have been done in the
tropics and subtropics regions which could be economical and technically effective
for household application. Technically effective implies that the solar dryer should
fulfil the following criteria: it should be simple in structure, should take minimum
power, should be highly effective, and should have long life. Again, in developing a
solar dryer, cost is a key influential factor when it is applied in rural or remote areas.
An Experimental Study on a Portable SPV-Integrated Forced … 235

In rural or remote household purpose, it is a primary necessity that the solar dryer
must be economically feasible. The present work thus concentrates on a low-cost
solar dryer that can be utilized in rural or remote household purposes. In this work,
a new design of solar dryer has been chosen which could be operated by SPV and
battery backup system. A portable-type solar dryer has been taken under considera-
tion which makes it beneficial as it could be relocated according to the availability
of the sun’s position.
The main aim of this work was to develop a small forced convective dryer utilizing
solar energy and fully run by renewable energy for household application. After
studying the literatures related to solar dryer, a new design has came into the mind,
which could be implemented for drying various kinds of fruits and vegetables. From
the literature, it was found that most of the authors worked with polyethylene sheet
and polycarbonate sheet for covering purpose of their dryers. In most of the cases,
transmissivity of a solar radiation of plastic lied in the ranges 60–65% [6–8], and
for polycarbonate sheet transmissivity lied in the range of 60–80% [4, 5, 9–11]. To
produce better results in drying, polycarbonate sheets have been chosen instead of
plastics. In present study, a clear glass-like polycarbonate sheet of 86% solar radiation
transmissivity has been taken to make the dryer cover. Continuous side openings on
two bottom walls of the greenhouse were made by which more wind can flow through
it. To make the dryer portable type, 1 m2 floor area has been selected. In this study,
the microclimate of the dryer and the drying characteristics of grapes (Sonaka) and
tomatoes (Plum) were examined during drying periods in a month.

2 Materials and Methods

Figure 1 shows the schematic diagram of the solar dryer. At the initial stage, a cubical
wooden frame having 1 m × 1 m × 1 m in volume has been constructed using 5-
cm (thickness and width) wooden block. Clear glass-like polycarbonate sheets with
1.3 mm thickness and 86% solar radiation transmittance were used as the glazing
material.
Those glazing sheets could filter UV radiation, which may degrade the colour
and nutritious value of some kinds of fruits and vegetables. The four sides of the
frame are covered with polycarbonate sheets. The base of the cubical box is covered
with a square ply-board providing a 4-mm-thick thermocol sheet on it for insulation
purpose. A square (1 m2 ) blackened aluminium sheet of one millimetre thickness is
firmly attached with the help of glue and nails on the thermocol sheet. This blackened
aluminium sheet acts as a heat absorber which receives heat when solar radiation is
coming from the sun. The thermocol sheet, beneath the absorber, prevents convective
heat transfer from the absorber towards the bottom. The central height of the dryer
was kept 1.282 m. Two roofs of the dryer were covered with polycarbonate sheets with
33° slope angle as maximum slope angle is about 8°–10° more than the latitude of
the location. Other two roofs were covered with isosceles trapezoidal polycarbonate
sheets of 33° base angle, 1 m base and 0.282 m height. The top position of the roof
236 D. Misra

Air m Air

Fan

 o

Rack 
m

  m

Rack 
m

Air Air
 m Absorber

Opening   m Opening

Fig. 1 Schematic diagram of the solar dryer

was flattened and covered with 1 m × 0.13 m polycarbonate sheet, and an exhaust
fan is mounted on its central position.
Two continuous openings were provided having 1 m × 0.075 m dimensions on
the bottom part of the two opposite side walls. Thus, outside air flows inside the
dryer through these openings/air gaps, gains heat from the absorber and ventilates
with the help of the exhaust fan. Two wooden racks were built inside the dryer to
support two perforated wire mesh trays with a base area of 0.9 m × 0.8 m. The first
rack was placed at 0.35 m height from the base, and the second rack was kept 0.35 m
apart from the first rack to ensure proper air circulation and distribution all around
the products. Two gaps, each of 0.1 m × 1 m, were provided on a side wall of the
dryer to insert or to remove trays during loading or unloading the products. The gaps
were covered with polyethylene sheets and velcro double-sided tapes.
When the dryer is exposed to sunlight, blackened absorber becomes hot. It heats
the surrounding air which is flowing from bottom to top through the perforated trays.
The solar dryer has been integrated with an SPV module of 21 Wp. The SPV module
generates DC electrical power to drive a DC exhaust fan of 0.114 m outer diameter
for force convection mode operation.
Various inside and ambient microclimatic parameters, which influenced the dryer
performance, were measured. Two-way K-type digital thermometer (BW-T326273)
An Experimental Study on a Portable SPV-Integrated Forced … 237

was used for measurement of dryer temperature; one sensor probe was placed at
central position of the dryer to measure the drying air temperature, and another sensor
probe was placed near the fan to measure the exit air temperature. The velocity of
wind was measured with an anemometer (HTC AVM06), and the exit velocity of the
dryer was measured using hot wire anemometer (Benetech GM8903). A solarimeter
(Waco 206) was used to measure the intensity of global solar radiation. The ambient
air temperature and relative humidity were measured using a digital psychrometer
(HTC HD304). Weight loss of the products in the drying times was measured using
an electronic balance.

3 Experimental Procedure

Sonaka seedless grapes and plum tomatoes were purchased from a local market. The
grapes and tomatoes were manually chosen during marketing to get almost uniform
shape and size. The grapes had average dimensions of 2.7 ± 0.2 cm in length, and the
tomatoes were of 4.2 ± 0.3 cm in diameter and 6.0 ± 0.3 cm in length, measured with
a Vernier calliper. At the initial stage, grapes and tomatoes were thoroughly rinsed
for removing dust and foreign materials. Then, tomatoes were cut longitudinally
into halves, which made more exposure and improved water diffusion, thus reducing
the drying time. Tomatoes were plain salt dipped as salt acts as antimicrobial. The
grapes were immersed in warm 4% citric acid bath for 5 min and then 5% potassium
metabisulphite (K2 S2 O5 ) bath around five minutes to remove wax content from the
peel. This reduced the drying time of the products as well as reduced the toxicity
without affecting the nutritious value. The dryer is of 15 kg capacity. As drying was
conducted in the dryer for experimentation only, 1 kg grapes and 1 kg tomatoes were
taken under consideration.
The study has been conducted in an engineering institute located at Hooghly
(23.2055° N, 88.4050° E), India. The dryer was placed on a concrete path, shown in
Fig. 2, so no shadow by trees or buildings appeared during experimental run from
9:00 am to 3:00 pm. Experiments were conducted in two phases in a hot summer
month of May in 2019; the first phase was 6–11 May, and the second phase was
16–21 May. The experiments were conducted on clear sunny days. During each
experimental phase, 1.0 kg of tomatoes was put in a tray and 1.0 kg of grapes was
put in the other tray. The inside and outside temperatures were recorded on every
hour during the drying time. The weight losses of the products were measured after
two hours for both of the products. The experiments were begun at 9:00 am and
ended at 3:00 pm as solar radiation is high enough in this period. The drying was
continued on subsequent days until the desired moisture content levels were reached
by wet basis.
238 D. Misra

Fig. 2 Experimental solar

dryer

4 Results and Discussions

The experiments were conducted on clear sunny periods of the days. Typical exper-
imental values found after the drying of grapes and tomatoes are presented in the
following section. Table 1 shows recorded climatic data upon which solar dryer had
been worked. May was the hottest month in Kolkata as well as Hooghly in 2019. In
the first phase, sometimes the ambient temperature crossed 40 °C and in the second
phase, sometimes the ambient temperature reached 38 °C. The relative humidity
remained within 52–60% in most of the times. During the experimentation, the wind
velocity was not steady; it was always fluctuating. In the table, data captured for both
the phase-I and phase-II are clearly shown.
Figure 3 shows intensities of global solar radiation during the experimental run
of solar drying of grapes and tomatoes in the dryer. In the six days of drying periods
of each of the phases, the intensity of solar radiation increased rapidly starting from
9 am to 2 pm and then decreased. It was observed that in very rare case, the intensity
of radiation fluctuated abruptly. In most of the cases, solar radiation was maximum
at noon and rarely it is changed due to sudden appearance of cloud.
Figure 4 shows the dryer air temperature for the two different phases of the exper-
iments. In the first phase, as the ambient temperatures and solar radiation intensi-
ties were higher than the phase-II, the dryer temperature was also higher. It is also
observed that temperatures were increasing with time up to 1 pm, and thereafter, up
to 3 pm, it slightly decreased. This kind of trend was almost uniform as the climatic
condition was almost the same.
Figure 5 shows the variation of moisture content for grapes and tomatoes with
time. The products had been placed inside the dryer for consecutive days until it
An Experimental Study on a Portable SPV-Integrated Forced … 239

Table 1 Climatic data in the drying periods

Time Phase-I Phase-II
Ta (°C) RH (ambient) Wind speed Ta (°C) RH (ambient) Wind speed
(m/s) (m/s)
Day-1
9 am 33.7 49.7 0.42 31.8 54.4 0.18
10 am 34.6 52.3 0.47 32.5 57.2 0.31
11 am 35.2 52.6 0.72 33.9 56.3 0.26
12 pm 36.7 53.4 0.61 34.8 57.5 0.34
1 pm 38.2 52.1 0.45 36.7 57.9 0.56
2 pm 38.5 56.9 0.35 36.6 58.1 0.42
3 pm 36.2 60.1 0.15 35.3 55.4 0.33
Day-2
9 am 30.2 52.9 0.21 31.8 56.1 0.23
10 am 34.3 51.6 0.45 32.4 56.7 0.16
11 am 37.9 53.7 0.58 33.4 55.4 0.09
12 pm 38.8 56.3 0.62 34.3 57.9 0.26
1 pm 41.2 54.5 0.73 35.7 57.3 0.43
2 pm 39.3 57.8 0.38 35.9 58.6 0.27
3 pm 36.6 58.1 0.17 34.1 57.4 0.19
Day-3
9 am 33.7 54.8 0.13 32.3 53.7 0.32
10 am 34.1 56.7 0.22 33.2 54.4 0.29
11 am 35.4 56.7 0.37 34.2 54.8 0.12
12 pm 37.9 57.4 0.21 36.5 56.9 0.37
1 pm 38.5 56.2 0.48 37.4 56.6 0.26
2 pm 38.8 58.4 0.15 37.8 57.1 0.08
3 pm 37.5 60.1 0.08 35.6 56.2 0.12
Day-4
9 am 32.4 55.3 0.46 31.8 55.5 0.13
10 am 33.5 55.6 0.48 32.7 55.7 0.35
11 am 34.3 57.4 0.23 33.5 56.3 0.21
12 pm 36.2 59.3 0.14 35.8 58.1 0.27
1 pm 36.8 56.7 0.33 35.6 56.2 0.46
2 pm 36.7 57.9 0.27 36.2 56.4 0.17
3 pm 35.2 56.8 0.14 35.6 55.7 0.32
Day-5
9 am 34.1 52.8 0.36 33.2 55.1 0.26
(continued)
240 D. Misra

Table 1 (continued)
Time Phase-I Phase-II
Ta (°C) RH (ambient) Wind speed Ta (°C) RH (ambient) Wind speed
(m/s) (m/s)
10 am 34.8 54.2 0.17 34.3 56.6 0.32
11 am 36.5 56.8 0.38 35.6 55.4 0.24
12 pm 36.9 56.2 0.44 35.9 56.2 0.33
1 pm 38.6 57.1 0.19 38.1 56.6 0.26
2 pm 38.1 58.4 0.07 37.2 57.5 0.22
3 pm 36.3 57.6 0.18 35.1 55.8 0.17
Day-6
9 am 33.5 52.8 0.37 32.8 53.5 0.18
10 am 35.3 54.2 0.44 33.7 55.4 0.31
11 am 37.9 56.8 0.62 34.9 54.3 0.22
12 pm 38.8 56.2 0.47 36.5 56.1 0.16
1 pm 40.3 57.1 0.25 38.8 56.3 0.18
2 pm 39.7 58.4 0.27 38.3 57.2 0.16
3 pm 37.6 57.6 0.36 35.7 55.4 0.15
Global solar radiation intensity (W/m2)

1000
Phase I
Phase II
900

800

700

600

500

400 Day  Day  Day  Day  Day  Day 

--

--

--

--

--
10am
12pm
2pm

10am
12pm
2pm

10am
12pm
2pm

10am
12pm
2pm

10am
12pm
2pm

10am
12pm
2pm

Time (hr)

Fig. 3 Intensity of global solar radiation with time

dried. The moisture content decreased continuously in progress of time. It is clearly

seen that grapes dried earlier than the tomatoes as more moisture had to be removed
from tomatoes. For the phase-I, due to the hottest climatic condition, grapes took
26 h, whereas tomatoes took 32 h for drying and for the phase-II, those times were
28 and 36 h for grapes and tomatoes, respectively.
During the experiments, grapes were dried to moisture content of 15% on wet
basis and tomatoes were dried within moisture content of 10% on wet basis. Rathore
An Experimental Study on a Portable SPV-Integrated Forced … 241

64
Phase I
Air temperature in the dryer ( C)
62 Phase II
o

60
58
56
54
52
50
48
46
44
42
40
38
Day Day Day Day Day Day
--

--

--

--

--
10am

12pm

2pm

10am

12pm
2pm

10am

12pm
2pm

10am

12pm

2pm

10am

12pm
2pm

10am

12pm

2pm
Time (hr)

Fig. 4 Dryer air temperature with time

1.0
(Grape) phase I
(Grape) phase II
0.8 (Tomato) phase I
(Tomato) phase II
Moisture content (wb)

0.6

0.4

0.2

0.0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36
Drying time (hrs)

Fig. 5 Products moisture content during drying time

and Panwar [10] in their solar tunnel dryer equipped with solar chimney found drying
air temperature within 55–70 °C for drying chemically untreated grapes. They carried
out their experiments from 9 am to 3 pm in a hot climatic zone (Udaipur, Rajasthan)
and reported that grapes took 6–7 days for raisin formation. Chemically pretreated
grapes require less time for drying due to the removal of wax content from the peels.
Thus, present solar dryer provides satisfactory performance as it took 26–28 h (about
5 days) to dry grapes when drying air temperature range was 38–59 °C.
Figure 6 presents grapes and tomato products before drying. This picture was taken
after treatment of the products. Figures 7 and 8 were taken after the completion of
the drying process.
242 D. Misra

Fig. 6 Products after predrying treatment

Fig. 7 Grapes after complete drying

5 Conclusion

In the present solar dryer, air temperature in the drying zone remained 38–59 °C
in a peak summer month which was suitable for drying agricultural products like
tomatoes and grapes as this dryer took reasonable drying time (26–36 h). The main
feature of the dryer is that it can be run by SPV and battery backup to provide uniform
ventilation. This kind of solar dryer might be used to dry other agricultural products
for household purpose as it is moderate in weight and portable type.
An Experimental Study on a Portable SPV-Integrated Forced … 243

Fig. 8 Tomatoes after complete drying

Acknowledgements The author would like to thank the authority of Saroj Mohan Institute of
Technology to provide support during the entire work. Special thanks go to the workshop staff and
some other members of the college for their active help.

References

1 Sodha MS, Dang A, Bansal PK, Sharma SB (1985) An analytical and experimental study of
open sun drying and a cabinet type dryer. Energy Convers Manage 25(3):263–271
2 Sharma VK, Colangelo A, Spagna G (1995) Experimental investigation of different solar dryers
suitable for fruit and vegetable drying. Renew Energy 6(4):413–424
3 Barnwal P, Tiwari GN (2008) Grape drying by using hybrid photovoltaic-thermal (PV/T)
greenhouse dryer: an experimental study. Sol Energy 82:1131–1144
4. Janjai SV, Khamvongsa Bala BK (2007) Development, design, and performance of a PV
ventilated greenhouse dryer. Int Energy J 8:249–258
5 Sethi VP, Arora S (2009) Improvement in greenhouse solar drying using inclined north wall
reflection. Sol Energy 83:1472–1484
6 Tiwari S, Tiwari GN, Al-Helal IM (2016) Performance analysis of photovoltaic–thermal (PVT)
mixed mode greenhouse solar dryer. Sol Energy 133:421–428
7 Kumar A, Prakash O, Kaviti A, Tomar A (2013) Experimental analysis of greenhouse dryer in
no-load conditions. J Environ Res Dev 7:1399–1406
8 Nayak S, Kumar A, Mishra J, Tiwari GN (2011) Drying and testing of mint (Mentha piperita)
by a hybrid photovoltaic-thermal (PVT)-based greenhouse dryer. Dry Technol 29:1002–1009
9. Nayak S, Tiwari GN (2008) Energy and exergy analysis of photovoltaic/thermal integrated
with a solar greenhouse. Energy Build 40:2015–2021
10. Rathore NS, Panwar NL (2010) Experimental studies on hemi cylindrical walk-in type solar
tunnel dryer for grape drying. Appl Energy 87:2764–2767
11. Ramos IN, Silva CLM, Sereno AM, Aguilera JM (2004) Quantification of microstructural
changes during first stage air drying of grape tissue. J Food Eng 62:159–164
244 D. Misra

12. Chauhan PS, Kumar A (2016) Performance analysis of greenhouse dryer by using insulated
north-wall under natural convection mode. Energy Rep 2:107–116
13 Ringeisen B, Barrett MD, Stroeve P (2014) Concentrated solar drying of tomatoes. Energy
Sustain Dev 19:47–55
14. Mariem SB, Mabrouk SB (2014) Drying characteristics of tomato slices and mathematical
modeling. Int J Energy Eng 4(2A):17–24
15. Misra D, Ghosh S (2018) Thermal modelling and performance assessment of a circular
greenhouse with solar chimney assisted ventilation and fog cooling. Agric Eng Int CIGR J
20(4):108–118
16. Misra D, Ghosh S (2018) Enhancement of natural ventilation of a circular greenhouse with
double wall solar chimney. IOP Conf Ser Mater Sci Eng 377(1):012205
17 https://agriexchange.apeda.gov.in
A Review on Energy-Efficient
and Sustainable Urban Buildings

Bishnu Kant Shukla, Harun Or Rashid, Anit Raj Bhowmik,

and Pushpendra Kumar Sharma

Abstract This study reviews works done on different types of energy-efficient

houses and experience of users which focuses on indoor climate as well as atti-
tude of users, their satisfaction level of such buildings and technological operations.
Energy-efficient buildings are generally rated much better than conventional build-
ings due to pleasant indoor climate. Based on results obtained from user assessments,
it was concluded that features of buildings vary. Discerned control of individuals and
sufficient information on use and operation is critical for on the whole encouraging
understanding of the building and further research in this area could be identified.
Previous information of environmental issues, social environment and the proce-
dure of moving into such building, largely control assessment of building. Specific
architectural solutions may cause influence on such energy competent buildings and
further works may be undertaken toward aesthetic and architectural features in evolu-
tion. Though there have been recent increase in research activities on operation and
use of energy-efficient buildings, there still remain gaps in different ways of providing
information and training in this area.

Keywords Energy-efficient building · Sustainability · Passive cooling · User

evaluation

B. K. Shukla (B) · H. O. Rashid · A. R. Bhowmik · P. K. Sharma

Lovely Professional University, Phagwara, India
e-mail: [email protected]
H. O. Rashid
e-mail: [email protected]
A. R. Bhowmik
e-mail: [email protected]
P. K. Sharma
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 245

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_23
246 B. K. Shukla et al.

1 Introduction

The need for sustainable development is ever increasing and humanity is at a point
where each man must learn to see beyond his animal instincts and selfish desires.
There are three masts of sustainable growth that includes economical development,
environmental attention and social insertion to be carried from corner to corner of
all sectors of growth, from farming to cities, energy development, infrastructure,
transport and water. At this crucial juncture where man must learn that his actions
have a consequence which is beyond his control, engineering comes as an increasing
threat (Fig. 1). Technology acts as a weapon which, in the hands of inept creatures
as man will turn into a force of destruction eliminating not only species of animals
but perhaps even man himself as well [1]. Thus, the time has come when not only
moral philosophers and preachers of good but engineers, men with the creative will
to either change or destroy the world to take the problem of morality into his own
hands and into engineering. It shall be the engineer who shall take the rest of mankind
into an age of enlightenment, an age where man learns to value consequence rather
than desire; it shall be men and women who teach the world of the growing problem
of climate change, or the depleting resources. It is the responsibility of human beings
even at individual level to focus more toward making environment a better place [2].
In the recent years, it has become a necessity to opt for the construction, which is
toward sustainability. More and more people are trending toward introducing basic
principles and laws and incorporate passive techniques of cooling and heating into
building even at domestic level. Existing research shows that there is a direct relation
between construction activity and impact on the environment which makes sustain-
ability a mandatory implementation [3]. Construction of zero energy buildings, green

16%,
34%

30%,

20%

Transportaon Residenal buildings

Industry Commercial buildings

Fig. 1 Energy consumed by different sectors

A Review on Energy-Efficient and Sustainable Urban Buildings 247

buildings, sustainable buildings, passive buildings etc. are having a global impact.
Rapidly industrialists including common people are switching toward this direction.
Not only these building give us an edge toward conserving environment but they are
also economic as compared to conventional building. Initial investment may show
higher investment up to 5–7% from a conventional buildings but the amount is easily
recoverable within 3–5 years after completion of building. Sustainable buildings and
constructions are more focused toward using renewable energy and that gives an
edge toward making the world a better and safer place for future generation [4].
Sustainable building and construction (SBC) is an all encircling and a multidis-
ciplinary approach. Such an approach is gradually being pushed for buildings and
construction. There is one more approach to convey this progress to regard as SBC
speaking to a “sustainable fabricated condition” encompassing the construction and
framework we create, the measures used to construct them and the various part-
ners included. Subsequently, “construction” is just piece of the sustainable building
process [5]. Evaluations of the time measurement that we use in the manufactured
condition—and based on it, in the vehicles–go from 80 to 90%. Other than the asset
and contamination issues encompassing the structure segment, guaranteeing that the
fabricated condition is refreshing and charming for people is starting to be seen as
an essential profitability issue. In the event that present examples do not change,
extension of the manufactured condition will pulverize or bother normal territories
and untamed life on more than 70% of the Earth’s territorial surface up to 2032,
driven predominantly by increments in populace, urbanization and financial action
[6].
The total population has dramatically increased since 1950. A large portion of this
development has occurred in the creating scene. In the following two decades, around
98% of total populace development will happen in creating countries. Seventy-five
percent of the general population in created nations officially lives in urban settle-
ments [7]. In creating nations, the offer of the populace living in urban areas is relied
upon to achieve 40% preceding the finish of this decade, contrasted and under 20%
out of 1950. Somewhere in the range of 60% of the world’s quickest developing
bigger urban areas (750,000 or more) are in low-wage nations. Regarding farmland
alone, urbanization guarantees as much as 40,000 km2 every year [8].

2 Passive Solar Design—An Approach Toward

Sustainability

The primary favorable circumstances of a latent sun powered home incorporate lower
warming and cooling costs and additionally the possibility to produce your own
particular sun-oriented vitality to supply parts or even a large portion of your family
unit’s vitality needs. On the off chance that you are thinking about a home that fuses
uninvolved sun-based outline, it is critical to work with modelers and contractual
workers who are knowledgeable about this sort of a development venture [9].
248 B. K. Shukla et al.

An effective uninvolved sun-oriented home needs an unhindered perspective of

the sun on its south-bound exterior. Consider whether adjacent development activ-
ities or trees could hinder your home’s introduction to daylight. A uninvolved sun-
based outline works by having windows arranged toward the south with the goal
that daylight can be saddled. Sunlight-based boards can be joined to the rooftop to
catch the sun amid the daytime. There are many significant detached sunlight-based
outline components to be considered, including protection, air fixing and warm mass
[10]. The building’s warm mass ought to be developed to assimilate warm amid the
day and keep that warmth inside the development materials. Uninvolved sun-based
homes require specific conditions with respect to their area and shading; moreover,
they require the coating of windows and in addition helper warming and aerating
and cooling frameworks that give a reinforcement wellspring of indoor solace on
overcast days [11].
One technique for aloof sun-oriented home plan is with coordinate increases.
The southern windows get daylight that goes through the windows to cement or
block deck and dividers. The stone work ingests the warmth and stores it. During the
evening, the warm mass transmits the warmth into the house [12].

3 Impacts of Building and Construction Sector

Both the current constructed condition and the way toward adding to it have various
natural and social effects. While most accessible measurements identified with these
effects are for created nations, specialists accept in general that these effects are more
terrible in creating than in created nations. The building up world’s offer of world
development exercises was 10% out of 1965, 29% out of 1998 and as yet developing
[13].
The general monetary commitments of the development area are impressive. Its
overall market volume adds up to over US dolor 3 trillion and witness for up to
10% of GDP of the world, contingent upon how the part is characterized. Develop-
ment is the biggest mechanical segment in Europe (up to 11% GDP) and in the US
(up to 12%). In creating scene, it speaks to 2–3% of GDP. Development likewise
represents more than half of national capital interest in many nations. It is approxi-
mately 7% of world business (nearly 28% of mechanical business) which constitute
a workforce of nearly 111 million, out of which; nearly 74% are in low-wage nations.
Creating nations represent 23% of worldwide development action—ultimately; the
development business is more work serious in poorer nations [14].
In many nations, the building and construction segment is the biggest single boss.
It is presumably the world’s biggest industrial business. Its exercises involve a high
multiplier impact: the International Council for Research and Innovation in Building
and Construction (CIB) gauges that a dollar spent on construction may create up to
three dollars of monetary movement in different areas [15].
A Review on Energy-Efficient and Sustainable Urban Buildings 249

4 Need of Energy-Efficient Buildings

Structures represent a critical extent of the aggregate vitality and carbon outflows
around the world, and assume a vital part in figuring manageable advancement tech-
niques. There is a developing enthusiasm for zero energy buildings (ZEBs) as of late.
A few nations have embraced or considering setting up ZEBs as their future building
vitality focuses to help reduce the issues concerning the exhaustion of vitality assets
and the crumbling of the earth. Extensively, ZEBs include two outline techniques—
limiting the requirement for vitality use in structures (particularly to heat and cooling)
through energy-efficient measures (EEMs) and receiving renewable energy and other
technologies (RETs) to meet the rest of the vitality needs [16].
The introduction of sustainability into the building sectors will also benefit
the contractors as will enhance their sustainability performance. This can only be
achieved by proper implementation of sustainable construction framework. Due to
increment in cost, certain long-lasting tasks and resource utilization to implement
environment affable construction practices, the contractors get demotivated from
appealing into the activities in humanizing the sustainability performance of the
buildings. Their primary focus is only economic loss [17]. Although, there are some
sectors that employ the environmental regulation system in the building sector but
this approach is still suffering from a lot of turbulences. Some studies suggest that
construction firms should be forced that they impose government guideline (legisla-
tion) and establish appropriate EMS standards. These strategies are well thoughtout
as peripheral pushing forces for service providers/builders to apply sustainable
construction practice without much consideration of its correct worth. On the other
hand, improved efficiency will be achieved if the implementation of sustainable
practice is accepted as inner motivating forces for included competitiveness policy
(Fig. 2).

5 Conclusion

Vitality productivity is the rush with no limits. The humankind is quickly moving
in the direction of vitality orderliness. During the intervening time, civilization is
trying to re-build up the connection it previously had with nature. An energy effec-
tive habitat is an advance of individual in the direction of sustainable source of
power, natural insurance and manageable livelihood. Possessions of such a habitat
allow people to decrease their expenditure and give an excellent project. Moreover,
energy effectiveness involves more beneficial and more contented with living that is
compliant with nature.
Building or moving up to a vitality productive home requires an underlying venture
that is higher than the cost of a customarily developed home. Be that as it may, there
are government gifts and motivators that can kick you off and balance a portion of
250 B. K. Shukla et al.

Fig. 2 Framework for sustainable construction practices

the cost. After you live in your vitality productive house for a couple of years, your
forthright venture will pay for itself.
The task life cycle must be assessed agreeing to numerous criteria considering the
general viewpoints of development effect on condition. The best procedure could be
chosen from accessible situations and data. In key choices, managing vulnerability,
the estimations of criteria could be resolved in interims– from skeptical incentive to
idealistic esteem.

References

1. Bakhtiar KAB, Shen LY, Misnan SHB (2008) A framework for comparison study on the major
methods in promoting sustainable construction practice. J Alam Bina 12(3):55–69
2. Toms S (2000) Environmental management, environmental accounting and financial perfor-
mance. CIMA, London
3. Tan Y, Shen L, Yao H (2011) Sustainable construction practice and contractors’ competitive-
ness: a preliminary study. Habitat Int 35:225–230
4. Lee KH, Ball R (2003) Achieving sustainable corporate competitiveness: strategic link between
top management’s (green) commitment and corporate environmental strategy. Greener Manage
Int 44(Winter):89–104
5. Bentivegna V, Curwell S, Deakin M, Lombardi P, Mitchell G, Nijkamp P (2002) A vision
and methodology for integrated sustainable urban development: BEQUEST. Build Res Inf
30(2):83–94
A Review on Energy-Efficient and Sustainable Urban Buildings 251

6. Ding GKC (2005) Developing a multicriteria approach for the measurement of sustainable
performance. Build Res Inf 33(1):3–16
7. Ding GKC (2008) Sustainable construction-the role of environmental assessment tools. J
Environ Manage 86:451–464
8. Rohracher H (2001) Managing the technological transition to sustainable construction of
buildings: a socio-technical perspective. Technol Anal Strat Manag 13(1):137–150
9. Zimmermann M, Althaus HJ, Haas A (2005) Benchmarks for sustainable construction-a
contribution to develop a standard. Energy Build
10. Medineckiene M, Turskis, Zavadskas EK (2011) Sustainable construction taking into account
the building impact on the environment. J Environ Eng Landsc Manage 18(2):118–127
11. Castro-Lacouture D, SefairJA, Flórez L, Medaglia AL (2009) Optimization model for the
selection of materials using a LEED-based green building rating system in Colombia. Build
Environ 44(6):1162–1170
12. Lombera JTSJ, Aprea IG (2010) A system approach to the environmental analysis of industrial
buildings. Build Environ 45(3):673–683
13. Zavadskas EK, Ustinovičius L, Turskis Z, Peldchus F, Messing D (2002) LEVI 3.0—multiple
criteria evaluation program for construction solutions. J Civil Eng Manag 8(2):184–191
14. Hill RC, Bowen PA (1997) Sustainable construction: principles and a framework for attainment.
Constr Manage Econ 15(3):223–239
15. Wyatt DP, Gilleard JG (1994) Deconstruction: an environmental response for construction
sustainability. In: Proceedings of first international conference of CIB TG 16 on sustainable
construction, Tampa, Florida, 6–9 Nov, pp 113–22
16. Yap NT (1989) Sustainable community development: an introductory guide. Ontario Environ-
mental Network, Canada
17. Wyatt DP (1994) Service life profiling an approach to designing for sustainable construction.
In: W88 proceedings architectural management. Notts, pp 1–10
Role of Optimization Algorithms
in Enhancing the Performance
of Photovoltaic Thermal (PV/T) Systems

Sumit Kumar, Sourav Diwania, Anwar S. Siddiqui, and Sanjay Agrawal

Abstract For reducing the gap between demand and supply of electricity generation,
many sustainable energy sources have been utilized to fulfill the expanding energy
demand. The researchers and industries have key concentration on the innovative
expansion and improvement on power generation and their efficiencies. Photovoltaic
thermal module system application of renewable energy resource is effectively used
worldwide because of easy availability of sun energy. Sun energy is consumed by
PV/T system to generate the electricity and useful thermal (heat) energy. The PV
module approximately gives electrical efficiency in scope of 7–12%, the remainder
of energy being dispersed in the form of heat losses. An endeavor has been made
to display and optimize the parameter of hybrid single-channel PV/T-module like
length of the channel, depth of the channel, velocity of the fluid, and so forth. There
are numerous other such type of parameters were used which affect the efficiency of
PV/T system. Aim of this paper is to improve the exergy efficiency by optimizing the
design parameter with the help of gravitational search algorithm (GSA) and compare
the results with genetic algorithm (GA) approach. By comparing the other methods,
we found that GSA extremely successful technique to optimize the design parameter
of hybrid single-channel PV/T module.

Keywords GSA · Exergy · Energy · Photovoltaic thermal module

1 Introduction

In the cutting edge world, the demand of electricity generation expands step by
step as the populace is expanding. The fulfillment of desired electricity demand is
basic requirement of any developing country but the installed capacity is fixed. To

S. Kumar · S. Diwania (B) · A. S. Siddiqui

Department of Electrical Engineering, JMI, New Delhi, New Delhi 110025, India
e-mail: [email protected]
S. Agrawal
SOET, IGNOU, New Delhi 110068, India

© Springer Nature Singapore Pte Ltd. 2021 253

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_24
254 S. Kumar et al.

reduce the gap between demand and supply of electricity, many non-conventional
energy resources have been emerged and extensively used these days. As conven-
tional energy sources are unable to meet the growing demand and also the fossil
fuel like coal, petroleum, and natural gas will be depleted in few hundred years. The
emerging technology of hybrid photovoltaic thermal module (PV/T) is mostly used
application of solar photovoltaic system in which wasted heat energy is used for
various applications and helps in improving the efficiency as shown in Fig. 1. New
progression and enhancement for power efficiency of PV system have a noteworthy
center theme by many research network and ventures (Table 1).
Zhang et al. [1] presumed that the utilization of sustainable power source came into
existence because the fuel used in conventional power plants is depleting; therefore,
in contrast to demand, the majority of the utilization is sustainable power source
nowadays. The PV/T system comes into existence with a plan to use the thermal
energy along with electrical energy. Singh et al. [2] have done investigations on
various kind of PV/T air collector and inferred that the overall annul thermal energy,
exergy gain, and ïEx of unglazed hybrid PV/T was improved by 32%, 55.9%, and
53%, respectively, over traditional air collector. Various kinds of PV modules and
their applications are discussed by Tiwari et al. [3] and they clarify and discuss the
modeling in detail. Singh et al. [4] have been done experimental work on PV/T
module; they have given design and simulation of smart control MPPT methods
for PV module using MATLAB. Bunches of hypothetical and exploratory work
have been done been done on hybrid PV/T system and shown in literature. The
performance of polycrystalline PV/T flat plate collector has been investigated by
Amori et at. [5]. From the experimental analysis, it has been observed that at a constant
air velocity of 0.0991 kg/s, an average reduction of 15.52 °C in cell temperature is
observed with single pass air channel as compared to conventional PV. Slimani et al.
[6] presented the comparative analysis of four different types of PV models. The
selected configurations are photovoltaic module (PV-I), hybrid PV/T air collector
(PV/T-II), glazed hybrid PV/T air collector (PV/T-III), and glazed hybrid double
pass PV/T air collector (PV/T-IV). The experimental investigation was conducted in
Algiers city.
The experimental model consisted of mono-crystalline silicon-based PV module,
an air duct, protective glass layer, and tedlar below PV module. The prototype of
PV/T tile was designed by Agrawal et al. [7] which consists of a channel beneath the
cell to eliminate the heat from the cell. The author did an experimental investigation in

Glass layer
Solar cell

Tedlar

Air In Air Out

Insulaon

Fig. 1 Hybrid PV/T air collector

Role of Optimization Algorithms in Enhancing the Performance … 255

Table 1 Nomenclature used

Symbol Description (Unit)
in the paper
ïele Electrical efficiency
ï0 Efficiency at STC
β0 Temperature coefficients of efficiency (K−1 )
Tc Temperature of solar cell (K)
T0 Ambient temperature (K)
ïT Total efficiency
ïth Thermal efficiency
Qth Usable heat (W)
Gt Incident solar radiation (W/m2)
Ac Aperture area (m2 )
ṁ Mass flow rate of fluid (kg/s)
Cp Specific heat of fluid (J/kg K)
T in Input temperature (K)
Tout Output temperature (K)
Im Current at MPP (A)
Vm Voltage at the MPP (V)
ηE x Exergy efficiency
Am Area of PV module (m2 )
TSUN Sun temperature (K)
L Duct length used in the paper
d Depth of the duct
V Fluid velocity used

which the output of one tile is connected to the input of other tile and the observations
were taken for both the tiles.
It has been observed that the electrical efficiency (ïele ) reduces significantly when
the temperature of solar cell increases. By using the channel below the cell, the heat
energy associated with the cell is extracted which reduces the temperature of the solar
cell and hence increases its electrical efficiency. Further, this thermal energy can also
be utilized for domestic applications. This unique concept of PV/T air collector
application has been introduced in 1973 and tested at solar test house, University of
Delaware [8].

2 System Description

The PV/T module consists of a channel or duct between PV solar cell and tedlar. The
schematic view of PV/T module is shown in Fig. 2.
256 S. Kumar et al.

Air out

Solar cell
Tedlar

Air in Insulaon

Micro-channel

Fig. 2 Photovoltaic/thermal tile

In the channel, an air or any other fluid is circulated to absorb the heat from the cell
by convection phenomenon. In order to improve the ratings of this module, many tiles
are connected in series and in parallel. This type connection is called a single-channel
photovoltaic thermal (SCPV/T) module. In order to maximum exergy efficiency,
different parameters of SCPV/T module have been optimized using different opti-
mizations algorithms like GSA and the results are compared with GA. In PV/T appli-
cation, solar thermal part is employed for collecting the sun’s energy and converting
it to heat energy for application of like water heating, air heating, etc., while solar
photovoltaic application is based on utilization of solar energy and converted it into
electrical energy.

3 Modeling of PV/T Air-Collector

While driving the energy equation for PV/T—system, the following assumptions are
kept in mind:
1. Temperature gradient beside the width of solar cell is neglected.
2. Heat capacity of solar cell is neglected.
3. Air specific heat remains invariable.
4. The system is assumed to be in quasi-steady state.
5. Unity packing factor is considered for the course of analysis.
Role of Optimization Algorithms in Enhancing the Performance … 257

The SCPV/T module is shown in Fig. 2. Thermal area of SCPV/T is bdx. Following
Tiwari and Sodha [3], the energy balance equation for photovoltaic/thermal tile can
be written as:

[Rate of sun energy existing on solar cell]

= [Rate of heat loss from upper surface of solar cell to ambient]
+ [Rate of heat transport from solar cell to flowing fluid i.e. air]
+ [Rate of electrical energy formed ]

We know that
 
αc In∗ bdx = [h SCA (TSC − TA )bdx] + [h SCF (TSC − T F)bdx] + [ηT C In ∗ bdx]
(1)

where αeff = αSc − ηTC .

When the air flows into the channel, the energy equation for elemental area bdx
cab be derived
dTF
HSCFT (TSC − TF )bM dx = m F Cair dx + HFA (TF − Tamb )bM dx (2)
dx
where the mass flow rate can be calculated as:

m F = ρ × L M × d × VF

By equating Eqs. 1 and 2, we obtained

αeff In + HSCAG Tamb + HSCFT TF

TSC = (3)
HSCAG + HSCFT

On solving Eqs. 1–3, we will obtain an expression for output temperature, thermal
efficiency, and electrical efficiency
   Nb L H   Nb L H 
h p αeff M M L
− m C M
M M L
− m C M
TFON = In + Tamb 1−e F air + TFI e F air (4)
HLM

The thermal efficiency is calculated as shown in Eq. 5

   
m F Cair −NC bM L M HLM (TFI − Tamb )
ηth = 1 − exp h p αeff − HLM
HLM NC ASC m F Cair In
(5)

The expression for electrical efficiency was calculated according to Agrawal and
TiwariI [9].
258 S. Kumar et al.


αeff In HSCFT αe f f In h p
ηele = η0 1 − βo − (TFO − Tamb ) +
HSCAG + HSCFT HLM (HSCAG + HSCFT )
    
NC bM L M HLM HSCFT NC bM L M HLM
1 − exp −
m F Cair
+
HSCAG + HSCFT
1 − exp −
m F Cair
( Tamb − TFI )} (6)

The ïEx is the sum of thermal exergy and exergy electrical as shown in Eq. (7)
  
Exout = Exth + Exele (7)

while the exergy thermal and exergy electrical are calculated as:

  
Tamb + 273
Exth = Q U,N 1 − (8)
TFO + 273
  
η A m Im
and ExEL = (9)
1000

The input exergy is calculated as:

 
 4 Tamb Tamb 4
ExIN = ASC Nc ∗ In 1 − + 1/3 (10)
3 Tsun Tsun

ExOUT
ηEx = × 100 (11)
ExIN

4 Gravitational Search Algorithm (GSA)

GSA is the one of the most sought physics-based meta-heuristic optimization algo-
rithm used to solve number of practical optimization problems of daily life. GSA
proposed by Rashedi [10] is most suitable for particularly solving nonlinear prob-
lems. GSA is based on Newton’s Gravitational Law which states that every object
attracts another object with a force called gravitational force which is directly propor-
tional to their masses and conversely proportional to the square of distance between
their centers of masses. In the proposed communication, particle is considered as
an object and their mass is considered as their performance criteria. Each particle
in GSA is characterized by four parameters viz. position of particle, inertial mass,
active gravitational mass, and passive gravitational mass. The solution is a function
of particle position while gravitational and inertial masses are calculated with the
help of fitness function.
Role of Optimization Algorithms in Enhancing the Performance … 259

Exploration and exploitation are the two axioms of every population-based algo-
rithm. GSA uses exploration at start to obviate local maxima and then uses exploita-
tion. The other particles are attracted by the use of time function, namely K best . Explo-
ration and exploitation control the performance of GSA. The K best values associated
with all particles linearly decays with time and finally only particle with highest
K best value (highest mass) survives and represents the final solution. The flowchart
explaining the working of gravitational search algorithm has been shown in Fig. 3.

5 Results and Discussion

The proposed work involves the optimization of four different PV/T- system param-
eters which will have an impact over the PV/T—system’s performance. While opti-
mizing these four parameters, the other design parameters kept fixed. A gravitational
search algorithm is useful for proposed problem to determine the optimize values
of the said parameter by maximizing the overall ïEx . Initially, the parameters were
identified on which the ïEx of the PV/T module depends. Then, the upper and lower
bound limits of the parameters are set. After that, the objective function for the
problem is defined. The flowchart for the entire process is depicted in Fig. 3.
The analysis has been done on intensity and ambient temperature data of New
Delhi, India, taken from IMD, Pune, for a particular day. The intensity and ambient
temperature vary with time as depicted in Fig. 4. It has been observed from the input
data that both intensity and ambient temperature changes with change in time.
The exergy efficiency (ïEx ) is calculated according to Eq. 11 and it has been
found that the maximum overall exergy efficiency of the system is 14.633% which
is obtained at 11 A.M when the intensity of sunlight is 680.73 W/m2 . The optimized
value of variable factors is shown in Table 2. When the ïEx is maximum.
The convergence curve showing that the rate of convergence of GSA is much better
than GA and it takes lesser time in indentifying the optimized value of parameters.
Figure 5 shows the deviation in ïEx at each time instant for the day. It has been
observed that the overall ïEx of PV/T module using GSA algorithm is far better than
the ïEx obtained by using GA algorithms.
The % ïEx at each time instant and the optimized value of four selected variable
factors are shown in Table 3.
It has been observed from the above bar graph that GSA having higher overall
exergy efficiency as compared to optimization done by the genetic algorithm. The
optimized design parameters are given in Table 3. It is clear from the table shown
below that the design parameters are approximately equal. So it is better to take the
average value of the parameter for best results (Fig. 6).
260 S. Kumar et al.

Fig. 3 Flowchart for GSA

5.1 Conclusion

In this communication, four of the variable parameters which affects the effectiveness
of PV/T air collector has been optimized using different optimization algorithms and
the results are compared. It has been observed intensity of sunlight is the major factor
Role of Optimization Algorithms in Enhancing the Performance … 261

Fig. 4 Solar radiation and amb. temperature data

Table 2 Optimized value of parameters

S. No. Optimized factor Optimized value
1 Duct length (L) 0.3000 m
2 Width of duct (d) 0.00100 m
3 Fluid (air) velcoity (V) 1.45559 m/s
4 Input temperature (T f1 ) 4.86086 °C

Table 3. Optimized value of design factors at each time instant

Depth (d) Length (L) Fluid (V ) Temp (T fi ) % ïEx Time
0.00100 0.30000 0.25943 1.49816 4.92799 8:00
0.00100 0.30000 0.29444 1.50000 10.88154 9:00
0.00100 0.30000 0.29748 1.48754 16.23553 10:00
0.00100 0.30000 0.24347 1.34811 19.35030 11:00
0.00100 0.30000 0.28381 1.40254 19.55750 12:00
0.00100 0.30000 0.30000 1.50000 18.40786 13:00
0.00100 0.30000 0.23306 1.43365 18.98991 14:00
0.00100 0.30000 0.26274 1.11486 13.85664 15:00
0.00100 0.30000 0.28373 1.37271 9.97839 16:00
0.00100 0.30000 0.26377 1.16563 6.94043 17:00

which affects the ïEx of PV/T system. The analysis by using GSA and GA shows
that the ïEx of PV/T system goes on increasing with increase in intensity of sunlight.
The ïEx achieved from the system is 14.663% which is much higher when the design
factors of PV/T were optimized using GA and other optimization technique. It has
262 S. Kumar et al.

Fig. 5 Exergy efficiency comparison using GSA and GA

Fig. 6 Variation in exergy efficiency

Role of Optimization Algorithms in Enhancing the Performance … 263

been concluded that the pace of iteration convergence using GSA is much faster than
GA.

References

1. Zhang X, Smith S, Zhao X, Xu J, Yu X (2012) Review of R&D progress and practical application
of the solar (PV/T) technologies. Renew Sustain Energy Rev 16:599–617
2. Singh GK, Agrawal S, Tiwari GN (2012) Analysis of different types of hybrid photovoltaic
thermal air collectors: a comparative study. J Fundamentals Renew Energy Appl. https://doi.
org/10.4303/jfrea/R120305
3. Tiwari A, Sodha MS (2007) Parametric study of various configurations of hybrid PV/T Thermal
air collector: experimental validation of theoretical model. Sol Energy Mater Sol Cells 91:17–28
4. Singh S, Swati Mathew L, Shimi SL (2013) Design and simulation of intelligent control MPPT
technique for PV module using MATLAB/SIMSCAPE. Int J Adv Res Electr Electron Instrum
Eng
5. Amori KE, Abd-Al Raheem MA (2011) Field study of various air based photovoltaic/thermal
hybrid solar collectors. Renew Energy 63:402–414
6. Slimani MEA, Amirat M, Kurucz I, Bahria S, Hamidat A, Chouch WB (2017) A detailed
thermal-electrical model of three photovoltaic/thermal (PV/T) hybrid air collectors and photo-
voltaic module: comparative study under Algiers climate condition. Energy Convers Manage
133:458–476
7. Agarwal S, Tiwari GN, Pandey HD (2012) Indoor experimental analysis of glazed hybrid PVT
tiles air collector connected in series. Energy Build 53:145–151
8. Garg HP (1980) Treatise on solar energy. Wiley , Chicester, p 608
9. Agrawal S, Tiwari A (2009) Experimental validation of glazed hybrid micro-channel solar cell
thermal tile. Sol Energy 85:3046–3056
10. Rashedi E, Nezamabad H, Saryazdi S (2009) Gravit Search Alg Inf Sci 179:2232–2248
Selection of the Best Material for Coating
on Solar Cell and Optical Filter

Sanjay Kumar and Lillie Dewan

Abstract The perfect selection of materials plays a vital role in the decision-making
task process of the manufacturing industry. The selection of the material does affect
many aspects of product and manufacturing process too. There are various range and
different criteria affecting the determination of material for a particular application.
These criteria go from electrical, optical, and physical properties to economic delib-
eration of the materials. The issue of choosing a material for a design application
from numerous substitutes based on a few criteria can be treated as a multicrite-
rion decision-making task. In this exploration work, emphasis is largely on choice
for the best material for coating on solar cell and optical channel (filter) applica-
tions. By utilizing “Technique for Order Preference by Similarity to Ideal Solution”
and “Simple Additive Weighting” method, the correlation and best ranking of the
materials have been observed.

Keywords TOPSIS and SAW method · Material types · Refractive index · Melting
point · Transmission

1 Introduction

The selection of materials [1, 2] for specific applications is a tedious task, as it affects
the cost and the quality of the item. Selection of materials is a non-stop strategy,
coordinating to pick the best material for an offered application. The material choice
is made in the beginning phase of the item life cycle, i.e., when the module is first
planned or when it is overhauled. An off-base picked material can lead not exclusively
to an early disappointment of the segment, but also to a pointless expense. Over the
most recent couple of decades, numerous new materials and material sorts have been

S. Kumar (B) · L. Dewan

National Institute of Technology Kurukshetra, Kurukshetra, Haryana 136119, India
e-mail: [email protected]
L. Dewan
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 265

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_25
266 S. Kumar and L. Dewan

created like earthenware production, plastics, ceramics, composites, and so on. This
offers a chance to the designer for advancement by using those materials in items
that give improved execution at a lower cost.
In this way, a designer faces a lot of problems in materials selection to make a
proper design [3] among the various materials considering that the selected materials
are best in terms of technology, economic, and environment. The primary assignment
of materials choice is to perceive the best appropriate criteria among different choices
to get a proper design under certain circumstances. In this manner, while choosing
the best materials for specific requirement, the creator needs to consider different
quantitative and qualitative criteria. Choosing the material for a industrial application
from among at least two substitutes based on various conflicting criteria can be
considered as a multicriteria decision-making task [4] issue.
There are many alternatives to choose the material for an optical channel appli-
cations and coating on solar cell. After carrying the literature survey, the number of
alternatives that can be used in designing optical filter for required transmission in
prescribed design wavelength has been observed. In this paper, main focussing is on
main dielectric materials like as aluminum oxide (Al2 O3 ), Silicon dioxide (SiO2 ),
hafnium oxide (HfO2 ), niobium oxide II (Nb2 O5 ), tantalum oxide (Ta2 O5 ), titanium
(IV) oxide (TiO2 ), zirconium dioxide (ZrO2 ), barium fluoride (BaF2 ), magnesium
fluoride (MgF2 ), and yttrium fluoride (YF3 ) that are mainly used for the optical
coating. Optical coating consists of thin-film layers of transparent dielectric mate-
rials, and such type of material numbers of layer is deposited on a substrate (glass) by
the ion excitation, thermal evaporation, electron beam vacuum coating equipment,
and the performance of these coating was checked by the using various sophisti-
cated spectroscopic technique such as UV–Vis-NIR spectrometer, Gonio spectropho-
tometer etc. [5–7]. For farther studied, if we coat the dielectric material on the surface
of the bare glass, then coated glass transmits and reflects more light as compared
to the bare glass, that results in the increase in the efficiency of transmittance and
reflectance of the light in the solar cell and optical filters. These types of optical
coating find applications in various optical systems like display systems, optical
instrumentation, avionics, military, defense sector, solar cell, and different optical
channel applications.
Due to changes in the transmitting and reflective properties of glass after the
coating of the dielectric materials and increasing the efficiency of components, the
main focus of this paper is to select the best material for coating on glass.
Selection and design of material can be resolved by the multiple decision making
[8]. There are number of methods SAW, WPM, AHP, FAHP, and TOPSIS [9] method
are possible decision-making criterion, two method has been discussed.

2 Research Methodology

From the possible different decision-making criterion, TOPIS and SAW criteria and
flowchart of the multiple decision-making task have been discussed below. Figure 1
Selection of the Best Material for Coating on Solar Cell … 267

Fig. 1 Flowchart of the multiple decision-making task

represents the flowchart of the decision-making task process for the perfect selec-
tion of materials and also represents the two popular multicriteria decision-making
algorithms, and those are used to find out the ranking of the best materials.
(a) Technique for order preference by similarity to ideal solution (TOPSIS)
method:
TOPSIS extent investigation strategy is used to calculate the weight of the materials
and its properties. In the fuzzy methodology, we found out the ranking of the materials
after calculating the weight of material by using this method. The procedural steps
of TOPSIS method are discussed below.
Stage 1: Determine the materials and arrange the materials according to its significant
evaluation properties.
Stage 2: Make a decision matrix: Make a choice matrix dependent on the data,
design, and module that we have. Each row of the decision matrix is assigned out
to one materials and every column to one property. Accordingly, an element M mn of
the decision matrix demonstrates the presentation of mth material with respect to
the nth properties.
Step 3: Obtain the normalized decision matrix, dmn : Find the normalized decision
matrix, dmn using the following equation:

Mmn
dmn =   1/2 (1)

M
2
Mmn
n=1
268 S. Kumar and L. Dewan

Step 4: To find the weight of different materials or the relative position: It is used
to find out the weight of materials properties by using Eq. 2. Calculation of finding
the weight of materials was developed by analytic hierarchy process method [4].
Step 5: To find the weighted normalized matrix, Wmn : Obtain weighted normal-
ized matrix as

Wmn = Wi ∗ dmn (2)

Step 6: To find out the ideal solutions: Obtain the best and worst positive and
negative ideal solution of the desired material by using the following equations:
 max   min 
 Wmn  Wmn
+ 
W = ∈n , ∈ n (m = 1, 2, 3 . . . N ) (3)
m
n m
n
 min   max 
 Wmn  Wmn
−  
W = ∈n , ∈ n (m = 1, 2, 3 . . . N ) (4)
m
n m
n

Step 7: To find the separation measures: The separation alternative ideal solutions
can be calculated by using the following equations.
(a) For positive ideal solution
 0.5

N
 2
Si+ = Wmn − Wn+ , m = (1.2, 3 . . . N ) (5)
n=1

(b) For negative ideal solution

 0.5

N
 2
Si− = Wmn − Wn− , m = (1.2, 3 . . . N ) (6)
n=1

Step 8: To find the relative closeness: The relative closeness to a specific option in
contrast to the perfect arrangement process is as follows:

Si−
Pi = (7)
Si+ + Si−

Step 9: According to relative closeness value, all alternatives are arranged in the
descending order.
Based on all these steps, the results of different materials are as follows:
Selection of the Best Material for Coating on Solar Cell … 269

3 Results and Discussions for TOPSIS

See Table 1.
Step 3: To find normalized decision matrix (Table 2)
Step 4: To find the weight of require parameters
Wn = (0.4 0.1 0.5) be the weight of refractive index, melting point, and transmission.
Step 5: To find weighted normalized matrix (Table 3)
Step 6: To find positive and negative ideal solutions
W + = (0.09104, 0.020127, 0.2757).
W − = (0.168124, 0.049002, 0.1820).
Step 7: To find separation measures (Table 4)
Step 8: To find the relative closeness and ranking (Table 5)

(b) Simple additive weighting (SAW) method

SAW method is easy to use as scoring or weighted linear combination [4] and most
usually used as multiproperty choice technique [10]. The main advantage of this
method is that it is proportional to linear transformation of the raw data which means

Table 1 For Step 1 and Step

Materials Refractive Melting point Transmission
2
index (%) (550 nm)
Hafnium oxide 1.9 2812 83.2
Niobium 2 1512 79.4
oxide(II)
Silicon dioxide 1.46 1713 96.9
Tantalum 2.1 1880 75.8
oxide
Titanium(IV) 2.4 1775 65.7
oxide
Zirconium 2.05 2700 77.5
dioxide
Barium 1.3 1280 99.5
fluoride
Yttrium 1.5 1155 96.1
fluoride
Aluminum 1.63 2046 92.1
oxide
Magnesium 1.38 1266 98.6
fluoride
270 S. Kumar and L. Dewan

Table 2 Normalized
Materials Refractive Melting point Transmission
decision matrix
index (%) (550 nm)
Hafnium oxide 0.3327 0.49002 0.23073
Niobium 0.3526 0.26348 0.2201
oxide(II)
Silicon dioxide 0.2556 0.29851 0.26851
Tantalum 0.3677 0.327614 0.21021
oxide
Titanium(IV) 0.42031 0.30931 0.1820
oxide
Zirconium 0.3590 0.47051 0.2147
dioxide
Barium 0.2276 0.2230 0.2757
fluoride
Yttrium 0.2626 0.20127 0.2662
fluoride
Aluminum 0.2854 0.3565 0.2552
oxide
Magnesium 0.24168 0.22061 0.27322
fluoride

Table 3 Weighted
Materials Refractive Melting point Transmission
normalized matrix
index (%) (550 nm)
Hafnium oxide 0.13308 0.049002 0.23073
Niobium 0.14104 0.026348 0.2201
oxide(II)
Silicon dioxide 0.10224 0.029851 0.26851
Tantalum 0.14708 0.0327614 0.21021
oxide
Titanium(IV) 0.168124 0.030931 0.1820
oxide
Zirconium 0.1436 0.047051 0.2147
dioxide
Barium 0.09104 0.02230 0.2757
fluoride
Yttrium 0.10504 0.020127 0.2662
fluoride
Aluminum 0.11416 0.03565 0.2552
oxide
Magnesium 0.096672 0.022061 0.27322
fluoride
Selection of the Best Material for Coating on Solar Cell … 271

Table 4 Separation measures

N 
0.5
 2
Si+ = W mn − Wn+
n=1

0.00046139
0.00563005
0.00027115
0.1019634
0.0147726
0.0071959
0.00000471
0.00028625
0.00116855
0.7668222191

N 
 2 0.5
Si− = W mn − Wn−
n=1

0.049958
0.002698353
0.07703558
0.00387617
0.000326542
0.001674522
0.015433680
0.071007521
0.13051999
0.04079199

Table 5 Relative closeness Si−

and ranking Pi = Si+ +Si−

0.09164 (8)
0.83274 (5)
0.99649 (4)
0.03661 (9)
0.02162 (10)
0.17681 (6)
0.99969 (3)
0.14131 (7)
1.00104 (2)
1.97983 (1)
272 S. Kumar and L. Dewan

that the relative order of magnitude of the standardized scores remains equal. The
procedural steps of SAW method are discussed below.
Step 1: To obtain a decision matrix: We find a decision matrix which includes M
materials and N properties of material and thus calculate the normalized decision
matrix is:
For positive solution
rmn
n mn = (m = 1, 2, 3 . . . M)(n = 1, 2, 3 . . . N ) (8)
rn∗

And for negative solution

rnmin
n mn = (m = 1, 2, 3 . . . M)(n = 1, 2, 3 . . . N ) (9)
rmn

Step 2: Evaluate for every alternative, Am by following equations.


Am = wn ∗ xmn (10)

where x mn is the score of mth materials with respect to the nth properties of material,
wn is the weight of material properties. In this work, we defined the weight parameters
according to their priority in optical coating applications such that their sum is to be
1.
Wn = (0.4 0.1 0.5) be the weight of refractive index, melting point, and
transmission.
Based on all these steps, the results of different materials are as follows:

4 Results and Discussions for SAW

See Table 6.
Step 2: Obtained decision matrix by using Eq. 11 and ranking will be shown in Table
7.

Am = wn ∗ xmn (11)
Selection of the Best Material for Coating on Solar Cell … 273

Table 6 Table for decision

Materials Refractive Melting point Transmission
matrix
index (%) (550 nm)
Hafnium oxide 0.0137 0.49002 0.230734
Niobium 0.0144 0.263485 0.220195
oxide(II)
Silicon dioxide 0.01057 0.298512 0.2687277
Tantalum 0.015205 0.327614 0.210212
oxide
Titanium(IV) 0.0173 0.309316 0.1822024
oxide
Zirconium 0.0148 0.470510 0.2147525
dioxide
Barium 0.00942 0.223056 0.275938
fluoride
Yttrium 0.010860 0.201273 0.266509
fluoride
Aluminum 0.011 0.35654 0.25541
oxide
Magnesium 0.0099 0.220617 0.2734422
fluoride
Hafnium oxide 0.0548 0.0490 0.1153
Niobium 0.0057 0.0263 0.1100
oxide(II)
Silicon dioxide 0.0042 0.0298 0.1343
Tantalum 0.0608 0.0327 0.1051
oxide
Titanium(IV) 0.0069 0.0309 0.0911
oxide
Zirconium 0.0059 0.0470 0.1073
dioxide
Barium 0.0037 0.0223 0.1377
fluoride
Yttrium 0.0434 0.0201 0.1332
fluoride
Aluminum 0.0044 0.0356 0.1277
oxide
Magnesium 0.0396 0.0220 0.1367
fluoride
Hafnium oxide 1.9 2812 83.2
Niobium 2 1512 79.4
oxide(II)
Silicon dioxide 1.46 1713 96.9
(continued)
274 S. Kumar and L. Dewan

Table 6 (continued)
Materials Refractive Melting point Transmission
index (%) (550 nm)
Tantalum 2.1 1880 75.8
oxide
Titanium(IV) 2.4 1775 65.7
oxide
Zirconium 2.05 2700 77.5
dioxide
Barium 1.3 1280 99.5
fluoride
Yttrium 1.5 1155 96.1
fluoride
Aluminum 1.63 2046 92.1
oxide
Magnesium 1.38 1266 98.6
fluoride

Table 7 Ranking of the 

Am = wn ∗ xmn
decision matrix
0.2191 (2)
0.3787 (1)
0.1683 (6)
0.1986 (3)
0.1309 (10)
0.1602 (9)
0.1637 (8)
0.1967 (5)
0.1677 (7)
0.1983 (4)

5 Correlation

Hence, after solving Eq. (12), we get a correlation between the degree of relationships
(D) and rankings of the materials. Ranking of the materials was found out by the
SAW and TOPSIS methods, and the degree of relationships (D) is calculated by the
difference between the both rankings of the material. In the Table D2 term represent
the square of the each cross ponding value of materials D.
Upcoming resultant value of the best material is always should be near by the
value of r. That is shown in Table 8.

6 nm=1 D 2
r =1− (12)
n(n 2 − 1)
Selection of the Best Material for Coating on Solar Cell … 275

Table 8 Different rankings of the materials

Materials SAW ranking TOPSIS ranking d= r1 − r2 D2
Hafnium oxide 2 8 −6 36
Niobium oxide(II) 1 5 −4 16
Silicon dioxide 6 4 2 4
Tantalum oxide 3 9 -6 36
Titanium(IV) oxide 10 10 0 0
Zirconium dioxide 9 6 3 9
Barium fluoride 8 3 5 25
Yttrium fluoride 5 7 -2 4
Aluminum oxide 7 2 5 25
Magnesium fluoride 4 1 3 9
r =1− 6∗164
10(102 −1)
r = 0.00606

6 Conclusion

In this paper, the main idea was to find out the best materials out of many materials
that are available to use for coating for solar cell and optical filter applications. We
found the titanium(IV) oxide is best material because the resultant value of the degree
of relationships (D) is approximate near to the value of r. The ranking of the materials
has been observed by using TOPSIS and SAW method. In future work, we can also
find out the effectiveness ranking of material by another method.

References

1. Edwards K (2005) Selecting materials for optimum use in engineering components. Mater Des
26:469–473. https://doi.org/10.1016/j.matdes.2004.07.004
2. Deng Y-M, Edwards K (2007) The role of materials identification and selection in engineering
design. Mater Des 28:131–139. https://doi.org/10.1016/j.matdes.2005.05.003
3. Ermolaeva NS, Kaveline KG, L Spoormaker JL (2002) Materials selection combined with
optimal structural design: concept and some results. Mater Des 23:459–470. 10.1016/S0261-
3069(02)00019-5
4. Saaty TL (1990) How to make a decision: the analytic hierarchy process. Eur J Oper Res
48(1):9–26 (1990)
5. Macleod H (2001) Thin-film optical filters, 3rd edn. Institute of Physics Publication, Bristol
and Philadelphia
6. Thelen A (1989) Design of optical interference coating. McGraw-Hill, New York
7. Kumari N et al (2015) Broadband antireflection coatings for multifunctional avionic displays.
In: International Conference on Optics & Photonics 2015. International Society for Optics and
Photonics
8. Chan JWK, Thomas TKL (2007) Multi-criteria material selections and end-of-life product
strategy: grey relational analysis approach. Mater Des 28(5):1539–1546 (2007)
276 S. Kumar and L. Dewan

9. Opricovic S, Tzeng G-H (2004) Compromise solution by MCDM methods: a comparative

analysis of VIKOR and TOPSIS. Eur J Oper Res 156(2):445–455
10. Ashby MF (2000) Multi-objective optimization in material design and selection. Acta Mater
48(1):359–369
Microcontroller-based PID Temperature
Controller Using Genetic Algorithms

Bharat Bhushan Sharma, Anuj Banshwar, Mohit Pathak,

Naveen Kumar Sharma, and Aman Joshi

Abstract In this work, a hardware setup has been designed to control the temper-
ature in a varying environment. For this electronic circuitry, namely, temperature
controller, an 89C51 microcontroller, has been fabricated, and for online controlling
the parameters Genetic Algorithm based fuzzy controller has been implemented in
the Visual-Basic5. A lot of work has been done on the fuzzy controllers as they are
being used in the industries. But these controllers are unable to acquire the data and
have to rely to a great extent on empirical and heuristic data which in many cases,
especially in the varying environment, cannot be objectively elicited. For adaptive
learning, a Genetic Algorithm technique has been employed which is modeled on
the process of Natural Evolution. For adaptive learning, the environment iterative
rule learning (IRL) approach has been utilized. With this approach the knowledge
base table is modified after learning the environment.

Keywords PID · Genetic Algorithm · Visual basic

1 Introduction

This work is a microcontroller-based PID Temperature Controller using Genetic

Algorithms has been designed and fabricated to control the temperature of 1000 ml

B. B. Sharma (B)
School of Automation, Banasthali Vidyapith, Banasthali, Rajasthan, India
e-mail: [email protected]
A. Banshwar
Government Polytechnic, Puranpur, Pilibhit, Uttar Pradesh, India
M. Pathak
Noida Institute of Engineering & Technology, Greater Noida, Uttar Pradesh, India
N. K. Sharma
IKG Punjab Technical University, Main Campus, Kapurthala, Punjab, India
A. Joshi
National Institute of Technology, Kurukshetra, Haryana, India

© Springer Nature Singapore Pte Ltd. 2021 277

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_26
278 B. B. Sharma et al.

water beaker. A PID controller is an essential item for an industry to function properly
and produce quality goods. As we know about the PID controller, its working and
tuning by different methods but we are unable to understand, how PID controller
controls the temperature or any other parameter practically. As well as how one
can tune the PID parameters using the standard methods. To give a feel about PID
controller an electronic circuit has been considered and developed with the help
of 89C51 microcontroller which not only acquires the temperature data from the
Platinum Resistance Thermometer (PRT) but also controls the power fed to the
heater.
The general equation for a continuous PID controller is given in Eq. (1).
⎛ ⎞
t  
1 de ⎠
u = K p ⎝e + edτ + K d (1)
Ki dt
0

where u is the output of the controller, proportional gain is given by K p , integral

reset time is given by K i , derivative time is given by K d , and e is the error.
A step response technique is used for tuning of PID parameters. The PID param-
eters are provided to the PID control program. The PID controls the system effi-
ciently. PID controllers are the most popular controller system used in industries;
robust PID controller gives more robust output than classical PID controller. Despite
of increasing the use of PID (almost 60–80 years), there are hundreds of publications
on control delayed processes collected [1, 2].

2 Experimental Setup

The experimental setup of the project is shown in Fig. 1. As seen in the block
diagram, a PID controller has been considered. 89C51 microcontroller-based system
is designed for temperature control in the beaker. Platinum Resistance Thermometer
(PRT) is used for acquiring the required temperature data of the water inside the
beaker and it provides an output in analog or voltage as output. Amplification of
voltage is done to provide the temperature as direct output. 4½ Analog-to-digital
converter is used and provided the voltage obtained as its input data. The output of
Analog-to-digital converter which is a digital output is provided to the computer by
the help of the microcontroller.
The coding in computer will provide the desired calculation using the PID control
algorithm. After doing all the calculations it would generates the data to be sent to
8bit DAC. The output of DAC which is an analog output is used for firing angle
control. This helps the microcontroller to trigger the TRIAC and control the energy
in the water beaker. Stirrer is used for providing the uniform heat to the system.
Microcontroller-Based PID Temperature Controller Using … 279

Fig. 1 Block diagram of the

PID demonstrator

3 Working

At first, microcontroller receives the commands from the program in PC and acts
accordingly. For acquiring the temperature data from Platinum Resistance Ther-
mometer (PRT) also called Pt-100 first a constant source of 1 ma has been designed
[3]. This current acts as excitation current to the PRT. The voltage across the PRT
is then conditioned (amplified using instrumentation amplifier) in such a way that
the voltage can be easily be accepted by the 14-bit ADC and also its variation will
give us the temperature change (i.e., 10.0 mV = 1.00 °C). This signal will then be
converted in to digital data by 14 bit ADC and will be transferred to the 89C51
Microcontroller. The microcontroller then passes this temperature data to the VB 5.0
for processing. It passes the binary data, after receiving from program in PC, to the
TRIAC firing circuit. The TRIAC firing circuit then fires the TRIAC accordingly.
Figure 2 illustrates the block diagram for PID controller.
The TRIAC firing circuit has Zero-Crossing Detector, Sawtooth Wave generator
triggered by Zero-crossing pulse, Comparator Optically Isolated TRIAC gate trigger,
and TRIAC circuit. In Fig. 3, change of response for varying K p is shown. A control

Fig. 2 Block diagram for

PID controller
280 B. B. Sharma et al.

Fig. 3 Change of response

for varying K p

Fig. 4 Change of response

for varying K i

system is a device or set of devices to manage, command, direct or regulate the

behavior of other devices or systems.
Figures 4 and 5 shows the response for varying K i and K d

4 Optimization Technique

Genetic Algorithm is a computational model based on Natural Evolution. A system

to be optimized is represented by a binary string which encodes the parameters of
the system [4]. A Population of strings which encodes the parameters is used. In
implementing the concept of Genetic Algorithm, the rule base table designed taking
into account the number of counts or in other words the amount of energy to be given
Microcontroller-Based PID Temperature Controller Using … 281

Fig. 5 Change of response

for varying K d

to maintain the temperature of the bath. The rule base has to be optimized with the
help of GA used as a learning tool [5]. The following operators are applied to a set
of population and the best fit individuals are selected to be a part of the rule base.
This is how amount of energy to be given to the heater is regulated. Figure 6 shows
the block diagram for genetic system.
The designing part of the circuit includes PRT (platinum resistor thermometer)
which is used as sensor to measure the temperature change. The PRT sends analog
data to the ADC through instrumentation amplifier (which increases the voltage
gain). The ADC further sends the digital bits to the IC-89C51. The interfacing of
ADC with IC-89C51 also includes over range, Run/Hold, polarity and busy signals to
be checked along with digital data D1 (LSB), D2, D3, D4, D5 (MSB). The interfacing
of binary bits is done at port P0 and port P2 respectively. The ADC (7135) works at

Fig. 6 Block diagram for

genetic System
282 B. B. Sharma et al.

172.8 kHz frequency, so IC-393 is connected with crystal oscillator of 11.059 MHz
which is the required frequency of 89C51. Thus IC-393 used as a frequency divider
of microcontroller frequency, which is further used by ADC.
The output of signal is send to the port P1, after processing operations on the
received data from ADC [6]. The output bits are then sending to the DAC for regen-
erating analog signals. This signal is further passed to TL-071 to convert output
voltage of DAC into current. The current is compared with the sawtooth wave. On
comparing through comparator the TRIAC is fired in order to turn ON the heater.
The pulses act as a timer to the heater, for which the heater is ON for a particular
duration.
GA-based algorithm is done to calculate the tuning using the computer. A firing
angle is generated using the obtained calculation. The GA-based fuzzy control
program in PC continuously monitors the environment. If there is any undesired
change is seen in the system, it modifies the input parameter by self-learning through
GA. Visual Basic is used for writing the codes for the GA-based control [7].

5 Conclusion

The following conclusions are drawn after the up-gradation and automation of bath
such as the automated system would be more reliable as compared to the manually
controlled system. An automated machine is capable of handling a large amount of
data and performs complex calculations without any human intervention. Moreover,
it gives accurate results thereby consuming less time. Thus, the quality of output
results increases.
A computer-controlled machine is less error-prone as compared to the manual
operation. Automation of this machine would lead to better performance increased
efficiency, increased throughput rate and accurate processing of large amount of
data. In the present scenario time saving is the most important consideration. Using
upgrades and automated version of bath saves the user’s time. Thus more samples can
be tested taking lesser amount of time and this increases the profit of organization.

6 Result

The temperature variations in water bath lead to the development of program for inter-
facing the various components ADC, Triac firing circuit and the PC with 89C51. The
program has been tested to be working properly, i.e., it is possible now to send the data
from one component to another in the proper format and way. The GA-based adaptive
fuzzy temperature controller system has been tested at various setpoint temperatures
in varying environments. It can easily adapt itself to the varied environment and work
efficiently as no tuning is required after the change of environment.
Microcontroller-Based PID Temperature Controller Using … 283

References

1. Singh PP, Singh JP, Roy BK (2017) NAC-based synchronisation and anti-synchronisation
between hyperchaotic and chaotic systems: its analogue circuit design and application. IETE J
Res 63:853–869
2. Balas V, Jain L, Balas M (eds) (2017) Soft computing applications. SOFA 2016. Advances in
intelligent systems and computing, Springer, pp 354–364
3. Kumar SMG, Jayaraj D, Kishan AR (2010) PSO based tuning of a PID controller for a high
performance drilling machine. Int. J. Comput. Appl. 0975–8887(1): 12–18
4. Singh JP, Roy BK, Wei Z (2018) A new four-dimensional chaotic system with first Lyapunov
exponent = 22, hyperbolic curve and circular paraboloid types of equilibria and its switching
synchronization by an adaptive global integral sliding mode control. Chin Phys B 27:040500–
040514
5. Gaing ZL (2004) A particle swarm optimization approach for optimum design of PID controller
in AVR system. IEEE Trans Energy Convers 19:384–391
6. Zadeh LA (1973) Outline of new approach to the analysis complex systems and decision
processes. IEEE Trans Syst Man Cybern 3:28–44
7. Chow MY, Tipsuwan Y (2003) Gain adaptation of networked DC motor controllers based on
QOS variations. IEEE Trans Industr Electron 50:936–943
Optimal Energy Dispatch
and Techno-economic Analysis
of Islanded Microgrid with Hybrid
Energy Sources

V. V. S. N. Murty and Ashwani Kumar

Abstract Islanded microgrids with hybrid energy sources are being adopted in
remote areas to ensure reliability, energy security, cost-effective and reduce carbon
emissions. This paper investigates the techno-economic benefits of islanded micro-
grid with hybrid energy sources. Optimal energy dispatch of hybrid system with
photovoltaic (PV), wind turbine (WT), battery energy storage system (BES) and
diesel generator (DG) is determined hourly subject to the minimization of net present
cost (NPC). Various feasible case studies performed with hybrid energy sources and
comparative analysis of techno-economic benefits are demonstrated for each case.
Comparative analysis of techno-economic benefits is given with different configu-
rations of PV + WT + DG + BES, PV + DG + BES and WT + DG + BES.
A simulation study has been carried out using HOMER Pro optimization tool, and
numerical results show that the hybrid microgrid system with PV + WT + DG +
BES has minimum NPC and cost of energy compared to other configurations. Simu-
lation results are obtained for the optimal capacity of PV, WT, DG, BES and SOC
of the storage system.

Keywords Hybrid energy system · Diesel generator · PV · Wind turbine · Battery

energy storage system · Techno-commercial benefits

1 Introduction

Even though the expansion of generation, transmission network and distribution

system are increasing day by day to meet growing load demand, however, today
approximately 14% of world population has no access to electricity. The main objec-
tive of the microgrid network operator is to provide cost-effective, energy security

V. V. S. N. Murty (B) · A. Kumar

Electrical Engineering Department, National Institute of Technology Kurukshetra, Kurukshetra,
Haryana, India
e-mail: [email protected]
A. Kumar
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 285

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_27
286 V. V. S. N. Murty and A. Kumar

and reliable power supply to customers. Microgrid provides significant benefits of

including low-cost system, reliability, integrating excess renewable power genera-
tion to microgrid, growth of rural areas by providing electricity to remote areas,
and emission reduction. Integration of renewable energy sources PV, wind and fuel
cell in distribution system was discussed using hybrid Nelder–Mead Particle Swarm
Optimization algorithm [1]. In [2], various nano-grid configurations are modeled
using HOMER software to evaluate technical and commercial implications. Hybrid
energy system with PV, wind, BES, diesel generator and hydroenergy sources were
simulated in HOMER software for coastal area of Saudi Arabia [3], Urumqi, China
[4], Kadayam, Tamil Nadu [5], Iran [6] to evaluate technical–commercial benefits. A
feasible solution obtained among different configurations of PV/BES/wind/fuel cell
energy sources Saudi Arabia using HOMER Pro software [7]. Techno-commercial
benefits of island grid with hybrid energy sources in Turkey were simulated using
HOMER software [8]. Authors have concluded that cost of energy in islanded grid
mode is higher than grid electricity. Also quoted that BES was economical compared
to fuel cell technology. A hybrid system was simulated using HOMER software to
quantify techno-economic benefits for mobile base transceiver station in Nigeria
[9]. PV + DG-based minigrid is simulated for Nigeria [10] using HOMER soft-
ware. Energy management in microgrids was addressed in [11] to minimize opera-
tional cost. Electricity consumption pattern, wind power output and electricity prices
were forecasted using seasonal autoregressive integrated moving average model.
Techno-economic advantage of islanded microgrid with hybrid energy sources is
presented using HOMER Pro software [12]. The average cost of a typical hybrid
energy system was determined using HOMER software [13, 14]. Major benefits of
BES in grid integrated with non-conventional energy sources were addressed for
reducing operation cost, frequency regulation, power quality improvement, power
fluctuation stabilization, peak load shifting and system stability [15]. Appropriate
technology and size of BES for microgrid expansion problem were solved using
mixed-integer linear programming [16]. Optimal sizing of BES was presented using
mixed-integer programming for commercial applications [17]. Energy dispatch of
DGs, BES in microgrids in presence of uncertainties presented in [18] using mixed-
integer quadratic programming. Optimal energy scheduling problem was solved
using an alternating direction method of multipliers algorithm [19]. Based on the
literature survey, it may note that major work published was pertaining to sizing and
energy dispatch. This paper addresses optimal sizing of PV/WT/DG/BES, energy
management and techno-commercial aspects of the islanded microgrid in Vellian-
gattu, Tamil Nadu, India. In Tamil Nadu, the potential availability of natural resources
of solar and wind is wealthy. Solar and wind energy sources are pollution-free, envi-
ronmentally friendly and unlimited energy resource to provide energy security. The
region has significant solar irradiation for almost 300 days of clear sun, therefore
suitable for solar power generation. Also, average wind velocity is 3–5 m/s which
is more suitable for wind power generation. For accurate analysis, real-time data of
solar irradiation and wind velocity for the site location (Tamil Nadu) is taken from
the weather portal and considered in the simulation study. The main contribution of
the paper is:
Optimal Energy Dispatch and Techno-economic Analysis … 287

• islanded microgrid system is modeled in HOMER Pro software to feed load

demand, and best configuration is identified based on technical, commercial and
emission results;
• explores techno-economic benefits of islanded microgrid configurations of PV +
DG + BES, WT + DG + BES and PV + WT + DG + BES;
• simulation results are obtained for optimal capacity of PV, WT, diesel generator,
converter, battery energy storage system and state of charge of BES;
• seasonal load variation and renewable power generation are suitably taken care
in the simulation study.
The paper is organized as follows: Section 2 explains the hybrid system schematic
model of HOMER Pro software. Simulation results are presented in Sect. 3. Finally,
the paper concluded in Sect. 4.

2 Islanded Microgrid with Hybrid Energy Sources

HOMER Pro software schematic configuration of standalone microgrid system with

PV, wind, DG, battery energy storage system and load is shown in Fig. 1. Optimal
capacity of each component is determined to meet load completely; therefore, unmet
electricity demand is zero throughout the planning period. The hybrid microgrid
system is modeled in HOMER Pro software, and best configuration is identified
based on technical, commercial and emission results.
Daily load profile and yearly load demand pattern is shown in Figs. 2 and 3.
Peak load demand and average daily load demand on the system are 13.23 kW and

Fig. 1 Schematic model of

islanded microgrid with
hybrid energy sources
288 V. V. S. N. Murty and A. Kumar

10
8
Load (kW)
6
4
2
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hours

Fig. 2 Daily electricity demand pattern

14
12
10
Load (kW)

8
6
4
2
0
1
367
733
1099
1465
1831
2197
2563
2929
3295
3661
4027
4393
4759
5125
5491
5857
6223
6589
6955
7321
7687
8053
8419
Hours

Fig. 3 Yearly load demand variation

127.17 kWh/day, respectively. Hourly variation of solar irradiation and wind speed
is shown in Figs. 4 and 5 for selected site location.

0.2
Solar irradiation

0.15
(kW/m^2)

0.1

0.05

0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Time (Hours)

Fig. 4 Hourly solar irradiation profile

Optimal Energy Dispatch and Techno-economic Analysis … 289

wind speed (m/sec) 6

0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Time (Hours)

Fig. 5 Hourly wind speed profile

3 Economic Modeling

HOMER Pro software optimizes various possible configurations and ranks each
feasible configuration based on net present cost (NPC). NPC of a system includes
present value cost and revenue earned. Capital cost, replacement cost, operation &
maintenance cost, and fuel cost are included in NPC calculation. NPC is calculated
using Eq. (1).

Cann, tot
NPC = (1)
CRF(i, N )

Cann,tot is total annualized cost ($/year), i is rate of interest (%), CRF is capital
recovery factor and N is project lifetime (years). Project lifetime is considered as
20 years in the simulation study. CRF is determined using the following equation.

i(1 + i) N
CRF(i, N ) = (2)
i(1 + i) N − 1
i − f
i= (3)
1+ f

i is the actual rate of interest, i  is the discount rate of interest and f is the inflation
rate.
Levelized cost of energy (LCOE) is the ratio of total annualized cost to total
electrical load served E served (kWh/year).

Cann, total
Levelized COE = (4)
E served

Polycrystalline-type Canadian Solar MaxPower makes PV panel considered in

the simulation study. The efficiency of PV panel is 16.94%, and lifetime is set as
25 years. Capital and replacement cost of PV system is considered as 900$/kW, and
O&M cost is 10$/kW/year. Bergey Windpower making wind generator is considered
290 V. V. S. N. Murty and A. Kumar

in this paperwork. Capital and replacement cost of wind turbine system is considered
as 15,000$/kW, O&M cost is 75$/kW/year, and lifetime is set as 20 years. Cut-in
and cut-out speeds are taken as 3 m/s and 20 m/s, respectively. Rotor diameter and
hub height of wind turbine is 15.81 m and 10 m, respectively. Diesel generator of
TDPS make is considered in the simulation study. Capital and replacement cost of
DG is considered as 900$/kW, O&M cost is 0.04$/kW/year, and fuel price is 1$/liter.
Trojan-make BES is included in the simulation study. Capital and replacement cost
of BES is considered as 220$/kW, O&M cost is 4$/kW/year, and lifetime is set as
10 years.

4 Results and Discussions

This paper simulates islanded microgrid with hybrid energy sources

PV/DG/wind/BES in remote village Velliangattu, Tamil Nadu, India, using
HOMER Pro software to quantify techno-commercial benefits, and the optimal
scheme is determined based on annual net present cost. Simulation results are
obtained for optimal capacity of PV, WT, diesel generator, converter, battery energy
storage system and state of charge of BES. Simulation results have been obtained
for following configurations:
Case 1: PV + WT + DG + BES.
Case 2: PV + DG + BES.
Case 3: WT + DG + BES.
System architecture details are given in Table 1 for three configurations. It is
observed that PV + WT + DG + BES system is most optimal among other scenarios.
Total annual load consumption of the system is 45,583 kWh/year which is met by
PV power production 15,885 kWh/year, DG power production 14,679 kWh/year and
wind turbine power production 23,948 kWh/year. It is evident from simulation results
that total electricity demand is being supplied by 67.8% renewable fraction and 32.2%
non-renewable fraction. Levelized COE of the optimal system is 0.7801$/kWh, and
NPC value is $459,662.80. The total initial cost is low for case-2 hybrid system PV
+ DG + BES and high for case-1 PV + WT + DG + BES. Total operating cost and
cost of energy (COE) are low for case-1 PV + WT + DG + BES compared to other
hybrid systems. NPC and cash flows of hybrid system case-1 are shown in Figs. 6
and 7.
Cost comparison of various hybrid energy systems is indicated in Fig. 8. Optimal

Table 1 System architecture details

PV (kW) WT (kW) DG (kW) BES (strings)
PV + WT + DG + BES 9.89 9 4 48
PV + DG + BES 10 – 5 8
WT + DG + BES – 9 5 43
Optimal Energy Dispatch and Techno-economic Analysis … 291

$300,000.00

$250,000.00

$200,000.00
NPC

$150,000.00

$100,000.00

$50,000.00

$0.00
WT Converter DG PV BES

Fig. 6 Annual net present cost of hybrid system case-1

150000
WT Converter DG PV BES

100000
50000
Cash Flows

0
-50000
-100000
-150000
-200000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Years

Fig. 7 Cash flows for hybrid system case-1

PV+WT+DG+BES PV+DG+BES WT+DG+BES

$1,000,000

$500,000
$637,403
$459,663
$563,821

$164,379

$153,799
$22,841
$41,935
$37,409

$21,711
$0.78
$0.96
$1.08

$0
NPC COE Operating Cost Initial Cost

Fig. 8 Cost analysis of different hybrid energy systems

292 V. V. S. N. Murty and A. Kumar

PV DG Load (kW) WT
15

10
Power (kW)

0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
-5

-10

Fig. 9 Daily power dispatch of hybrid system

Load PV WT DG BES
15

10

5
Power (kW)

0
1
304
607
910
1213
1516
1819
2122
2425
2728
3031
3334
3637
3940
4243
4546
4849
5152
5455
5758
6061
6364
6667
6970
7273
7576
7879
8182
8485
-5

-10

-15
Time (hours)

Fig. 10 Yearly power dispatch of the hybrid system

energy dispatch of the hybrid system is depicted in Figs. 9 and 10 to balance electricity
demand and power generation. The simulation results indicate that hybrid microgrid
system with PV + WT + DG + BES system reduces CO emissions by 72% and
CO2 emissions by 15.3% per year as compared to the islanded microgrid system
operating with DG alone. Rate of return (IRR) is 13.2%, return on investment (ROI)
is 9.4%, and payback period is 7 years for the hybrid system.

5 Conclusion

This paper evaluates techno-commercial benefits of islanded microgrid with hybrid

energy sources in remote village Velliangattu, Tamil Nadu, India. Different feasible
configurations of a hybrid system with PV/WT/DG/BES are studied in this paper,
and comparative analysis has been presented. Capital cost, operational cost, fuel cost,
cost of energy and total cost are determined for each configuration. From simulation
Optimal Energy Dispatch and Techno-economic Analysis … 293

results, it is observed that configuration PV + WT + DG + BES has lower NPC

and COE compared to other configurations considered in the simulation study. This
paper can be useful to microgrid operator for decision making, solid investment
toward rural electrification, design a competitive islanded microgrid and optimal
energy dispatch.

References

1. Senthil kumar J, Charles Raja S, Srinivasan D, Venkatesh P (2018) Hybrid renewable energy-
based distribution system for seasonal load variations. Int J Energy Res 42:1066–1087
2. Rajkumar RK, Ramachandaramurthy VK, Yong BL, Chia DB (2011) Techno-economical
optimization of hybrid pv/wind/battery system using neuro-fuzzy. Energy 36:5148–5153
3. Ramli MAM, Hiendro A, Al-Turki YA (2016) Techno-economic energy analysis of wind/solar
hybrid system: case study for western coastal area of Saudi Arabia. Renew Energy 91:374–385
4. Li C, Ge X, Zheng Y, Xu C, Ren Y, Song C (2013) Techno-economic feasibility study of
autonomous hybrid wind/PV/battery power system for a household in Urumqi. China, Energy
5. Amutha WM, Rajini V (2016) Cost benefit and technical analysis of rural electrification
alternatives in Southern India using HOMER. Renew Sustain Energy Rev 62:236–246
6. Baneshi M, Hadianfard F (2016) Techno-economic feasibility of hybrid diesel/PV/wind/battery
electricity generation systems for non-residential large electricity consumers under Southern
Iran climate conditions. Energy Convers Manag 127:233–244
7. Al-Sharafi A, Sahin AZ, Ayar T, Yilbas BS (2017) Techno-economic analysis and optimization
of solar and wind energy systems for power generation and hydrogen production in Saudi
Arabia. Renew Sustain Energy Rev 69:33–49
8. Duman AC, Güler O (2018) Techno-economic analysis of off-grid PV/Wind/Fuel cell hybrid
system combinations with a comparison of regularly and seasonally occupied households.
Sustain, Cities Soc
9. Olatomiwa L, Mekhilef S, Huda ASN, Sanusi K (2015) Techno-economic analysis of hybrid
PV-diesel-battery and PV-wind-diesel-battery power systems for mobile BTS: the way forward
for rural development. Energy Sci Eng 3:271–285
10. Akinyele D (2018) Analysis of photovoltaic mini-grid systems for remote locations: a techno-
economic approach. Int J Energy Res 42:1363–1380
11. Tavakoli M, Shokridehaki F, Marzband M, Godina R, Pouresmaeil E (2018) A two stage
hierarchical control approach for the optimal energy management in commercial building
microgrids based on local wind power and PEVs. Sustain Cities Soc 41:332–340
12. Krishan O, Suhag S (2019) Techno-economic analysis of a hybrid renewable energy system
for an energy poor rural community. J Energy Storage 23:305–319
13. Singh S, Gao DW, Giraldez J (2017) Cost analysis of renewable energy-based microgrids. In:
North American Power symposium (NAPS), Morgantown, WV, USA, pp 1–4
14. Weber J, Gao DW, Gao T (2016) Affordable mobile hybrid integrated renewable energy system
power plant optimized using HOMER Pro. In: IEEE 2016 North American power symposium
(NAPS), Denver, CO, USA, pp 1–6
15. Xie H, Chen H, Yan W, Gao W, Sun Q, Ma J (2017) Application of energy storage in high
penetration renewable energy system. In: IEEE international conference on electro information
technology (EIT), Lincoln, NE, USA, pp 188–193
16. Alsaidan I, Khodaei A, Gao W (2018) A comprehensive battery energy storage optimal sizing
model for microgrid applications. IEEE Trans Power Syst 33(4): 3968–3980
17. Alsaidan I, Gao W (2017) Amin Khodaei, battery energy storage sizing for commercial
customers. In: IEEE power & energy society general meeting, Chicago, IL, USA, pp 1–5
294 V. V. S. N. Murty and A. Kumar

18. Ji Y, Wang J, Yan S, Gao W, Li H (2015) Optimal microgrid energy management integrating
intermittent renewable energy and stochastic load. In: IEEE advanced information technology,
electronic and automation control conference (IAEAC), Chongqing, China, pp 1–5
19. Zhang H, Li Y, Gao DW, Zhou J (2017) Distributed optimal energy management for energy
internet. IEEE Trans Indus Inform 13(6):3081–3097
Performance Analysis of Carrier
Modulation Based H-Bridge Inverter
for Hybrid Microgrid

Mukh Raj Yadav and Navdeep Singh

Abstract Hybrid micro-grid system with cascaded hybrid-bridge multilevel inverter

(CHBMI) has been discussed on the analysis of a bi-directional DC/DC converter
for DC micro-grid and CHBMI-based AC grid system. There are two types of
controller have been used for controlling the unequal loading of load which is
connected with DC grid and second one controller has been used to control the
power quality of the system with the help CHBMI controller. This hybrid controller
attached with DC grid helps to manage of power between bi-directional DC/DC
converter and boost converter during the unequal loading condition. CHBMI has
many features over voltage source inverter as low weight due to large transformer
and higher efficiency as such low total harmonic distortion (THD). Carrier-based
bipolar modulation technique of the CHBMI plays a prominent role to achieve better
power quality performance. The analysis of different modulation techniques such
as phase disposed (PDPWM), phase opposition disposed (PODPWM), alternate
phase opposition disposed (APODPWM), phase shift (PS-PWM), COPWM and
variable frequency (VFPWM) has been discussed for load With AC micro-grid by
the MATLAB/Simulink with CHBMI. The performance of hybrid controller has
been discussed to verify the power management.

Keywords DC micro-grid · Bi-directional converter · Battery · Boost controller ·

Bi-directional controller · Multilevel inverter

1 Introduction

In current time-electrical power system, micro-grid can be explained as the combi-

nation of distribution generation, energy storage device, power electronics controller

M. R. Yadav · N. Singh (B)

Department of Electrical Engineering, Madan Mohan Malaviya University of Technology,
Gorakhpur, India
e-mail: [email protected]
M. R. Yadav
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 295

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_28
296 M. R. Yadav and N. Singh

and interconnection of variable loads worked as a single controlled unit from the
utility grid depend on grid connected or islanding mode [1]. In case of micro-grid,
there are several benefits to consume and utilized for improvement in energy effi-
ciency, compensation of renewable energy sources and synthesis of electric vehicles.
There are normally three types of micro-grid such as AC [2], DC micro-grid [3] and
hybrid micro-grid combination of AC with DC grid [4]. In case of hybrid micro-grid
combination, it involves the numerous AC & DC loads via an AC bus though power
converter, DC loads and battery devices a common DC bus. The DC micro-grid is
more dynamic, and asynchronous as compared with AC micro-grid [5, 6]. The power
factor losses are neglected and regulating the infiltration level.
During power imbalance in DC micro-grid, bi-directional converter (BDC) plays
an important role in coordinate energy storage device with the micro-grid. In case
of islanded mode of the DC micro-grid, the BDC has the following objective: (1)
To control the direction and amount of required power flowing from and toward the
battery. (2) Controller of DC link should help smoothly transition and manages flow
of power in both directions [7].
The CHBMI has many advantages as lower switching frequency, better reliability,
higher voltage level with higher power rating in comparison with voltage source
converter (VSC). The performance quality, operating switching frequency, power
balancing problem, conversion of voltage upgradation and lower input voltage rating,
reduction of filter size has been replaced by CHBMI in comparative between two
multilevel inverters, flying capacitor and neutral point clamped. CHBMI is connected
to the DC to AC micro-grid system to produced load/grid power distribution system
[6].
There are different types of PWM control strategy such as with and without carrier-
based modulation techniques has been generalized for mitigation of harmonics and
upgradation of quality of voltage and current for hybrid micro-grid. The classical
carrier-based modulation techniques sinusoidal pulse modulation such as phase
opposition disposition (POD), phase disposition (PD), phase shift carrier, alternate
phase opposition disposition (APOD), variable frequency pulse width modulation
(VFPWM) and carrier overlapping (COPWM) topology needs a higher switching
frequency [8, 9].
This paper is basically divided into different sections. Section 2 about the intro-
duction of micro-grid architecture and configuration and described about the hybrid
controller used to provide the switching pulses for managed the power of micro-
grid. Three different types of controller are used such as PV boost type controller, bi-
directional DC-DC converter and multilevel modulation controller. These controllers
are used to mitigate the power quality problem. Each controller discussed in below.
Hybrid controller used to provide the switching pulses for balancing the power during
islanding. Section 3 explains the working and comparison of different carrier-based
modulation of the converter followed by the results, conclusion and references. An
effort has been made to design five-level (5L) CHBMI with carrier-based modulation
for hybrid micro-grid system, for improvement of THD and efficiency.
Performance Analysis of Carrier Modulation Based … 297

2 Hybrid Micro-grid Architecture and Configuration

Figure 1 describes about hybrid micro-grid architecture contains hybrid energy

storage DC system, DC load, multilevel inverter with AC linear and nonlinear load
to tie the grid or self-islanded at the time of need.
A. Hybrid Energy Storage DC System model:
Hybrid energy storage DC system (HESDS) model explained about the two different
types of sources having two converter technology. An UDC (unidirectional converter
works as a boost converter) with PV configuration and a bi-directional converter
(BDC) with battery storage system. Since UDC provides one direction power flow
maintained the power balance flow between them by BDC as shown in Fig. 2. The
PV array’s voltage will the input voltage for the boost converter. The output voltage

Fig. 1 Simplified hybrid micro-grid architecture

Lbc D

Lbc D
S1 C1
S1 C1
BOOST CONVERTER
BOOST CONVERTER

Luc
S2 Luc
S2

DC C3 S3 C2 DC C3 S3 C2

BDC BDC

(a) Power flow by BDC (b) Power consumed by BDC

Fig. 2 Operating mode of HESDS

298 M. R. Yadav and N. Singh

equations being associated with the boost converter are given by (2) [10],

Vs DT = (V0 − VS )(1 − D)T (1)

V0 1
MV = = (2)
VS 1− D

The bi-directional converter described the function of boost as well as buck

converter so it will be connected after DC sources. The HESDS four modes of
operations explain about the switching pattern of the switches as shown in Table1.
The PV system is variable in nature because of their dependency on cloud condi-
tions. Hence, to regulate the output voltage, boost converter is controlled using a
PI controller boosting variable voltage into the constant output to fulfill the load
requirements. The controlling of the converter mainly depends on the configuring of
the values of K p and K i PI controller. The unidirectional converter controller inter-
faces with the boost converter connected to the PV system. Since, the presence of
battery will lead to the interfacing of the bi-directional DC/DC converter helps in
charging and discharging of the battery as shown in Fig. 3. The operation of the
converter is between two voltages, i.e., the voltage of the battery and the voltage of

Table 1 Switching pattern

MODE S1 S2 S3
according to mode of
operation I Off Off On
II On Off On
III On Off Off
IV On On Off

Fig. 3 Hybrid controller for DC grid

Performance Analysis of Carrier Modulation Based … 299

the DC bus. The controller for the battery charger corresponds to the power/current
signal for the battery charger as well as discharging of the battery. It also regulates
the DC bus voltage in case of islanded micro-grid. The value of Kp and Ki decides
the charging and discharging mode of the battery. Bi-directional converter controller
is used to operating the converter in the buck and boost mode so that the system can
be operated in the heavy load condition and the light load condition [11].
B. CHBMI:
CHBMI needs distinct DC voltage at the input side contribute the better output qual-
ities and minimized the requirements of filter parasitic elements [10]. The structure
of CHBMI consists of two single phase full bridge with four different switches as
Sw1–Sw4 to obtain sixteen operating states for bi-directional current flow allows
with desired output voltage. The individual H-bridge generates, as shown in Fig. 4,
the three types of voltage range 0, + V /2, V and desired result in combination of five
different states 0, + V /2, V, −V /2, and –V which are shown in Table 2. The oper-
ating pulses operates the switches directly implemented to PWM pulse generator for
generating the output voltage at frequency, modulation index and carrier frequency
as 50 Hz, 1 and 1 kHz, respectively.
Multiple carrier pulse width modulation (MCPWM) used for multilevel inverter.
Multiple carrier pulse width modulation is distinct triangular carrier wave compared

Fig. 4 Circuit diagram of CHBMI

Table. 2. Switching pattern of CHBMI

Output voltage (V o ) Sw1 Sw2 Sw3 Sw4 Sw5 Sw6 Sw7 Sw8
0 Off Off On On Off Off On On
+V /2 Off Off On On On Off Off On
+V /2 On Off Off On Off Off On On
V On Off Off On On Off Off On
−V /2 Off On On Off Off Off On On
−V /2 Off Off On On Off On On Off
−V Off On On Off Off On On Off
300 M. R. Yadav and N. Singh

with an alternating reference wave. There is a various carrier signal used for
improvement of power quality [8, 9].
• Phase disposition PWM (PDPWM)
• Phase opposition disposed (POD) bipolar modulation technique for (CHBMI)
• Alternative phase opposition disposition PWM (APODPWM)
• Phase shift (PS) PWM
• Carrier overlapping PWM strategy (COPWM)
• Variable frequency PWM strategy (VFPWM).
In the phase disposition pulse width modulation types controller for (5-L) inverter,
the carrier wave amplitude is equal for all carrier signal such as frequency, phase
and amplitude, vary with the time interval between minimum to maximum with DC
offset equal to the carrier amplitude. The modulation and carrier signal for (PDPWM)
controller with varying carrier signal magnitude as shown in Fig. (5a). The phase
opposition disposed modulation techniques are said to be whenever the carrier signal
analogous to negative half of reference signal are 180° out of phase reference to the
positive half carrier signal. For the first half and second half, the carrier signal has a
positive and negative slope with corresponding to the reference axis with below and
above, respectively. The modulation and carrier signal for (PODPWM) controller
with varying carrier signal amplitude as shown in Fig. (5b). The APODPWM are
defined as the modulation and carrier for alternate phase opposition disposed control
techniques with varying carrier signal magnitude is shown in Fig. (5c). The PS-PWM
is equal as phase disposed PWM but there are two distinct modulation waves will be
situated with a phase difference. In this case, 45° phase delay between two distinct
modulation signals. The modulation and carrier signal for 5-L multilevel inverter
PWM-phase shift techniques with varying carrier signal magnitude are represent
in Fig. (5d). In carrier overlapping pulse width modulation overlapping each other.
The overlapping of the vertical distance occurs half of the carrier amplitude between
each carrier with varying carrier signal magnitude are shown in Fig. (5e). In this
topology, the multilevel inverters which are the switching frequency of the lower
and upper condition are must be more compared with the intermediate switching
condition. Hence, (m − 1) number of carrier wave of the equal magnitude and phase,
but variation of the frequency is to be used in the m-level inverter as shown in Fig. (5f)
techniques, the carriers with the equal supply frequency and maximum magnitude
AC.
C. DC Grid System/AC Dynamic Load:
The DC grid system generally fulfilling the requirement of the load and has very
attractive feature as low cost, easy controlling without any reactive power concern,
higher efficiency of energy conversion with less electronic components. The residen-
tial AC type load generally involves the different types of home appliances such as
DC refrigerators, laptops batteries, mobile phones batteries, etc., considered as light
loads.
Performance Analysis of Carrier Modulation Based … 301

(a) Phase disposition (PD) bipolar Modulation (b) Phase opposition disposed(POD) bipolar
technique for CHBMI Modulation technique for CHBMI

(c) Alternate Phase opposition disposed (d) Phase Shifted (PSPWM) bipolar
(APOD) bipolar Modulation technique for Modulation technique for CHBMI
CHBMI

(e) Carrier overlapping PWM (COPWM) (f) Variable frequency PWM (VFPWM)
bipolar Modulation technique for CHBMI bipolar Modulation technique for CHBMI

Fig. 5 Multiple carrier pulse width modulation (MCPWM)

3 Simulation Results

Hybrid micro-grid has been simulated for DC grid and AC grid system. It has divided
as three parts given below:
• Performance of DC micro-grid system for constant load and variable micro-grid
current
• Performance of micro-grid system with CHBMI for RL load
• Comparative analysis of modulation technique.

A. Performance of DC micro-grid system for constant load and variable micro-

grid current
302 M. R. Yadav and N. Singh

The result of the system is being described accordingly in the below section. This
part of discussion is about the constant load and the variation in the current. This
case is divided into three parts, i.e., t 1 (0–1.5 s), t 2 (1.5–3 s) and t 3 (3–5 s). During t 1
time the bi-directional converter produces current 2.9 A. During t 2 time the islanding
operation of micro-grid takes place. Hence, the current produce by the bi-directional
converter is 0.8 A. During t 3 time operation the current will be 1.2 A. The operating
mode for given constant load and variable micro-grid current is divided into three
parts as shown in Fig. 6a, b and Table 3.
B. Performance of micro-grid system with CHBMI for RL load
The different modulation techniques are design in MATLAB software. On the basis
of THD, we can find the best modulation techniques among these techniques. The
output current for RL load is 1.180 A and output voltage is 229.6 V, from CHBMI
as shown in Fig. 7a, b. THD of current and voltage VFPWM is 3.59 and 18.99% for
RL load as shown in Fig. 8k, l, respectively.
C. Comparative analysis of modulation technique

(a) Output Voltage waveform of the Load (V)

Islanding Mode

(b) Output Power waveform in the System (W)

Fig. 6 Performance of DC micro-grid system for constant load and variable micro-grid current

Table 3 Management of load system for constant load and variable micro-grid current
Time duration (s) Pload (w) PBDC (w) Pgrid (w) PBoost (w)
t1 1000 −850 −200 750
t2 1000 −250 0 750
t3 1000 −250 600 750
Performance Analysis of Carrier Modulation Based … 303

(a) Single phase of CHBMLI in Simulink (b): Simulation of PD

(c) Output current of CHBMI for RL Load (d) Output Voltage of CHBMI for RL Load

Fig. 7 Performance of micro-grid system with CHBMI for RL load

Shows the output response of voltage and current related to THD for phase disposed
(PDPWM), phase opposition disposed (PODPWM), alternate phase opposition
disposed (APODPWM), phase shift (PS-PWM), COPWM and variable frequency
(VFPWM) types controller varying carrier signal magnitude. The THD for voltage
and current has to substantially minimize 24.06–4.89% for PDPWM as shown in
Fig. 8a, b, on the other hand, THD for PODPWM is 23.43–3.76% in Fig. 8c, d. After
applying APODPWM technique, THD of converter become 25.48–5.45%, respec-
tively, as described in Fig. 8e, f. At last, there are three technique PS-PWM, COPWM
and VFPWM has simulated at 1 kHz switching frequency, then THD of converter
has value for voltage and current 21.15–3.84% in Fig. 8g, h, 29.92–8.98% in Fig. 8i,
j and 18.99–3.59% in Fig. 8k, l, respectively.
The comparison of different carrier modulation is shown in Table 4. It can show
that VFPWM better than PD, POD, APOD, COPWM and PS-PWM. The THD for
voltage and current of VFPWM are the 18.99%, 3.59%, respectively.
304 M. R. Yadav and N. Singh

(a) THD for Voltage for PD (b) THD for current for PD

(c) THD for Voltage for POD (d) THD for Current for POD

(g) THD for Voltage for PSPWM (h) THD for Current for PSPWM

(i) THD for Voltage for COPWM

(j) THD for Current for COPWM

(k) THD for Voltage for VOPWM

(l) THD for Current for VOPWM

Fig. 8 FFT spectrum analysis of different MLI strategy

Performance Analysis of Carrier Modulation Based … 305

Table. 4 Comparison of
Topology Voltage (THD) (%) Current (THD) (%)
results
PD 24.06 4.89
POD 23.43 3.76
APOD 25.48 5.45
Phase shift PWM 21.15 3.84
COPWM 29.92 8.98
Variable frequency 18.99 3.59
PWM

4 Conclusion

This paper involves designing and analysis of hybrid micro-grid system interfacing
of a DC and AC micro-grid containing a boost converter with PV and BDC with
battery through DC grid. DC grid connected with DC load and CHBMI for AC load
system. Modeling of DC grid and CHBMI with AC grid has been discussed with
its controller for controlling required load power for constant load with variable
micro-grid current. On the other hand, AC micro-grid with CHBMI is simulated for
various carrier modulation techniques. The performance of the hybrid DC controller
technique has been discussed to verify the power management between the systems.
The performance of the CHBMI is analyzed for POD, PD, APOD, COPWM and
PS-PWM. VFPWM provide the THD of voltage and current is 18.99% and 3.59%,
respectively, to provide better power quality performance with power balance during
unbalanced loading for DC grid controlled by hybrid controller.

References

1. Mahmood H, Michaelson D, Jiang J (2014) A power management strategy for PV/battery

hybrid systems in islanded microgrids. IEEE J Emerg Sel Top Power Electron 2(4):870–882
2. Gui Y, Wei B, Li M, Guerrero JM, Vasquez JC (2018) Passivity-based coordinated control for
islanded AC microgrid. Appl Energy 229:551–561
3. Han Y, Chen W, Li Qi, Yang H, Zare F, Zheng Y (2019) Two-level energy management strategy
for PV-Fuel cell-battery-based DC microgrid. Int J Hydrogen Energy 44(35):19395–19404
4. Vargas U, Lazaroiu GC, Tironi E, Ramirez A (2019) Harmonic modeling and simulation of a
stand-alone photovoltaic-battery-supercapacitor hybrid system. Int J Electr Power Energy Syst
105:70–78
5. Kondrath N (2017) Bidirectional DC-DC converter topologies and control strategies for
interfacing energy storage systems in microgrids: an overview. In: 2017 IEEE international
conference on smart energy grid engineering (SEGE), Oshawa, pp 341–345
6. Chintala LR, Peddapelli SK, Malaji S (2016) Improvement in performance of cascaded multi-
level inverter using triangular and trapezoidal triangular multi carrier SVPWM. Adv Electr
Electron Eng 14(5):562–570
7. Yadav M, Shukla A, Singh N (2019) Control and implementation of bi-directional converter for
power management of unbalanced DC microgrid. In: International conference on computing,
power and communication technologies GUCON, NCR New Delhi, India, pp 343–349
306 M. R. Yadav and N. Singh

8. Kavali J, Mittal A (2016) Analysis of various control schemes for minimal total harmonic
distortion in cascaded H-bridge multilevel inverter. J Electr Syst Inf Technol 3(3):428–441
9. Mohan Harin M, Vanitha V, Jayakumar M (2017) Comparison of PWM techniques for a three
level modular multilevel inverter. Energy Procedia 117:666–673
10. Kumar S, Kumar R, Singh N (2017) Performance of closed loop SEPIC converter with DC-DC
converter for solar energy system. In: 2017 4th international conference on power, control &
embedded systems (ICPCES), Allahabad, pp 1–6
11. Mirzaei A, Forooghi M, Ghadimi AA, Abolmasoumi AH, Riahi MR (2017) Design and
construction of a charge controller for stand-alone PV/battery hybrid system by using a new
control strategy and power management. Sol Energy 149:132 144
Optimal Energy Management in Hybrid
Microgrid with Battery Storage

Masood Rizvi, Bhanu Pratap, and Shashi Bhushan Singh

Abstract Microgrid, consisting of photovoltaic and battery storage, integrated with

a diesel generator is considered to be capable of providing energy supply that is
economically optimal and technically feasible for current and future use in relation
to increases in energy demands and depletion of conventional sources. In this paper,
optimal sizing and energy management of a hybrid PV/Diesel generator with battery
energy storage is proposed for a grid-connected microgrid of a dental college and its
hospital. The paper takes account of sensitivity analysis performed to examine the
impact of the cost of investment of PV, solar insolation, cost of converters, and cost of
fuel price on the optimum result of energy management. Sensitivity analysis is done
by considering mean repair time for utility grid outage of 2 and 5 h and the fuel price
of $0.97, $1.1, and $1. 35. Net present cost of the energy management system, fuel
consumption, and emissions are minimized by using the software HOMER (Hybrid
Optimization Model for Electric Renewables) designed at the National Renewable
Energy Laboratory US.

Keywords Renewable energy · Cash flow · HOMER · Optimization

1 Introduction

Worldwide non-renewable resources are main sources of electrical energy which

results in environmental pollution [1]. The use of microgrids as a platform to incor-
porate distributed generation systems such as renewable energy sources (RES) and

M. Rizvi (B)
School of Renewable Energy and Efficiency, National Institute of Technology Kurukshetra,
Kurukshetra, India
e-mail: [email protected]; [email protected]
KIET Group of Institutions, Delhi-NCR, Ghaziabad, India
B. Pratap · S. B. Singh
Department of Electrical Engineering, National Institute of Technology Kurukshetra,
Kurukshetra, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 307

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_29
308 M. Rizvi et al.

conventional energy sources in grid-connected as well as in remote power systems,

has increased due to their less ecological impact and no harmful emissions. Further,
the depletion of reserves of coal and crude oil and an increase in demand for elec-
trical energy necessitates the development of a distributed generation system that
produces electrical energy based on renewable resources [2]. The use of a diesel
generator and battery storage system has become inevitable due to the intermittency
in the nature of RES [3]. Diesel generator (DiG) and battery energy storage are used
for ensuring continuous power supply during periods of low sunshine and load shed-
ding at the grid. The energy consumption pattern of the hospitals is less sensitive
to climatic conditions and requires special conditions than other types of buildings
[4]. Therefore, special consideration is required in the management of the energy
for the hospital of the dental college to save and minimize the cost of the energy [5,
6]. Moreover, in modern hospitals, the comfort of the patients carries great impor-
tance and therefore, the quality of service can’t be compromised to implement the
cost-cutting measures. However, effective measures can be taken to improve energy
efficiency and reduce the cost of energy while upholding a high standard of service,
comfort, and productivity [7–9]. In general, for healthcare facilities in hospitals, it is
necessary to develop a baseline of energy consumption by its effective management
and setting goals to reduce consumption of fossil fuels. In this research two scenarios
are analyzed. In scenario 1, the existing model consisting of DiG and grid is simu-
lated and in scenario 2, the proposed model consisting of Photovoltaic (PV), DiG,
Battery Storage (BS), and the grid is analyzed. Sensitivity analysis for 55 annual
grid outages is done by considering mean grid repair time of 2 and 5 h and fuel
price of $0.97, $1.1, and $1. 35. Research indicates that the system reliability can be
improved and energy cost and harmful emissions can be reduced by integrating PV
in micro-grid (MG). HOMER software is used for optimizing a PV/DiG/BS energy
management system of grid-tied micro-grid of a dental college and hospital [10]. The
energy consumption data was taken from the audit report tabulated in Table 1 and
PV data for GHI is taken from the website of NREL (National Renewable Energy
Laboratory), United States [11–13].

2 Architecture

2.1 Specifications of Components

The components used in the two models are as follows:

Scenario 1 (Existing Model): DiG and Grid.
Scenario 2 (Proposed Model): PV, DiG, BS, and Grid (Fig. 1).
Optimal Energy Management in Hybrid Microgrid … 309

Table 1 Load data of the college and its hospital

Departments Annual load
Av (kW) Peak (kW)
Conference rooms 5.3 10.73
Lounge 2.72 9.68
Corridors 0.18 0.47
Cast room 0.23 0.47
Chiller plant energy including cooling fans and pumps 7.85388 10.58
Lifts/elevators 0.70586 1.44
Reception/waiting area 2.3 5.5
Nurse/work station 5.26 11.75
Dermatology 4.41 17.97
Domestic hot water 0.82002 2.38
Elevators 0.70586 1.44
Employee health/occupational health, medical staff development 3.72 5.9
Exam rooms 7.18 13.83
Exhaust fans 2.57801 4.44
Eye procedures, hospice, medical transcription 2.64 7.44
Head and neck surgery 2.57 6.71
Hemo/oncology and infusion center 4.47 7.7
Humidification 3.85274 5.45
Imaging/linear accelerator 3.9 10.97
Internal medicine 5.57 12.41
Offices 3.08 7.03
Labs/medical storage 5.56 10.33
Material services/biomed 0.83 1.62
Mechanical/electrical rooms 3.19 4.29
Medical imaging 0.3672 1.35
Medical secretaries, health education admin 4.85 9.03
Neurology 1.3 6.17
Oncology 7.37 11.24
Miscellaneous electric loads 18.4317 27.06
Orthopaedics & podiatry 0.51 1.87
Paediatrics 1.91 2.8
Pharmacy and medical-legal 1.91 2.8
Physical/occupational therapy pediatric rehab 5.35 10.22
Post-op 1.83 2.52
Preheat 5.19406 7.88
(continued)
310 M. Rizvi et al.

Table 1 (continued)
Departments Annual load
Av (kW) Peak (kW)
Prep area, pre-op 6.91 13.69
Primary care 0.66 1.66
Procedure/operating rooms 3.96 11.93
Public waiting area 0.33 0.54
Pumps for hot water and heat recovery 0.30822 1.32
Reception 1.83 4.82
Reheat and heating 20.1161 28.2
Return fans 1.33181 3.3
Shipping/receiving, tele equip. room, injection room 3.26 7.8
Steam piping losses 10.8447 14.25
Supply fans 5.54985 7.6
Surgery-OB/Gyn 2.98 5.27
Urology, neurology, Eeg/dleep lab 1.96 4.45

Fig. 1 Model of grid-tied

microgrid

2.2 Model Description

In the scenario, the dental college and its hospital are connected to the grid and
in addition, a 500 kVA diesel generator is installed in the premises to meet the
load during load shedding and blackout at the grid. In scenario 2 it is proposed to
incorporate PV system and battery energy storage to the DC bus.
Operational Mode: The system consists of PV panels for harnessing solar energy,
DiG, batteries for energy storage, inverters for delivering power to ac load, and
controller. The DiGs produces AC power therefore, connected to AC bus whereas
PV produce DC power which requires an inverter to convert DC to AC. At the time
of low solar power generation, the load demand is met from batteries instead of DiG
due to environmental and economic reasons as there will be no fuel consumption and
Optimal Energy Management in Hybrid Microgrid … 311

harmful emissions. DiG is used as a power back up only when PV, Grid and Batteries
are not able to meet the load demand. Load following strategy is used during DiG
operation. It produces enough power to meet the demand it does not charge the
battery. Load following tends to be optimal in systems with a lot of renewable power
that sometimes exceeds the load.
Economic Modeling: Economics plays a crucial role in which simulations in
HOMER are carried out to find the optimal configuration of the system to mini-
mize the total net present cost (NPC) [14]. For a number of prospective design
configurations HOMER carries out thousands of simulations. On examination of
every design, it selects the one which meets the load at the least cycle cost consid-
ering all the system constraints. Sensitivity analysis and optimization are performed
by HOMER across all stated components and their resources, system constraints,
and data sensitivity of exogenous variables over a certain range, and technical and
cost parameters. The best-suited hybrid MG for Dental College and its hospital is
compared with the conventional option of the grid and DiG.
Grid Outage: Load shedding from the grid is scheduled from 0900 to1100 h
and from 1400 to 1600 h every day, and the utility grid is out randomly at other
times. Annual Grid outages of 55 days are considered. Sensitivity analysis is done
by considering mean repair time for grid of 2 and 5 h and the fuel price of $0.97,
$1.1, and $1. 35.

3 Load Data

The load of the building is tabulated in Table 1; Fig. 2.

Fig. 2 Annual load frequency of dental college and its hospital

312 M. Rizvi et al.

Fig. 3 Average monthly load variation of college and its hospital

Monthly Variation of load: Monthly variation of load for one year is shown in
Fig. 3.

3.1 Global Horizontal Irradiance (GHI)

Daily Average of Solar radiation data at a location of 28°46.3 N latitude and 77°30.4
E longitude was taken from NREL website. The average annual clearness index for
this location and the average daily radiation are 0.679 and 5.19 kWh m−2 respectively
(Fig. 4).

4 Cost Summary of the Hybrid System

The Annual Operating cost reduces by 21.05% from $143,919 to $113,622 due to the
free availability of solar energy and its transformation into electricity by the process
of photovoltaic does not require any fuel.
Optimal Energy Management in Hybrid Microgrid … 313

Fig. 4 Global horizontal irradiance (GHI)

Table 2 Cost summary of

Existing model Proposed model
the hybrid system
NPC (Net present cost) $736,300 $859,900
Operating cost/year $143,919 $113,622

4.1 Cost of Energy (COE)

The results of the cost of energy by considering the sensitivity of the variation in fuel
price and mean grid repair time during blackout for the existing model and proposed
model are tabulated in Table 2.

5 Results

In this section, the results of economical, technical, and environmental analysis for
existing model and proposed model are presented and discussed.

5.1 Diesel Fuel Consumption

The diesel fuel consumption in existing model decreased by 49.71% from 83,587 to
42,031 L/year
Annual fuel consumption: Average monthly variation of diesel fuel is shown in
Fig. 5.
Emissions: The emission reduces due to PV penetration and the result is tabulated
in Table 5
Battery storage system: Deep-cycle batteries (lithium-ion and lead-acid batteries)
are used since with continuous use their life cycle and efficiency are uncompromised.
Towards the end of life, lithium-ion batteries have higher energy density as compared
314 M. Rizvi et al.

Fig. 5 Annual fuel consumption of diesel generator

Fig. 6 SOC of battery

to a lead-acid battery. The efficiency of Li-ion batteries and lead-acid batteries are
78% and 70% respectively [15]. However, the cost of lithium-ion batteries is double
the cost of lead-acid batteries. In this paper, lead-acid batteries are used. The battery
energy storage comprises of 10 strings each of size 30 connected in parallel. SOC of
the battery reaches a minimum value of 40% only in June and October (Fig. 6).

6 Conclusion

In this paper distributed generation system connected to the grid is optimized for
dental college and hospital. Analysis of the study shows that the inclusion of PV
systems reduces carbon emissions and reduces Levelized cost of energy (LCOE).
It has been observed that there is a need to meet the demand for peak load from
the utility grid and it is found to be financially favorable. This was accomplished
by optimization with HOMER. Thousands of design options were evaluated from
the different capacities of the combination of chosen battery sources, converters,
and capital costs by taking into account the technical specifications. The results are
tabulated in Tables 2, 3, 4 and 5.
The main objective of this paper was to ensure the potential of integrating PV and
diesel generators with the utility grid with battery storage to supply critical loads in
a Dental Hospital. The DiG power was utilized during an outage at utility grid or for
periods of low sunshine or at night when renewable energy input is low or zero or
during high load demand. To meet the peak load demand the DiG is optimally sized
with 80% of its rated power. The PV module’s size is based on its characteristics,
Optimal Energy Management in Hybrid Microgrid … 315

Table 3 COE (Cost of energy) for sensitivity in diesel price and grid mean repair time
Grid mean repair Diesel fuel price Cost of energy (COE) in $ % reduction in cost
time (in $) Scenario-I Scenario-II
2h 0.97 0.175 0.154 13.63
1.1 0.18 0.158 13.9
1.35 0.191 0.165 15.74
5h 0.97 0.18 0.157 14.63
1.1 0.186 0.161 15.52
1.35 0.197 0.17 15.88

Table 4 Diesel fuel consumption

Components Scenario-I Scenario-II
Diesel generation Grid Diesel generation Grid PV
Rated capacity (kW) 500 500 500 500 300
Production (%) 27.3 72.7 15.2 44.8 40
Fuel consumption (L/yr) 83,587 – 57,075 – –

Table 5 Emissions from diesel generators

Emissions Carbon Carbon Unburned Particulate Sulphur Nitrogen
dioxide monoxide Hydrocarbon matter dioxide oxide
(kg/year) (kg/year) (kg/year) (kg/year) (kg/year) (kg/year)
Scenario-I 718,870 293 15.9 1.25 2169 4403
Scenario-II 470,681 147 7.99 0.630 1564 2446

the voltage of the system, and the daily balance of energy, between the load and
PV array. Sizing of the battery was done in such a way it could supply load by
considering both seasonal and daily energy deficits. Apart from cutting the cost of
energy (COE) hybrid system’s maintenance cost is lower. The study reveals that
for a specified hybrid configuration, with an increase in PV capacity the number of
hours of operation of DiG decreases. Compared to only DiG system, the saving in
fuel is 49.71% in Hybrid PV microgrid. The hybrid system is more environmentally
friendly because CO2 emission is reduced to more than 34.52%.

References

1. Morf H (2011) The stochastic two-state cloud cover model STSCCM. Sol Energy 85(5):985–
999
2. Nejat P, Jomehzadeh F, Taheri MM, Gohari M, Muhd MZ (2015) A global review of energy
consumption, CO2 emissions and policy in the residential sector (with an overview of the top
316 M. Rizvi et al.

ten CO2 emitting countries). Renew Sustain Energy Rev 43:843–862

3. Amrollahi MH, Bathaee SMT (2017) Techno-economic optimization of hybrid photo-
voltaic/wind generation together with energy storage system in a stand-alone micro-grid
subjected to demand response. Appl Energy 202:66–77
4. Anglani N, Oriti G, Colombini M (2017) Optimized energy management system to reduce fuel
consumption in remote military microgrids. IEEE Trans Ind Appl 53(6):5777–5785
5. Shen J, Jiang C, Liu Y, Qian J (2016) A microgrid energy management system with demand
response for providing grid peak shaving. Electr Power Compon Syst 44(8):843–852
6. Jiang Q, Xue M, Geng G (2013) Energy management of microgrid in grid-connected and
stand-alone modes, vol 28(3):3380–3389
7. Pascual J, Barricarte J, Sanchis P, Marroyo L (2015) Energy management strategy for a
renewable-based residential microgrid with generation and demand forecasting. Appl Energy
158:12–25
8. Kanchev H, Lazarov V, Francois B (2012) Environmental and economical optimization of
microgrid long term operational planning including PV-based active generators. In: 15th inter-
national power electronics motion control conference exposition EPE-PEMC 2012 ECCE
Europe, pp 1–8
9. Minchala-Avila LI, Garza-Castanon L, Zhang Y, Ferrer HJA (2016) Optimal energy manage-
ment for stable operation of an islanded microgrid. IEEE Trans Ind Inf 12(4):1361–1370
10. Wu J, Yan J, Jia H, Hatziargyriou N, Djilali N, Sun H (2016) Integrated energy systems. Appl
Energy 167:155–157
11. Sheppy M, Pless S, Kung F (2014) Healthcare energy end-use monitoring healthcare energy
end-use monitoring
12. Oulis Rousis A, Tzelepis D, Konstantelos I, Booth C, Strbac G (2018) Design of a hybrid AC/DC
microgrid using HOMER pro: case study on an islanded residential application. Inventions
3(3):55
13. Hu SC, Chen JD, Chuah YK (2004) Enery cost and consumption in a large acute hospital. Int
J Archit Sci 5(1):11–19
14. Arcos-Aviles D, Pascual J, Guinjoan F, Marroyo L, Sanchis P, Marietta MP (2017) Low
complexity energy management strategy for grid profile smoothing of a residential grid-
connected microgrid using generation and demand forecasting. Appl Energy 205:69–84
15. Chalise S, Sternhagen J, Hansen TM, Tonkoski R (2016) Energy management of remote
microgrids considering battery lifetime. Electr J 29(6):1–10
Optimal Placement of DG with Battery
Energy Storage Using CPLS
and Combined Dispatch Strategy

Bharat Singh and Ashwani Kumar Sharma

Abstract This paper presents a novel application of General Algebraic Modeling

System (GAMS) and MATLAB interfacing for optimal placement and size of distri-
bution generation (DG) with battery. In this paper, the main contribution is: (i) optimal
position of DG based on combined power loss sensitivity (CPLS) method, (ii) optimal
placement of battery energy storage using combined dispatch strategy, (iii) optimal
size of DG and battery have been carried out in such a way to minimize the total
power loss without violating the constraints, (iv) there are two types of DGs that
have been considered. The impact of DG with and without battery placement on
total power loss and voltage improvement have been carried out. Voltage profile,
real and reactive power flow pattern, total power loss, fuel cost, total cost, state of
charge (SOC) status of battery, and power obtained by DG with the battery have been
determined. The result shows the importance of DG with battery size and location.
The results are obtained on IEEE-33 bus radial distribution test system and compare
with another existing method also.

Keywords Distribution generation · Battery · GAMS · Radial distribution system

1 Introduction

The distributed generations (DGs) with renewable sources and its penetration into the
distribution network have introduced a lot of challenges to the distribution network
operator for better management and planning. In this context, the distribution system
analysis needs to be analyzed with DGs. The storage devices have become part of
the AC/DC micro-grids with the renewable sources to meet out the deviations in
the power requirement during their intermittency. Therefore, there is utmost need to

B. Singh (B) · A. Kumar Sharma

Department of Electrical Engineering, National Institute of Technology Kurukshetra, Kurukshetra
136119, Haryana, India
e-mail: [email protected]
A. Kumar Sharma
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 317

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_30
318 B. Singh and A. Kumar Sharma

analyze such hybrid power system for optimal dispatch strategy satisfying the criteria
of loss minimization.
In this paper work, we have considered a 33-bus radial distribution for analysis
that consists of diesel generator and battery storage. For simplicity, current modeling
is used, for the expressions of input and output power.
A lot of research paper based on the Particle Swarm Optimization (PSO) algorithm
are used for optimal location as well as size of distribution generations [1]. The
single-variable and multivariable approach have been done on the basis of the PSO
algorithm. Murty and Kumar [2] considered as the total cost of power loss. However,
in [2] the optimal location and size of different types of DGs at different buses by
minimizing the power distribution loss. Dufo-López and Bernal-Agustín [3] carried
out with the multi-objective evolutionary algorithm along with a genetic algorithm
(GA). A lot of analysis work have been drained the realm of best placement of DGs
[4–6], and also the various optimization tools have been used.
The direct approach for distribution load flow solution has been utilized by using
two topological characteristic matrices represented by Teng [7]. The various methods
and comparisons have been done for the location and sizing of DG using combined
PLS [8], but the battery with DG is not considered. In the paper [9], Babu et al. do
not consider the battery storage.
In this paper, the DG size and location with and without storage device has been
determined. The combined power loss sensitivity approach has been developed for
finding the setting of DGs. The MATLAB and GAMS interfacing has been utilized for
obtaining the optimal solution. On the other hand, the battery energy storage device
has been integrated with DG in such a way to meet out the power requirement. The
objective function is to minimize the total power loss without violating constraints.
In this paper, the two cases are taken: in the first case—DGs at unity pf, 0.9
lagging, power factor have been considered. In the second case, the DG with battery
storage has been consider.

2 Mathematical Model and Dispatch Strategy

GAMS software system is meant for modeling linear, nonlinear, and mixed whole
number optimization issues. In this paper, the two objectives are formulated as:
• Location of DGs with and without battery energy storage system (BES)
• Position of DGs with and without a battery energy storage system.
The MATLAB is used for calculation of combined PLS for the 33-bus radial
distribution system and GAMS tool for optimized the size of DGs with and without
BES.
Optimal Placement of DG with Battery Energy Storage Using CPLS … 319

Fig. 1 Equivalent circuit

model of radial distribution
system

2.1 System Modeling

The network consists of the load for 24 h, diesel generators and batteries. The
schematic representation of the radial distribution power system is shown in Fig. 1.

2.2 Mathematical Model Formulation

1. Optimal load flow for integrated renewable sources

Optimal load flow is developed by minimizing the whole power loss that is given by
the equations.
Minimize


k
OF = PLi j (1)
i, j=1

Subject to


n
    
Pik = Pik − Pdik = Vik V jk G ikj cos δik − δ kj + Bikj sin δik − δ kj ,
j=1

i ∈ SB and k ∈ ST (2)


n
    
Q ik = Q ik − Qdik = Vik V jk G ikj sin δik − δ kj − Bikj cos δik − δ kj ,
j=1

i ∈ SB and k ∈ ST (3)

where ∀i = 1, 2 . . . nb, ∀i j = 1, 2 . . . nl, nb is a number of buses and nl is the

total number of line ∀k = 1, 2 . . . 24 is the set of time. Pik is the total active power
injected at the ith bus for kth hours. Pdik is the total active power demand at ith bus
for kth hours. The limits of capacity on the distribution system are:
min
PGi ≤ PG I ≤ PGi
max
(4)
320 B. Singh and A. Kumar Sharma

Ri ≤ Q Ri ≤ Q Ri
Q min max
(5)

Limits for voltage and angle

Vimin ≤ Vi ≤ Vimax , i ∈ SB (6)

δmin kj ≤ δi ≤ δmax kj , ∀i = 1, 2 . . . nb (7)

 k  k   k k k     
Pl  =  P  = V V G cos δ k − δ k + B k cos δ k − δ k − (V k )2 G k 
ij i j ij i i ij i i i ij

≤ Plkmax , l ∈ SL (8)

where SB represent the set of Bus i.

2. The combined power loss sensitivity method (CPLS)
This method [8] considers for which normalized sensitivity value is less than 0.0050
is chosen as the setting for placement DG.
3. Battery model
The battery storage works as backup storage. It can either be charged as well as
discharged according to the dispatch strategy [3]. The state of charge has been
calculated for each battery.
 
1
SOC(k + k) = SOC(k) · (1 − δk ) + Pch · ηch − Pdis · · k (9)
ηdis

Constraints:

SOCimin (k) ≤ SOCi (k) ≤ SOCimax (k) (10)

Pch_min ik ≤ Pch ik ≤ Pch_max ik

Pdis_ min ik ≤ Pdis ik ≤ Pdis_ max ik (11)

where SOC is state of charge in per unit, SOCimin (k) and SOCimax (k) is minimum and
maximum value of SOC at kth time, respectively. Pch ik and Pdis ik is the charging and
discharging power of battery at the kth time, respectively. ηch and ηdis is the charging
and discharging efficiency of the battery, respectively.
Optimal Placement of DG with Battery Energy Storage Using CPLS … 321

2.3 Dispatch Strategy

In this model, the DG and storage device as a battery has been with an objective to
minimize the total power loss and cost of the system. In order to maximize the reli-
ability and yet maintain the cost along with minimum power loss, different dispatch
strategies like as load following strategy, cycle charging strategy, or combined
strategy can be used [3]. In this paper, the combined strategy is implemented for
the location of energy storage device as batteries.

3 Algorithm Used

The following methodology has been used

• The location of DGs is based on the combined PLS using MATLAB [9].
• The position of battery energy storage is based on the dispatch strategy using
MATLAB [3].
The following steps are used:
Step 1: Read the system data for IEEE 33-bus radial distribution system.
Step 2: Run the load flow program for the 24 h dynamic load data and obtain the
base case total power losses.
Step 3: Obtain the PLS at each node for 24 h. Select the node having the highest PLS
factor.
Step 4: Obtain the position of batteries at each node for 24 h using dispatch strategy,
in the form of a matrix (bus, hours).
Step 5: Transfer all control parameter from MATLAB to GAMS using interfacing in
a matrix of (bus, hours).
Step 6: Solve the optimal load flow for optimal size of DGs with and without batteries,
and calculate the SOC using MINLP solver to minimize the total power loss.
Step 7: Transfer the objective variables form GAMS to MATLAB.
Step 8: Print the results.

4 System Data

In this paper, the 100 kVA 12.66 kV, IEEE 33-bus radial distribution system has
been taken; it has 33 buses and 32 branches. The load with 24 h variation has been
taken. In this paper, the battery having a nominal capacity of 170 Ah, operation and
322 B. Singh and A. Kumar Sharma

maintenance cost of 4.64 (e/hour), acquisition cost (e) of 464, max current (I max )
of 34, depth of discharge of 0.03, and no. of the cycle of 1110 has been used for
analysis purpose. In this paper, diesel generator having operation and maintenance
cost of 0.236 (e/hour), acquisition cost (e) of 3237.6, and lifespan of 10,000 in
hours has been used for analysis purpose.

5 Result and Discussions

The energy storage system incorporated with diesel generation in radial distribution
system under consideration has been designed by GAMS optimization tool.

5.1 Case 1: DGs Without an Energy Storage Device

The 33-bus radial distribution of a 100 kVA, 12.66 kV, 33-bus and 32 branches are
taken. The total average load of 3.0834 + j*1.909 MVA is taken from the daily load
profile.
The comparison of direct approach method [7] without DGs and proposed method
with DGs for voltage profile is shown in Fig. 2. The voltage profile has been improved
considerably in the proposed method.
From Fig. 3, the highest PLS is obtained for at bus number 8, 24, and 25.
The optimal location and placement for DGs type-1 (at unity power factor) and
type-3 (at lagging power factor) are given in Table 1, and the selection criteria are
based on PLS.
The higher value of PLS indicates the higher losses, for selection of single DG,
the losses are larger, and it has been decreasing simultaneously with increasing the
number of DGs. At bus number 25, the size of DG is 684.3 kW, but the losses are
higher in this case, and on the other hand at bus number 8, 24, and 25, the losses are
reduced.

Fig. 2 Comparison of voltage profile for proposed with DGs and direct approach method
Optimal Placement of DG with Battery Energy Storage Using CPLS … 323

Fig. 3. 33-bus radial distribution system PLS profile

Table 1 Result of comparison of existing method and purposed method with DG

Method Best location Best DGs size Losses (kW)
(kW) Without DG With DG
PSO [2] 6 2590 211.20 110.10
ELF [2] 6 2600 211.20 111.10
NM [8] 6 2494.80 210.98 111.14
CPLS [8] 8 1800 210.98 111.14
Proposed DG 8, 24, 25 [1424.9, 448.84, 210.98 79.356
type-1 420.22]
Proposed DG 8, 24, 25 [1499.4, 46.73, 210.98 + j*143.02 57.834 + j*43.57
type-3 418.68]

5.2 Case 2: DGs with Batteries 33 Bus System

The second part of this paper is to the analysis of battery energy storage with DGs
to share the real power injection as well as to minimize the total power loss. The
location and sizing of the battery energy storage are based on the combined load
cycling method [3]. In Table 2, the position of DGs type-3 with battery energy
storage (BES) is 8, 20, and 24 is based on the highest value of PLS.
In Fig. 4, the SOC profile is shown at bus number 8, 20, and 24, and the DGs are
placed at the bus no 8, 20, and 24.

6 Conclusions

This paper has conferred a unique application of GAMS computer code to perform
the dispatch strategy with the best location of decigram at the side of battery energy
device. The two-phase scheme has been proposed for DG and battery allocation in the
distribution system. In the first phase, the candidate nodes are selected as the best DG
324 B. Singh and A. Kumar Sharma

Table 2 Result for DG with battery storage system

Without DG DG + battery
DG type-1 DG type-3
DG location – 24, 25 8, 20, 24
DG size (kVA) – Location Size (kW) Location Size (kW)
24 253.22 8 482.23
31 105.09 20 262.88
– 24 90.713
Total power loss 3080.5 761.19 835.83
(kW)
Fuel used – 2648.9 L 2.057e + 05 L
Total fuel cost – 7946.7 e 6.1709e + 05 e
Installation cost of – 6978.1 e 6978.1 e
DG + battery +
regulator
Total cost – 14,925 e 6.2407e + 05 e

Fig. 4 Result and comparison for SOC at node 8, 20, and 24 for DG type 3

setting along with battery placement, and in second phase, best DG with battery size
are computed. The two techniques are used. In first, combined power loss sensitivity
(CPLS) has been used for optimal DG location, and in second, the combined dispatch
strategy has been used for placement of the battery. The proposed method is tested on
IEEE 33-bus test system. In terms of voltage profile and total power loss reduction,
the purposed method has superior results than existing methods.
Optimal Placement of DG with Battery Energy Storage Using CPLS … 325

References

1. Prakash DB, Lakshminarayana C (2016) Multiple DG placements in distribution system for

power loss reduction using PSO algorithm. Procedia Technol 25:785–792. https://doi.org/10.
1016/j.protcy.2016.08.173
2. Murty VVSN, Kumar A (2015) Optimal placement of DG in radial distribution systems based
on new voltage stability index under load growth. Int J Electr Power Energy Syst 69:246–256.
https://doi.org/10.1016/j.ijepes.2014.12.080
3. Dufo-López R, Bernal-Agustín JL (2005) Design and control strategies of PV-diesel systems
using genetic algorithms. Sol Energy 79(1):33–46. https://doi.org/10.1016/j.solener.2004.
10.004
4. Al Abri RS, El-Saadany EF, Atwa YM (2013) Optimal placement and sizing method to improve
the voltage stability margin in a distribution system using distributed generation. IEEE Trans
Power Syst 28(1):326–334. https://doi.org/10.1109/TPWRS.2012.2200049
5. Mistry KD, Roy R (2014) Enhancement of loading capacity of distribution system through
distributed generator placement considering techno-economic benefits with load growth. Int J
Electr Power Energy Syst 54:505–515. https://doi.org/10.1016/j.ijepes.2013.07.032
6. Borges CLT, Falcão DM (2006) Optimal distributed generation allocation for reliability, losses,
and voltage improvement. Int J Electr Power Energy Syst 28(6):413–420. https://doi.org/10.
1016/j.ijepes.2006.02.003
7. Teng J-H (2003) A direct approach for distribution system load flow solutions. IEEE Trans
Power Delivery 18(3):882–887. https://doi.org/10.1109/tpwrd.2003.813818
8. Murthy VVSN, Kumar A (2013) Comparison of optimal DG allocation methods in radial distri-
bution systems based on sensitivity approaches. Int J Electr Power Energy Syst 53(1):450–467.
https://doi.org/10.1016/j.ijepes.2013.05.018
9. Babu PV, Singh SP (2015) Optimal placement of DG in distribution network for power loss
minimization using NLP & PLS technique. Energy Procedia 90:441–454. https://doi.org/10.
1016/j.egypro.2016.11.211
Comparison Among Conventional
and Adaptive Gain Scheduling PID
Controllers for Stabilizing an Inverted
Pendulum System

Vikram Chopra, Sunil K. Singla, and Lillie Dewan

Abstract This paper presents a comparison between conventional and adaptive gain
scheduling PID controllers for the stabilization an inverted pendulum mounted on a
cart. The gains of the adaptive PID controller are designed to be a function of time
as well as error that can vary within a specified range depending upon the transient
and steady-state part of the response. The performance of adaptive gain scheduling
PID controller is compared with conventional PID controller given in the literature.
Simulation results show improved performance of the proposed scheme over the
conventional scheme.

Keywords Inverted pendulum · Stabilization · Conventional PID controller ·

Adaptive gain scheduling PID controller

1 Introduction

The inverted pendulum is an under-actuated, nonlinear and non-minimum phase

system with lesser control inputs than the degrees of freedom [1]. The system is a
classical benchmark problem for designing controllers in control theory (Boubakar
2012) and resembles the control of many systems such as, robots, Segways and
space rocket launchers. Due to the gravity effect, controlling an inverted pendulum
is a challenging task. The control of an inverted pendulum is divided into three parts,
i.e. the swing up control [6, 7], the stabilization control [3, 7] and the tracking control

V. Chopra (B) · S. K. Singla

Department of Electrical and Instrumentation Engineering, Thapar University, Patiala, Punjab
147004, India
e-mail: [email protected]
S. K. Singla
e-mail: [email protected]
L. Dewan
Department of Electrical Engineering, National Institute of Technology Kurukshetra,
Kurukshetra, Haryana 136119, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 327

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_31
328 V. Chopra et al.

[4]. However, from the application point of view, stabilization and tracking control
are the most commonly discussed.
A number of control strategies for stabilizing an inverted pendulum have been
reported in the literature. Lin et al. (1996) applied a linear state feedback controller
for obtaining a high value of gain margin to stabilize an inverted pendulum. A
linear quadratic regulator to keep the inverted pendulum in the upright position has
been presented by Prasad et al. [10]. Wai and Chang [11] proposed a sliding mode
controller for stabilizing an inverted pendulum system. Intelligent control of inverted
pendulum using an interval type 2 fuzzy PID controller has been reported by Nagar
et al. [8]. Zhao and Spoong [14] presented the hybrid control for global stabilization
of the cart-pendulum system while the energy-based control of the inverted pendulum
has been given by Fantoni et al. (2000).
Although a number of control techniques have been implemented, PID controller
is still the most commonly used controller because of its simple structure, easy
implementation and fewer numbers of tuning parameters [2]. However, very few
researchers have worked on the PID control of an inverted pendulum system [5, 10,
12] Also, as far as the author’s knowledge, a little or no efforts have been made in
designing PID controllers based on gain scheduling method for the stabilization and
tracking control of the inverted pendulum system. In the present work, adaptive gain
scheduling PID controllers have been designed for the stabilization of an inverted
pendulum system using a two-loop structure.
The paper is organized as follows: Sect. 2 will describe the state space model
of the inverted pendulum system. The designing of adaptive gain scheduling PID
controllers will be given in Sect. 3 followed by the simulation results in Sect. 4. The
stability analysis of the proposed controller will be discussed in Sect. 5, and finally,
the paper is concluded.

2 The Inverted Pendulum System

The inverted pendulum mounted on a cart is shown in Fig. 1. Stabilization of the

inverted pendulum in upright position depends upon the horizontal displacement of
the cart which in turn depends upon the applied force F. The parameters related to
the inverted pendulum are defined as follows:
M = 1 = Mass of the cart in kg.
m = 0.1 = Mass of the pendulum in kg.
l = 0.3 = Distance from the cart to the mass center of the pendulum in m.
g = 9.8 = Acceleration due to gravity in m/s2 .
x = Position of the cart from the reference.
(x, z) and (ẋ, ż) = Position and velocity of the cart in the xoz coordinates.
Comparison Among Conventional and Adaptive Gain Scheduling PID … 329

Fig. 1 Inverted pendulum

system

  
(x P , z P ) and (ẋ P , ż P ) = Position and velocity of the cart in x o z coordinates.
The Lagrange’s equations of the inverted pendulum system are given as [12]:
 
d ∂L ∂L
− = Fx (1)
dt ∂ ẋ ∂x
 
d ∂L ∂L
− =0 (2)
dt ∂ θ̇ ∂θ

where L is the inverted pendulum is defined as:

L=K−P (3)

K is the kinetic energy and P the potential energy of the system.

1
K = (M ẋ 2 + m(ẋ 2P + ż 2P )) (4)
2

P = mgz p (5)

Here x p = x + l sin θ and z p = z + l cos θ .

Using Eqs. (1)–(5)

(M + m)ẍ + ml cos θ θ̈ − ml sin θ θ̇ 2 = Fx (6)

330 V. Chopra et al.

cos θ ẍ + l θ̈ − g sin θ = 0 (7)

Defining x1 = x, x2 = ẋ, x3 = θ and x4 = θ̇ the state equations of the inverted

pendulum system can be written as [12]:

ẋ1 = x2 (8)

−mg cos x3 sin x3 + ml sin x3 x42 + F

ẋ2 = (9)
M + m sin2 x3

ẋ3 = x 4 (10)

−ml cos x3 sin x3 x42 − cos x3 F + (M + m)g sin x3

ẋ4 = (11)
Ml + ml sin2 x3

Equations (9) and (11) have been used in developing the Simulink model of the
inverted pendulum system.

3 Proposed Adaptive Gain Scheduling PID Controllers

Due to its simplicity and unexceptional capability of solving a number of practical

control problems, a PID controller is the most widely used controller. A conventional
PID controller is given by Astrom and Hagglund [2]:

de(t)
u(t) = kp e(t) + ki e(t)dt + kd (12)
dt

Here u(t) is the control signal. kp , ki and kd are the proportional, the integral and
the derivative gains of the controller, respectively. The gains of the PID controller are
fixed and can be obtained using different tuning methods such as Zeigler Nichols and
relay autotuning. However, for a nonlinear system, a fixed gain PID controller cannot
produce satisfactory performance under all operating regions. Hence, adaptive or
intelligent controller is preferred over the conventional controller in case of nonlinear
systems. In this paper, the adaptive gains of the PID controller have been scheduled
using time as well as error.
Comparison Among Conventional and Adaptive Gain Scheduling PID … 331

3.1 Time Adaptive Gain Scheduling PID Controllers

In the proposed scheme, the PID controller gains are designed to be a function of time,
which varies within the specified range. The functions of kp , ki and kd are designed
to fulfill the requirements of the system and make the system more efficient.
Since the value of kp should be higher during the initial part of the system response
to reduce the rise time of the system and low during the later stages to overcome
overshoot and the settling time. Therefore, the proportional gain is defined as:

kp (t) = kp min + (kp max − kp min )e−kt (13)

where k is a constant for deciding the rate at which kp varies between minimum and
maximum values. During the transient part of the response, the value of error is high;
hence, a large value for kp is required. So for the initial phase (i.e., t = 0) e−kt is
approximately one therefore, kp = kp max . As time t → ∞ e−kt → 0. So, kp = kp min .
Similarly, the main function of the integral term is to remove the offset which
comes in the steady-state part of the response. Hence, a large value of ki is required
as t → ∞ and minimum value is required during the initial stage. Therefore, the
integral gain can be defined as:

ki (t) = ki max − (ki max − ki min )e−kt (14)

While the derivative term is defined as:

kd (t) = kd min + (kd max − kd min )e−kt (15)

Since the derivative term is used to reduce oscillations, which are more prone
during the initial stage of the response; therefore, a high value for the derivative gain
is required during the initial part of the response, while during the steady-state part
of the response, a small value is sufficient.

3.2 Error Adaptive Gain Scheduling PID Controllers

During the transient part of the system response, the error is high. Therefore, a large
value of k p is required to reduce the rise time of the system. The value of k p should
be small during the steady-state part to overcome overshoot and the settling time.
Therefore, the proportional gain is defined as:

kp (t) = kp max − (kp max − kp min )e−ke(t) (16)

The integral gain can be defined as:

332 V. Chopra et al.

Table 1 Operating range of

Gain Max Min
gains for adaptive gains
scheduling PID controllers (a) Controller 1
kp 36 24
ki 3.6 1.5
kd 5 3
(b) Controller 2
kp −2.4 −3
ki −1 −1.2
kd −2 −4.8

ki (t) = ki min − (ki min − ki max )e−ke(t) (17)

Since a large value of ki is required as e(t) → 0 and minimum value is required

during the initial stage of the response.
While the derivative term is defined as:

kd (t) = kd max − (kd max − kd min )e−ke(t) (18)

A high value for the derivative gain is required when the value of error is high,
while during the steady-state part of the response where the value of error is low, a
small value is sufficient. The minimum and maximum values of PID gains are given
in Tables 1 and 2 for angle and position control, respectively.
These designed values of k p , ki and kd have been implemented on the inverted
pendulum system. The Simulink block diagram of the inverted pendulum system is
shown in Fig. 2.
As shown in Fig. 2, the control structure is a two-loop structure with one loop
for the angle control while the second loop for position control. The control actions

Fig. 2 Control structure of inverted pendulum

Comparison Among Conventional and Adaptive Gain Scheduling PID … 333

generated by the two controllers are opposite in sign because of the opposite move-
ment of the cart and the pendulum. The simulation results corresponding to the
position and angle control are presented in the next section.

4 Simulation Results

The simulation results are given in Fig. 3 for the stabilization of the inverted pendulum
without considering the effect of disturbance on the inverted pendulum system.
Figure 4 gives the simulation results considering the addition of outer disturbances
d1 = d2 = 20 sin(20π t) in Eqs. (9) and (11).
From Figs. 3 and 4, it can be revealed that the adaptive gain scheduling PID
controllers stabilize the inverted pendulum in a more effective manner, especially in
case of position control, than the conventional PID controllers proposed by Wang
[12].

0.4
Conventional PID Conventional PID
1.4 Time Adaptive PID Time Adaptive PID
0.3
Error Adaptive PID Error Adaptive PID
1.2
0.2
Angle (radian)

1
Position

0.8
0.1
0.6
0.4 0
0.2
-0.1
0
-0.2 -0.2
0 10 20 30 40 50 0 10 20 30 40 50
Time (sec) Time (sec)
(a) Position control (b) Angle Control

Fig. 3 Stabilization without disturbances

0 10 20 30 40 50
Conventional PID Conventional PID
Time (sec)
1.4 Time Adaptive PID 0.3 Time Adaptive PID
Error Adaptive PID Error Adaptive PID
1.2
0.2
1
Angle (radian)

0.1
Position

0.8
0.6 0
0.4
-0.1
0.2
-0.2
0
-0.2 0 10 20 30 40 50
0 10 20 30 40 50
Time (sec) Time (sec)

(a) Position control (b) Angle control

Fig. 4 Stabilization with disturbances

334 V. Chopra et al.

5 Stability Analysis

The stability analysis of the proposed system has been obtained using Lyapunov’s
stability criterion. The Lyapunov function is defined as a positive definite function
and is defined as the sum of kinetic energy and potential energy and is given as:

1 1
V = M(ẋ 2 ) + m(ẋ 2p + ż 2p ) + mgz p (19)
2 2
Therefore, according to Lyapunov’s stability criterion, for a system to be stable
V̇ ≤ 0

V̇ = M(ẋ ẍ) + m[(ẋ + l cos θ θ̇ )(ẍ − l sin θ θ̇ 2 + l cos θ θ̈)

+ (ż − l sin θ θ̇ )(z̈ − l cos θ θ̇ 2 − l sin θ θ̈ )] + mg(ż − l sin θ θ̇ ) (20)

Using Eqs. (6), (7) and (20), we get:

V̇ = Fx ẋ (21)

Therefore, for the system to be stable:

Fx ẋ ≤ 0 (22)

Equation (21) has been plotted as a function of time.

Figure 5 shows that the derivative of the Lyapunov’s function is tending toward
zero with the passage of time, which proves the stability of the proposed scheme.

4 6
Fx Fx
3 X dot X dot
4
2
1 2
Amplitude
Amplitude

0
0
-1
-2 -2

-3
-4
-4
-5 -6
0 5 10 15 20 25 0 5 10 15 20 25
time(sec) time(sec)
(a)Time adaptive gain scheduling PID (b)Error adaptive gain scheduling PID

Fig. 5 Stability analysis

Comparison Among Conventional and Adaptive Gain Scheduling PID … 335

6 Conclusion

In this paper, adaptive gain scheduling PID controllers are designed for the stabi-
lization and tracking control of the inverted pendulum mounted on a cart using a
two-loop control structure. The PID controller gains are designed as a function of
time as well as error that can vary within a specified range depending upon the tran-
sient and the steady-state part of the response. The performance of the proposed
scheme is compared with the conventional scheme given in literature, which has
fixed gains. Moreover, the robust nature of the proposed controllers has been tested
in the presence disturbance in the inverted pendulum system. Simulation results show
that the proposed scheme proves to be more suitable than the conventional scheme,
especially in case of position control.

References

1. Astrom KJ, Furuta K (1999) Swinging up a pendulum by energy control. Automatica 3:287–295
2. Astrom KJ, Hagglund T (1995) PID controllers: theory, design and tuning, 2nd edn. Instrument
Society of America, USA
3. Bloch AM, Leonard NE, Marsden JE (2000) Controlled Lagrangians and the stabilization of
mechanical systems I: the first matching theorem. IEEE Trans Autom Control 45:2253–2270
4. Chang LH, Lee AC (2007) Design of nonlinear controller for bi-axial inverted pendulum
system. IET Control Theory Appl 1:979–986
5. Ghosh A, Krishnan TR, Subudhi B (2012) Robust PID compensation of an inverted cart-
pendulum system: an experimental study. IET Control Theory Appl 6:1145–1152
6. Mason P, Broucke M, Piccoli B (2008) Time optimal swing-up of the planar pendulum. IEEE
Trans Autom Control 53:1876–1886
7. Muskinja N, Tovornik B (2006) Swinging up and stabilization of a real inverted pendulum.
IEEE Trans Ind Electron 53:631–639
8. Nagar A, Bardini M, Rabaie NM (2014) Intelligent control for nonlinear inverted pendulum
based on interval type-2 fuzzy PD controller. Alex Eng J 53:23–32
9. Nasir A (2007) Modeling and controller design for an inverted pendulum system. Master of
technology degree, Faculty of Electrical Engineering, Universiti Teknologi Malaysia
10. Prasad LB, Tyagi B, Gupta HO (2011) Optimal control of nonlinear inverted pendulum
dynamical system with disturbance input using PID controller & LQR. In: IEEE international
conference on control system, computing and engineering, pp 540–545
11. Wai RJ, Chang LJ (2006) Adaptive stabilizing and tracking control for a nonlinear inverted-
pendulum system via sliding-mode technique. IEEE Trans Ind Electron 53:674–692
12. Wang JJ (2011) Simulation studies of inverted pendulum based on PID controllers. Simul
Model Pract Theory 19:440–449
13. Wang JJ (2012) Stabilization and tracking control of X-Z inverted pendulum with sliding-mode
control. ISA Trans 51:763–770
14. Zhao J, Spong MW (2001) Hybrid control for global stabilization of the cart-pendulum system.
Automatica 37:1941–1951
A Review on Protection Devices
and Protection Techniques for DC
Micro-grid

Bholeshwar and R. S. Bhatia

Abstract DC micro-grid has interesting notable attention over last decade. DC

micro-grid has manifest dominance over AC micro-grid with respect to reliability,
efficiency, integration of renewable energy resources, etc. One of the major chal-
lenge problems of the DC system is proper protection systems and fast responsive
protective devices. In this review paper, different characteristics and operation-based
protective devices are explained and protection devices too. In this paper, differ-
ence between various protection device and method is also discussed which will be
helpful to select the reliable one. One of the main concerns in DC micro-grid is lack
of guideline and standard. Performance of the protection scheme and devices with
communication and without communication are discussed.

Keywords DC micro-grid · Protection techniques · Protection devices · Circuit

breaker · Fault

1 Introduction

Power generation via conventional fuel is linked with the various environmental
issues and low energy efficiency. With the use of non-conventional energy resources
as distribution generation (DG) can be showing the capable solution of these prob-
lems. Today, DC micro-grid is a reliable approach with majority of electronics loads
consuming DC power. There are various DC sources available such as photovoltaic
(PV) array, fuel cell, geothermal, biomass, and storage devices. As the integration of
distribution resource growing, their direct interconnection to the distribution system
has been difficult to manage due to their intermittency. DC micro-grid has been
proposed as an effective solution to associate DGs with distribution system [1].

Bholeshwar (B) · R. S. Bhatia

Electrical Engineering Deparetment, National Institute of Technology Kurukshtra, Kurukshtra,
Haryana, India
e-mail: [email protected]
R. S. Bhatia
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 337

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_32
338 Bholeshwar and R. S. Bhatia

Generally, DC micro-grid covers only a small geographical zone and distribution

line. Therefore, DC micro-gird is treated as the resistive network.
Unlike conventional energy system generators, micro-network systems use
converters (DC-AC, DC-DC, and AC-DC) to integrate light sources, wind energy,
fuel cells, and micro-turbines. Due to the nature of the source and converter, the
micro-grid system provides less physical inertia, which affects the stability of the
system during interference/fault. Therefore, the general performance parameters of
the DC micro-grid can be identified as, topologies (system and converter), control
strategies (voltage control, power sharing, maximum power point tracking (MPPT),
if PV or wind as DERs etc.), power management with energy storage devices, power
quality, communication protocols, physical and cyber-security, etc. This paper aims
to discuss all protection techniques and DC protective devices, which have been
suggested for the DC distribution network. The challenge of protective scheme and
protective devices is also discussed in this paper. There are various design criteria
for any protective systems which are describe below.
Reliability-protective system must be reliable it must be response when its require-
ment otherwise system should be stable. It must be preventing the unnecessary
tripping, such as for transient and noise. System should response during the fault
quickly.
Speed-protective system must be very fast, removing faults and restoring normal
condition rapidly.
Selectivity system should select the faulty part only. It should improve the
continuity of the service of the load and reduce the load disturbance during the
fault.

2 DC Micro-grid

Micro-grid is a small area power grid that can employ in separately or cooperatively
with various other small area power grid. The operation of using micro-grids is known
as distributed, dispersed, decentralized, district, or embedded energy production. A
micro-grid is a localized collection of electricity sources and loads that normally
operates connected to and synchronous with the traditional wide area synchronous
grid. Micro-grid is easily associated with various sources of distributed genera-
tion (DG), generally non-conventional resources such as PV array, wind energy
resources, biomass, battery, and super capacitor. A micro-grid can be connected in
grid-connected mode and isolated mode. In islanded mode, micro-grid can provide
the service to the rural areas with minimum cost and minimum power losses. In
grid-connected mode, micro-grid can support the main grid in many ways such as
voltage control and frequency control and can provide more flexible, control, and
reliability. Figure 1, shows the integrated micro-grid which is connected with the
distributed energy resources with the power electronics converters [2]. Communi-
cation signal plays a vital role in the DC micro-grid. Communication signal is fed
to the supervisor system with compute the sources voltage and current which are
A Review on Protection Devices and Protection Techniques … 339

Fig. 1 Integrated micro-grid model

provided to the load and control the output signal of the power electronics converters
and compare with the load profile. Depending upon the type of connected mode,
it may be grid connected form or may be islanding form, needful control plan and
protection schemes are compulsory.

3 DC Protection Devices

Recently, the traditionally accessible protection devices for DC micro-grids are fuses
and circuit breakers, but today there are various protection devices such as seniors,
directional element, protection relays, fuses, and solid-state circuit breaker. In this
section, researcher describes the following protection devices.
340 Bholeshwar and R. S. Bhatia

3.1 Sensors

Nowadays, all power converters use MOSFET or insulated gate bipolar transistor
(IGBT)-based solid-state devices. These devices have high switching frequency, high
current carrying, and voltage withstanding capability. By use of additional circuits,
these devices can improve the short-circuit current withstanding capability. This
reduces voltage transients during turn-off and improves system stability. Measure-
ment errors may lead to the failure of fault protection systems, and it may cause phys-
ical damage of DC micro-grid components. Generally, the measurement devices are
Hall-effect sensors, current transducers, direct current transformers, etc. For example,
a double-ended protection scheme uses the measured values at various locations to
analyze the fault conditions, and error in the measurement values may lead to the
incorrect operation of the protective devices. Therefore, validation of voltage and
current sensors using any standard validation algorithm is the most crucial part of
any protective algorithm. This helps to reconstruct the bad data, and it can be replaced
with calculated appropriate data. The errors or variations in the measurements are
generally due to disturbances in the DC micro-grid, and there are many schemes [3]
discussed in the literature to verify the parametric uncertainties in the measurement.
This method can be used for the calibration and testing of the sensors, and in actual
practice, this will increase the fault clearing time.

3.2 Directional Element

Directional comparison can be achieved by using a double-ended detection scheme

[4]. The direction of the fault current and communication network can be used to
differentiate the fault, and the coordination can be done in two ways:
1. The tripping scheme such as directional comparison unblocking (DCUB)
2. The blocking scheme such as directional comparison blocking (DCB).
A better protection coordination can be achieved using double-ended fault direc-
tion comparison methods. For example, in DCUB, during a feeder fault, if the adjacent
protection devices both detect a fault in forward direction, they communicate and
trip together. On the other hand, DCB schemes communicate faults to the upstream
device in the opposite direction of the fault to block them from operating.

3.3 Protective Relays

The relays attenuate many disadvantages of the fuse and circuit breakers. DC power
relays protects the systems from the over-current, over-voltage, under-voltage, time
derivatives, step changes in the current/voltages, and ground faults. Protective relays
A Review on Protection Devices and Protection Techniques … 341

are a smart device which give the command circuit breaker of the line segment
when any abnormal/unwanted condition occurs in the DC micro-grid systems. In
most of the relays, external voltage and current sensors monitor the real-time system
conditions, and if there is any fluctuation in the estimated value compared to the
critical value, a delay time is activated. If the quantify value is still maximum even
after the delay time run out, microprocessor in the relay will provide a trip signal to the
contact and will distinguish the fault. The relay can be implemented with or without
communication network based on the application and DC micro-grid configuration.
Digital relays with microprocessor are more versatile because they can record and
protect more than one fault condition in the DC micro-grid.

3.4 Fuses

Generally, fuses can be applied to DC systems having di/dt (low inductance) where
the time for the fuse to reach its melting point is minimized. Regarding the simplicity,
fuses are to a reliable device for the protection of the DC micro-grid. Fuses are
simplest and most easily available protection devices for the AC and DC systems
[4]. Fuses are the over-current device which protect the electrical and electronic
equipment from over-current by melting and disconnecting the circuit from main
supply and protect the equipment from serious damage and firing. Fuses serve two
main purposes:
1. To protect the components, equipment, and persons from the risk of fire and
electric shock.
2. Isolate the unhealthy systems form the main system.
In DC micro-grid systems, selecting fuses as a protective device is dependent
on the DC micro-grid elements, level of protection, and fault characteristics. Nowa-
days, high speed or ultra-fast fuses, generally known as the semiconductor fuses, are
achieving more attention for power electronics application [5].

3.5 Non-fuses DC Circuit Breaker (DCCB)

DC circuit breaker (DCCB): Another protective devices are DC circuit breaker which
are used in the DC micro-grid. DCCB can replace the fuses and improve the operating
condition the systems. The arc extinguishing time is longer for DCCB due to the
nature of the direct current systems, where voltage is continuous. In DCCB, the arc
voltage is a function of system voltage, self-induced voltage (beck EMF), and voltage
across the arc attenuator [6]. Most of the move forward DCCBs also offer ground
fault protection attributes. To provide accuracy and reliable operation, move forward
DCCBs use microprocessor-based digital sensing schemes. Hence, the selectivity of
the CB can be customized based on adjustable fault pickup settings and time delays.
342 Bholeshwar and R. S. Bhatia

In some DCCBs, move forward communication options help protection computing,

remote on/off trip mechanisms, etc., and this permits selectivity of CBs as a primary or
backup protection device. Molded case CBs (MCCBs) [7, 8], vacuum CBs, and hybrid
solid/vacuum interrupters are the common CBs used in DC micro-grid systems.

3.6 Solid-State DC Circuit Breaker

Solid-state circuit breakers are the most encouraging protective devices for DC micro-
grid causes of ultra-high speed operation whenever the fault occurs in the systems,
the fault clearing time is very less. Here, different levels of SSCBs are available for
DC micro-grid systems [9–12], which are given as, gate turn-off thyristor-based CB,
emitter turn-off thyristor-based CB, insulated gate bipolar transistor (IGBT)-based
CB, insulated gate commutated thyristor (IGCT)-based CB, and coupled inductor
SSCB. As we know that SSCB has the better performance for the DC micro-grid,
which has the following advantages such as high switching speed, fully control-
lable, high voltage blocking strength, high current blocking capability, low on-state
conduction loss, and no arcing. Solid-state circuit breaker technology has the lot
of advantages over the electro-mechanical and magnetic predecessors. Result of
mechanical wear and risks of arcing are completely eliminated, and estimation has
high reliability and longer lifetime. Faster switching is improving from milliseconds
to microseconds [13]. Solid-state circuit breakers can control its own affair sense
over-current and either hold it to limit or quickly open and drive it to zero.

3.7 Hybrid DC Circuit Breaker

Hybrid DC circuit breaker: In [14, 15], a thyristor-based hybrid DC circuit breaker is

explained. This hybrid DC circuit breaker has the combination of MCB and SSCB.
These HCBs have distinct limiting operation to identify the permanent and temporary
faults and interrupt the fault current within a short duration.

4 DC Micro-grid Protection Techniques

4.1 Impedance or Admittance-Based Protection

In this, techniques associated with voltage and current are estimated at the fault
location to measure the impedance. There are two distinct methods to compute the
impedance/admittance of the system. First simplest method is to directly calculate
the value of voltage and current, if the impedance of the line change during any
A Review on Protection Devices and Protection Techniques … 343

disturbance compared the threshold impedance with the change impedance and if
change impedance exceeds the threshold value that generates the trip signal. Another
method is active impedance estimation discussed in [16, 17]. In these techniques,
spread frequency current is injected with the help of power electronic converter which
is a short duration to estimate the impedance of the line. The computation is based on
the voltage and current response, and a fast Fourier of these techniques is that, it uses
the extra power transformer (FFT) to analyze the formations. The main drawback
of the electronics equipment is to identify the fault and to increase the fault clearing
time.

4.2 Superimpose Current-Based Unit Protection

A protection technique for the DC micro-grid is based in the superimposed of the

fault current of both end of line segment is estimated. Superimposed fault current
discriminates the external and internal fault of the line segment. Intelligent electronic
device (IED) is used in these techniques to generate the trip signal which is fed to
the SSCB, and SSCB has the fastest speed as compared to the other circuit breakers
as explained above.
An intelligent electronic device is installed in every segment as shown in Fig. 5,
to compute protection commitment using current of both terminals. Superimposed
current of the both end of line terminal is represented in the i-plane for the protection
determination [18].

4.3 Differential Current-Based Protection

In [19], a unique differential protection scheme is described for the DC micro-

grid protection. Instead of the conventional protection scheme, this method uses
the concept of differential zone protection, and the author developed an algorithm
for suitable location of sensor and relay.
Communication network and time synchronization of data are required for the
efficient operation of the system. These protection techniques determine the data
from the both ends of the line segment of the network and exchange the data through
the communication link as shown in Fig. 6. Data is compared with the threshold value
to detect the fault. Discussed method have disadvantage that it takes long pickup time
comparatively and decays the performance of the method while severity of the fault.
344 Bholeshwar and R. S. Bhatia

4.4 Over-current Protection Techniques (Directional)

In complex design DC micro-grid system, movement of the current could be at both

ends. Causes of this issue implemented directional over-current relay that plays an
important role in improving fault selectivity. Nowadays, this directional over-current
relay system contains the communication systems. Because of the purpose tech-
niques, once the disturbance occurs in the system’s amplitude and direction, the
distributed current will be changed, then direction of the current in all branches
will be checked with the help of communication system, and this will help in loca-
tion of fault in the line. Direct over-current protection is difficult in DC micro-grid,
because finding over-current is difficult in the faulty line because of power electronics
converter that reduces the magnitude of the faulty current.

4.5 Current Derivative Protection

Current derivative feature is considered to detect a fault in a very small time. Once the
disturbance occurs in the system, current derivative increases from zero to high value
that do not require any communication network to clear the fault. Current derivative
value depends upon the length of the cable, line loading, and faulty impedance.
Because of the given fact, it is hard to find out the proper critical value. This critical
value has to be redesigned for each operating condition. To find solution of this
problem, first order and second order of the derivative of the current find out the low
and high impedance fault. Furthermore, to measure the current derivative, high rate
sampling sensors are used to operate. Because of the high rate of sampling, noise is
amplified in the system and may result in false tripping. To overcome this problem, a
systematic filtering solution is needed to improve both the time delay and high noise
cancelation capability.

5 Conclusion

Because of the different characteristics of the DC micro-grids during faulty condition,

a well-defined protection system must be a key of interest. The behavior of the fault
current depends upon the power electronic converter connected in the network and the
interconnected cabling between them. In this review paper, various protection devices
and the protection techniques are explained. Various types of fault at possible fault
locations are identified, and suitable protection techniques are suggested. Conven-
tional DC circuit breaker contains many demerits as more fault clearing time and
arc, etc. Therefore, introduction of solid-state circuit breaker (SSCBS) and hybrid
circuit breaker diminishes the demerits and improves the whole performance of the
A Review on Protection Devices and Protection Techniques … 345

DC micro-grid system. This review paper explains distinct method of fault detec-
tion, isolation, and location. Each protection technique has own merits and demerits.
Most of the protection techniques “instead of superimpose current based protection
techniques” are applied for low impedance fault, radial DC system, and grounded
system.

References

1. Eaton-molded case circuit breakers, DC circuit breakers. [Online]. Available: https://www.

eaton.com/us/en-us/catalog/electrical-circuit-protection/dc-and-pvgard-molded-case-circuit-
breaker
2. Bui DM, Chen SL, Wu CH, Lien KY, Huang CH, Jen KK (2014) Review on protection coor-
dination strategies and development of an effective protection coordination system for DC
micro-grid. In: Asia-Pacific power and energy engineering conference, APPEEC, pp 1–10
3. Hansen S, Jordan B, Lang M, Walsh P (2013) Fuses versus circuit breakers for low voltage
applications. Mersen USA 1(1):12
4. Kim H (2013) Arcing characteristics on low-voltage DC circuit breakers. In: 15th European
conference on power electronics and applications, pp 1–7
5. Xu Z, Zhang B, Sirisukprasert S, Zhou X, Huang AQ (2002) The emitter turn-off thyristor-based
DC circuit breaker. IEEE Power Eng Soc Winter Meet Conf Proc 1:288–293
6. Maqsood A, Corzine K (2016) DC micro-grid protection: using the coupled-inductor solid-state
circuit breaker. IEEE Electrif Mag 4(2):58–64
7. Shen ZJ, Miao Z, Roshandeh AM (2015) Solid state circuit breakers for DC micrgrids: current
status and future trends. In: IEEE first international conference on DC microgrids, pp 228–233
8. Fregosi D et al (2015) A comparative study of DC and AC micro-grids in commercial buildings
across different climates and operating profiles. In: IEEE first international conference on DC
microgrids, pp 159–164
9. Atom Power, Inc. Solid state power atom switch. [Online]. Available: 54 https://www.atompo
wer.com/atom-switch
10. Cuzner RM, Venkataramanan G (2008) The status of DC micro-grid protection. In: IEEE
industry applications society annual meeting, pp 1–8
11. Mohanty R, Pradhan AK. A superimposed current based unit protection scheme for DC micro-
grid
12. Meghwani A, Srivastava SC, Chakrabarti S (2017) A non-unit protection scheme for DC
microgrid based on local measurements. IEEE Trans Power Deliv 32(1):172–181
13. Emhemed AAS, Burt GM (2014) An advanced protection scheme for enabling an LVDC last
mile distribution network. IEEE Trans Smart Grid 5(5):2602–2609
14. Zhang L, Woodley R, Song X, Sen S, Zhao X, Huang AQ (2018) High current medium voltage
solid state circuit breaker using paralleled 15 kV SiCETO. In: Conference proceedings—
IEEE applied power electronics conference and exposition—APEC, vol 2018, Mar 2018, pp
1706–1709
15. Saleh KA, Hooshyar A, El-Saadany EF (2015) Hybrid passive overcurrent relay for detection
of faults in low-voltage DC grids. IEEE Trans Smart Grid 1–10
16. Dhar S, Patnaik RK, Dash PK (2017) Fault detection and location of photovoltaic based DC
micro-grid using differential protection strategy. IEEE Trans Smart Grid 3053(c):1
17. Wang J, Berggren B, Linden K, Pan J, Nuqui R (2015) Multi-terminal DC system line protection
requirement and high speed protection solutions. In: CIGRE international symposium, pp 1–9
18. Jhunjhunwala A, Lolla A, Kaur P (2016) Solar-DC micro-grid for Indian homes: a transforming
power scenario. IEEE Electrif Mag 4(2):10–19
346 Bholeshwar and R. S. Bhatia

19. Prasai A, Du Y, Paquette A, Buck E, Harley R, Divan D (2010) Protection of meshed micro-
grids with communication overlay. In: IEEE energy conversion congress and exposition, pp
64–71
Positive and Negative Vibe Classifier
by Converting Two-Dimensional İmage
Space into One-Dimensional Audio Space
Using Statistical Techniques

Kushal Sharma and Sandeep Gupta

Abstract Everything in this world has a vibe/energy either positive or negative such
as Sun is considered to be positive and garbage is considered as negative. Now, these
vibes cannot be visualized but can be felt and no one can feel it by seeing it but if
some negative audio is playing such as some depressing song then it may be felt. This
paper describes how to classify negative and positive vibes in images which are based
on converting images to audios. Also, this paper describes what features to extract
in doing the audio classification of these types of audio signals and classifying them
using multilayer perceptron with special weight initialization and hyperparameters.

Keywords Deep learning · Audio classification · Statistical analysis · Vector space

1 Introduction

Humans describe most of the things using image signals and some things using audio
signals. But audio plays an important role when one describes something hidden
behind the walls. Also, every object has some energy in forms of vibrations that can
only be felt but cannot be visualized.
The flow of this paper starts with an insight to the data that is used followed by
the technique that converts images to audios, feature extraction, data preprocessing
for deep learning (Multilayer perceptron), and conclusion.

K. Sharma
KAI Institute of AI, Jaipur, Rajasthan, India
e-mail: [email protected]
S. Gupta (B)
Electrical Engineering Department, JECRC University, Jaipur, Rajasthan, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 347

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_33
348 K. Sharma and S. Gupta

Table 1 Example of text

Text ID/image ID Example Class
corpus that will be used for
extracting text 00 Anger Negative (0)
01 Love Positive (1)
02 Fight Negative (0)
03 Anxiety Negative (0)
04 Friendship Positive (1)

2 Data Acquisition and Audio Features

2.1 Data Acquisition

Data is divided into two parts; one set is of positive images and one set is of negative
images. Images are scraped/extracted from google images (https://www.google.com/
images). Also, used a text corpus (https://sentiment.christopherpotts.net/lexicons.
html#opinionlexicon) of positive and negative sentimental words, also listed some
positive and negative sentimental object names like sun, garbage, cactus, etc., and
labeled them accordingly. Using this corpus, all the images from google images are
extracted and saved to the machine.
Text ID/Image ID is illustrated in Table 1 that is an index of the array in which
corpus is kept, which will also refer to the image.

2.2 Converting Images to Audios

Some non-semantic methods of converting image signals to audio signals.

1. Piano Transform [1]: This is a classic method for the non-semantic image to
audio conversion by Dutch scholar Meijer. After this, they developed an assis-
tance system for visually impaired people called “Piano Transform” based on
the principle of conversion vOICe.
2. Image segmentation and edge detection-based conversion [2–4]: To process
visual image, they used image segmentation algorithm and edge detection
algorithms to process voice patterns.
3. Listen2dRoom [5]: It uses sound to help visually impaired to learn a strange
image layout and made the capability to do individual activities better by
assisting, done by Dr. Walker from Georgia Institute of Technology.

Image-audio conversion method adopted by the vOICe system: In this method,

the input image is of dimension (n × n) and each pixel has been corresponded to a
sine wave. Frequency of wave is kept as the pixel index, so farther the pixel location
higher the frequency. The amplitude of each sine wave is kept as pixel intensity/pixel
values. Image pixel intensity values range from 0 to 255 in which 0 is black and 255
Positive and Negative Vibe Classifier by Converting Two … 349

is white if considered a grayscale image. For colored images (RGB), there are three
channels each for red, green, and blue.
After getting all the sine waves for each pixel all the waves using inverse discrete
Fourier transform (IDFT) [6], it will be converted to one audio signal. To make it
computationally fast, we will use inverse fast Fourier transform (IFFT).

N −1
  
x(n) = 1/N + X (m)ej2πmn/N (1)
m=0

Using Euler formula

e−jθ = cos ∅ − j sin ∅ (2)

This can be rewritten as

N −1
     
1 2π mn 2π mn
x(n) = + X (m) cos + j sin (3)
N m=0 N N

2.3 Data Preprocessing for Multilayer Perceptrons

In the audio domain, there are many features that can be extracted out from audio
signals that help in doing further classification as they are highly correlated most of the
time with the target variable. There are a total of thirteen features that are extracted
from the audio in order to classify them using multilayer perceptrons and these
include mean frequency, standard deviation, median frequency, third-quartile range,
first-quartile range, spectral entropy, spectral flatness, kurtosis, skewness, spectral
flatness, mode frequency, frequency centroid, and dominant frequency.
1. Median frequency: This is the central value of the data extracted by first
converting it into the frequency domain from the amplitude domain.
2. Mean frequency: It is a statistical natural parameter and is the best estimate
of the distributions most of the times. First amplitude values are converted to
frequency values using discrete Fourier transform and then, mean is taken of
frequency values.
n
fi
fˆ = i=1
(4)
N
3. Standard deviation frequency: It gives the idea of expansion of the distribution
of the audio frequency. Average difference from the mean squared and rooted.
350 K. Sharma and S. Gupta



 n fi − fˆ
 i=i
σ = (5)
n

4. First quartile (Q25): Median of first 25% of the data.

5. Third quartile (Q75): Median of values above 75% of data.
6. Inter Quartile Range (IQR): Difference between Q25 and Q75 values.
7. Kurtosis: Outliers in the data mislead the statistical algorithms. Kurtosis gives
the idea of outliers in the data.
n  
1  xi − x̂ 4
K = (6)
n i=1 s(x)

8. Skewness: If skewness is positive or negative then there are outliers because it

creates a difference between the mean and median and that is what skewness
gives.

 n
 (x − xi )3
g= (7)
i=1
(n − 1)s 3

9. Spectral entropy: First power spectrum density of speech signal is calculated

and then, spectral entropy is calculated by using normalized Shannon entropy
property.

1 f2
SE[ f 1 , f 2 ] = − Pn ( f i ) log(Pn ( f i )) (8)
log[N [ f 1 , f 2 ]] f = f
i 1

10. Mode frequency: Frequency value with the highest number of repetitions.
11. Spectral flatness: Gives the noise parameter in the sound and gives the idea of
noisiness in sound. Calculated as the ratio of geometric mean and arithmetic
mean.
 
exp N1 kK log(xn )
Sflatness = 1 K
(9)
N k xn

12. Centroid of frequency: Centroid value of sound in the frequency domain.

13. Dominant frequency: Frequency possessing maximum energy. Calculated by
applying energy integral and getting the maximum one.
Positive and Negative Vibe Classifier by Converting Two … 351

2.4 Final Data for Training

After extracting all the features, the data will be of the shape (n × 13) n is the number
of audio signals and as there are 13 features that we will extract out of each audio
signal. This will be the input data in which each audio has a class either 1 or 0 encoded
as positive class and negative class, respectively.

3 Multilayer Perceptron Algorithm

Multilayer perceptrons also called neural networks are very sound algorithms when
there is a huge amount of data available and classification boundary is nonlinear to
find. It projects data to higher dimensions by extracting multiple features out of it
in another vector spaces and multiple dimensions and ultimately finds a nonlinear
hyperplane that classifies the data (Fig. 1).

3.1 Layers

Layers in neural networks are a combination of dot product of previous output and
random weights followed by addition with random bias value. In the first layer, dot
product of input data and random weight matrix (w1) is done and bias (b1) is added
to the dot product and gives output (o1). In the next layer, dot product of previous
output (o1) and random weight matrix (m2) is done and bias (b2) is added to that.
This process is followed until the desired number of layers is achieved. Higher the
number of layers higher is the dimension of projected data and hypothesis equation.

Fig. 1 Illustrative multilayer

perceptron
352 K. Sharma and S. Gupta

3.2 Activation Functions

As in layers doing dot product and adding bias, makes a linear hyperplane but it
does not make the hyperplane nonlinear which is crucial in classifying nonlinearly
separable dataset. So, activation functions make the multidimensional hyperplanes
nonlinear in order to get a good fit to the dataset. Some of the commonly used
activation functions are:
1. Sigmoid f (x) = 1
1+e−z
−2x
2. Hyperbolic Tan f (x) = 1−e
1+e2x
3. ReLU f (x) = max(0, z)
4. Leaky-ReLU f (x) = max(∈ x, x)
zj
5. Softmax f (z) = Ke ezk
k=1

Every function has its own importance and it is related to its two qualities.
1. It should not vanish (make zero) the optimization gradients.
2. It should give gradients both positive and negative so that multiple weights can
increase or decrease depending on the cost function value; so it should be flexible
to do that.

3.3 Error/Cost Function

Mean-squared-error cost function is used which is most common and easy to mini-
mize as it is a convex function so gradient-based methods work well on this. Here is
the equation of the mean-squared-error cost function:
n  2
i=1 yp − ya
y= (10)
n

3.4 Gradients

The gradients in neural networks are calculated as it is used in optimizing the error
function. It is calculated using the chain rule of differential calculus. Consider four
layers I, J, K, L. These derivatives are always calculated of the cost function with
respect to all the weights. The chain rule is followed and these are assembled in a
hierarchy.

∂ OL
= 2 × mean(w L × L − y A ) × L (11)
∂w L
Positive and Negative Vibe Classifier by Converting Two … 353

∂ OL δ OL δ IL
= × (12)
∂wK L δ IL δw K L
δy δ OL δ IL δ IK
= × × (13)
δw JK δ IL δ IK δW JK
δ OL δ OL δ IL δ IK δ IJ
= × × × (14)
δw I J δ IL δ IK δ IJ δwi j

4 Optimization: Adam Optimizer

Adam stands for adaptive moment estimation. It is a gradient-based method evolution

from momentum gradient descent and RMSProp algorithm.
α 

wt+1 = wt −  
vt (15)
st + ∈
vt
v̂ = (16)
1 − β1t
 st
st = (17)
1 − β2t
∂L
vt = β2 vt−1 + (1 − β1 ) (18)
∂wt

∂L 2
St = β2 St−1 + (1 − β2 ) (19)
∂wt

Equation (16) is an adaptive equation as the iterations (t) will increase, it will


selfreduce the gradients using a slight factor. st is also an adaptive parameter that
will also reduce with a slight factor at each iteration (i) and will increase the learning
rate expression slightly in order to avoid overshooting and finding the optimal balance
between learning rate and gradients for optimal stability.

4.1 Weight Initialization

Weight initialization is done using Xavier initialization [7]. In this Xavier initializa-
tionm, weights are sampled from continuous Gaussian distribution (CGD) with zero
mean and standard deviation:
2
σ = (20)
n in + n Out
354 K. Sharma and S. Gupta

Fig. 2 Error optimizing

curve on 70% training data

where n in is the number of inputs in layer and n out is the number of outputs in layer.

5 Training and Results

Input data is of the shape of (n × 13), n being the number of audio signals and 13
being the feature count. The architecture of a neural network is having 13 neurons
in each layer and there are eight such layers with dropout turned on which is a
regularization technique and reduces the risk of overfitting. Figure 2 shows the error
optimizing curve. Here, data are split into three parts and are distributed as:
1. 70% for training
2. 20% for testing
3. 5 sets of 2% each for cross-validation.
After training it for 2000 epochs with hyperparameters:
1. beta = 0.99
2. alpha = 0.001.
Accuracy achieved:
1. Training: 89.9583%
2. Testing: 89.09167%
3. Cross-validation (Averaged): 90.33217%.

6 Conclusion

After training it with Adam Optimizer for 2000 epochs with 0.01 and testing on
Testing as well as cross-validation data, this model is able to classify the negative
Positive and Negative Vibe Classifier by Converting Two … 355

and positive vibes in images. In the future, this research can be used to classify/predict
the intention of suspicious people found at crowded places by predicting the vibe
coming from their face which is not easy but not impossible also. It can save thousands
of lives and can create a world where humans can live peacefully as it will predict
the threat that could be caused to society by uncivilized people.

References

1. Meijer L (1992) An experimental system for auditory image representations. IEEE Trans Biomed
Eng 39(2):112–121
2. Tian YN, Lei HW, Wang X (2010) An automatic navigation system based on SVM image
segmentation for the blind. J Northeastern Univ Nat Sci 31(12):1681–1684
3. Tian YN (2011) A vision substitution method for the blind based on image edge detection and
sound mapping. In: Proceedings of third international conference on computational intelligence,
communication systems and networks, Shenyang, pp 209–212
4. Gupta S, Kumar K et al (2019) An IoT based approach to minimize air pollution. Int J Recent
Technol Eng (IJRTE) 7(6S5):596–599
5. Davison K, Suh H, Walker N (2012) Math GNIE: visually impaired students creating graphs
and number lines on a computer. In: Proceedings of the AER2012, Bellevue, pp 18–22
6. Lyons RG (2010) Understanding digital signal processing. Pearson Education, pp 167–308
7. Glorot X, Bengio Y (2010) Understanding the difficulty of training deep feedforward neural
networks. In: International conference on artificial intelligence and statistics, Italy, vol 9, pp
249–256
Coupled Field Method for Analyzing
Short-Circuit Electromagnetic Forces
in Power Transformers

Karan Bali and Deepika Bhalla

Abstract The electromagnetic forces can induce critical stress on transformer

winding during short circuit. The forces in axial direction can damage the insu-
lation and result in transformer insulation getting damaged. In this paper, axial elec-
tromagnetic forces in low voltage (LV) winding and high voltage (HV) winding of
transformer are analyzed during three-phase short-circuit condition. The flux density
and axial electromagnetic forces are analyzed using coupled field method. All simu-
lations of the coupled field theory are carried out for a three-phase 10 MVA 33\11 kV
power transformer. The cumulative axial forces during short-circuit condition in the
LV and the HV winding are computed. The net axial forces in both the windings are
assessed.

Keywords Axial force · Coupled field method · Electromagnetic forces ·

Short-circuit current

1 Introduction

When the short-circuit situation occurs in an operating electric power system, then
short-circuit current flows through the transformer and the electromagnetic forces in
the windings have large magnitude [1]. A short circuit can result in an outage of a
transformer due to these high electromagnetic forces in both the high voltage (HV)
and low voltage (LV) windings. The analysis of the short-circuit forces generated, and
these are resolved in radial F(x) and axial F(y) direction. The analysis of short-circuit
forces is done considering the leakage flux density at any portion of the transformer
winding. For the analysis of electromagnetic forces in a transformer, it is resolved in
two components, namely the axial component of the force in the y-direction which

K. Bali (B) · D. Bhalla

Department of Electrical Engineering, IKGPTU Jalandhar, Kapurthala, Punjab 144603, India
e-mail: [email protected]
D. Bhalla
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 357

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_34
358 K. Bali and D. Bhalla

is due to the radial leakage flux density and the radial component of the force in the
x-direction which is due to axial leakage flux density. The leakage field pattern at
the ends of windings due to the fringing of leakage field results in the reduction of
radial forces in a winding as it reduces the axial components of the field.
The coupled field method is also known as the finite element method (FEM), and
it is the most preferred mathematical technique because of its three prime features
and capability; firstly, its formulation can handle anisotropic, geometric complexity,
non-uniform and nonlinear media, secondly due to the availability of some commer-
cial software which have the applicability to simulate transformer problems easily,
thirdly because the method is able to solve problems which are a combination of
other physical fields and electromagnetic fields within the circuits [2]. Short-circuit
forces in a transformer have been analyzed for different type of transformer. Analysis
has been carried out on oil power transformers [3–7], dry-type power transformers
[8], and power distribution transformers [9] and on core-type transformers [10].
The ratings of transformers used in different research paper are 10 MVA, 6 MVA,
130 MVA, 200 MVA, 15 KVA [6, 8, 9, 11]. As per the literature review, the analysis
of short-circuit forces in different types of transformer using different software and
technique to study of the flux distribution and subsequent forces has been done by
2D FEM [3–7, 10, 11] and 2D and 3D ANSYS [8]. There is scope of optimization
techniques as well [12–14].
This paper described three-dimensional magnetostatic coupled field analysis of a
10 MVA power transformer, and the axial forces are computed for both of windings
of the transformer. The commutative axial short-circuit forces are calculated for both
the windings, and the direction of the forces is analyzed and compared with that
available in the literature [3, 10].

2 Electromagnetic Forces

To find the electromagnetic forces in the transformer winding, the knowledge of

magnetic flux density distribution inside the transformer winding and air-gap is essen-
tial. Analysis of short-circuit forces in each of the LV and HV winding is done using
the value of leakage flux density in transformer winding [8]. The LV winding is the
inner winding, and the HV winding is the outer winding of the transformer. The inner
winding lies next to the core which under symmetrical conditions experiences radial
forces that are directed towards the core and the outer winding experiences radial
forces that are directed towards the tank. Figure 1 shows the distribution of the field
and electromagnetic forces in transformer windings.
Coupled Field Method for Analyzing Short-Circuit Electromagnetic … 359

Fig. 1 Distribution of field

and electromagnetic forces
in transformer windings [16]

2.1 Radial Forces

The axial leakage field produces the radial forces that act in the outward direction of
the outer winding, tending to produce tensile stresses which stretch the conductor in
it. Tensile stress is also known as hoop stress.
The radial leakage field produces the radial forces that act in the inward direction
of the inner winding tending to crush the winding towards the core. Thus, compressive
stress is produced. For deriving the radial force in a winding, the fringing of the field
is neglected and the approximation is made justified because the maximum value of
the radial force is important, which occurs in the middle portion of the winding [2].
The value of the leakage field increases from a minimum (zero) at the outside
diameter to a maximum value at the inside diameter (at the gap between the two
windings) in case of outer winding [2]. In the gap, the peak value of flux density is
√
Bgp = 2μ◦ NI/Hw (1)

where NI is (RMS value) of windings in ampere-turns and Hw is the height of winding

in meters. The average flux density value in the whole winding is half of the gap
value. The total radial force acting on the winding with mean diameter of Dm (in m)
can be calculated by the equation of electromagnetic forces

μ◦ NI
F = IL × B as Fr = × π Dm (2)
Hw
360 K. Bali and D. Bhalla

For the outer winding, the conductors close to the gap experience a higher force as
compared to near the outside diameter (the force reduces linearly from its maximum
value at the gap to zero at the outside diameter) [2]. The force can be considered to be
transfer from conductors with high load (force) to those with low load if conductors
are wound tightly [4].

2.2 Axial Forces

Due to the radial leakage field at the end of windings, the axial forces are directed
towards the winding center. At the ends of windings, axial forces are high and
compressive force is high at their center as shown in Fig. 2. Under ideal condi-
tions, compressive forces with on end thrust on the clamping structure both inner
and outer windings. For asymmetry factor of 1.8, the total compressive force is acting
on outer and inner winding which is expressed by

91.8 × 50.8 × S
Fa = , (3)
Z pu × Hw × f

where S is the rated power per limb in kVA, Z pu is the per-unit impedance, Hw is the
winding height in meters, and f is frequency in Hz [2, 7].
Due to the higher radial flux, the inner winding being closer to the limb and
experiences higher compressive force as compared to the outer winding. When the
detailed analysis is not available, it is assumed that takes about 25–33% of the force
and the remaining 75–67% force is taken by inner winding [2, 6, 8, 9], and this is just
an assumption. The reasons for a higher value of radial field, and consequent axial
forces are mismatch of ampere-turn distribution between the LV and HV windings
[6]. As asymmetries are present, these cause end thrusts in which the inner winding
tends to move upward and the outer winding tends to move downward.

Fig. 2 Distribution of flux

density and axial forces in
inner and outer winding
Coupled Field Method for Analyzing Short-Circuit Electromagnetic … 361

2.3 Short-Circuit Current

The worst conditions that can take place in practice the first peak of a three-phase
short-circuit flowing through the winding to corresponding the transformer is linked
to a system of infinite faulty current capacity. Hence, it is standard practice to design
a transformer to hold up a three-phase short-circuit at its terminals, and the other
windings being supposed to be connected to infinite systems. The balanced short-
circuit current for a three-phase two-winding transformer is specified by

V
If = √ kA (4)
3(Z T + Z S )

where V is the rated line-to-line voltage in kV, Z T is the short-circuit (leakage)

impedance of the transformer, Z S is the short-circuit impedance of the system. The
system impedance is little as compared to the transformer impedance and is ignored
[15].
The nature of a short-circuit current can be highly asymmetrical, and it is
maximum when the short circuit occurs at zero voltage instant. An asymmetrical
short-circuit current has two components: a unidirectional component decreasing
exponentially with time and an alternating steady-state symmetrical component at
the fundamental frequency. The rate of decay of the exponential component is decided
by the X/R ratio of the transformer. IEC 60076-5 (third edition: 2006-02) for power
transformers specifies the asymmetry factor corresponding to switching at the zero-
voltage instant (the worst condition of switching). For X/R > 14, its value is specified
as 1.8 for transformers up to 100 MVA rating, whereas it is 1.9 for transformers
above 100 MVA rating, the peak value of the asymmetrical short-circuit current for
ratings up to 100 MVA can be taken as [11, 15].

Iasm = πr 2 × 1.8 × If (5)

3 Modeling of Transformer

To find the electromagnetic forces in the transformer winding, the knowledge of

distribution of magnetic flux density inside the transformer winding is essential. For
the analysis of the axial electromagnetic forces, the windings on the central limb
have been energized. The boundary conditions have been applied at the outer most
surface of transformer model, thus creating a region to avoid the leakage of current
and enclose the transformer model. According to construction, the transformer core
laminations are used to give it a cruciform shape. In structural analysis, only the inner
and outer disk windings of central limb have been considered to analysis the axial
362 K. Bali and D. Bhalla

Table 1 Specification of
Parameters Outer (HV) Inner (LV)
10 MVA three-phase power
winding winding
transformer
Phase Three Three
Type of coil Continuous disk Continuous disk
type type
Number of turns per 1352 124
phase
Voltage (V) 33,000 11,000
Per phase short-circuit 7068.867 77,073.45
current (A)
Yoke clearance 58.5 58.5
Interphase connection Star Delta
Frequency (Hz) 50 50

Table 2 Dimensions and

Parameters Outer (HV) winding Inner (LV) winding
design parameters of 10 MVA
33\11 KVA power Window height 880 792.5
transformer (mm)
Outside diameter of 776 577.5
coil
Number of sections 15 15
Core material M4 M4

forces which can be formed during the short-circuit current is energized in trans-
former winding [8]. The heights of both the windings are divided into sections/cells
of equal height with a gap of 2 mm between them. Table 1 gives the specifications
of the transformer used in this work, and the dimensions and design parameters are
in Table 2.
For analysis of electromagnetic axial forces during short-circuit, 10 MVA
33\11 kV core-type power transformer has been considered and Table 1 gives the
other transformer specifications. There are 1352 HV winding turns and 124 LV
winding turns; for this disk-type winding arrangement, the entire height of both the
windings is divided into 15 sections/cells. The per-unit impedance of the transformer
is 0.0824. The value of short-circuit current in outer and inner winding is 7068.867 A
and 77,073.45 A, respectively. The ampere-turns during the short-circuit conditions
are 9,557,108.825 AT. For plotting the magnitude of electromagnetic axial forces, the
values of forces in each of the 15 sections/cells are taken. For finding the cumulative
axial force in a winding, the force in the end section/cells added to that of the next
section/cell and soon. Values of all the sections cells are added to obtain total force
[8, 9].
Coupled Field Method for Analyzing Short-Circuit Electromagnetic … 363

Fig. 3 Short-circuit axial forces in inner winding

4 Result and Discussion

4.1 Short-Circuit Forces in Inner Winding

For the inner winding, the axial force exerted on the lower and upper half of the
winding is in opposite direction. Figure 3 shows the axial force in each of the 15-
sections/cells of the inner winding. For this winding, the axial force on the lower
half of the winding is negative and on the upper half of the winding is positive, and
thus, the winding is being compressed toward the center. Due to the fringing of the
radial flux, the magnitude of axial force is more towards the two ends; its value being
1.8154E+05N at the lower end and −1.9305E+05N at the upper end. Figure 4 shows
the cumulative axial force of the inner winding, and the curve shape is in agreement
with that given in the literature [15].
The net axial force in inner winding is found to be 4.71E+04N and is in the upward
direction; thus, during short circuit, the inner winding tends to move towards the yoke
in the upward direction and is of tensile nature and expands it.

4.2 Short-Circuit Forces in Outer Winding

For the outer winding, the axial force exerted on the lower and upper half of the
winding is in opposite direction and same as that in the outer winding. Figure 5 shows
the axial force in each of the 15 sections/cells of the winding. For this winding, the
axial-force on the upper half of the winding is negative and on lower half is positive;
thus, the winding is also being compressed toward the center. Due to the fringing of
364 K. Bali and D. Bhalla

Fig. 4 Cumulative short-circuit axial forces in inner winding

Fig. 5 Short-circuit axial forces in outer winding

the radial flux, the magnitude of axial force is more towards the two ends; its value
being 12,921 N at the lower end and −13,445 N at the upper end. Figure 6 shows the
cumulative axial force of the inner winding, and the curve shape is in agreement with
that given in the literature [8, 15]. The net axial force in outer winding is found to
be −3420.18 N and is in the downward direction; thus during short circuit, the inner
winding tends to move towards the yoke in the downward direction and compresses
it.
Coupled Field Method for Analyzing Short-Circuit Electromagnetic … 365

Fig. 6 Cumulative short-circuit axial forces in outer winding

5 Conclusion

In this paper, the coupled field theory has been used to calculate the electromagnetic
forces which act in the axial direction in power transformer windings during short
circuit. The magnitude of the axial forces in the inner winding/LV winding and outer
HV winding has been calculated. The cumulative axial forces are also calculated.
The magnitude of the axial forces at the end turns is maximum for both the windings.
Most of the total axial force is borne by the inner winding. Also, during short circuit
an upward thrust is exerted on the inner winding and downward thrust is exerted in
the outer winding of the transformer.

References

1. Li J (2019) Design and application of modern synchronous generator excitation systems. Wiley-
IEEE Press. ISBN 978-1-118-84087-0
2. Kulkarni SV, Khaparde SA (2012) Transformer engineering: design, technology, and diagnos-
tics. CRC Press, Taylor & Francis Group. ISBN 9781439853771
3. Ahn H-M, Lee J-Y, Kim J-K, Oh Y-H, Jung S-Y, Hahn S-C (2011) Finite-element analysis of
short-circuit electromagnetic force in power transformer. IEEE Trans Ind Appl 47(3):1267–
1272
4. Guttan M, Janura R, Jurcik J (2015) Determination of the electromagnetic forces caused by
short circuit currents
5. Waters M (1954) The measurement and calculation of axial electromagnetic forces in concentric
transformer windings. Proc IEE 101(Pt. II):35–46
366 K. Bali and D. Bhalla

6. Wang X, Yu S, Zhao Q, Wang S, Tang R, Yuan X (2003) Effect of helical angle of winding in
large power transformer. In: Sixth international conference on electrical machines and systems,
ICEMS 2003, vol 1, pp 355–357
7. Gee FW, Whitaker JD (1963) Factors affecting the choice of priestess applied to transformer
windings. In: IEEE summer general meeting, Paper No. 63-1012_1963, Toronto
8. Janssen ALJ, tee Passkey LH (2000) Short circuit testing experience with large power
transformers, CIGRE 2000, Paper No. 12-105_2000
9. Kulkarni SV, Khaparde SA (2012) Transformer engineering: design technology, and diagnos-
tics. CRC/Taylor & Francis, Boca Raton
10. Bakshi A, Kulkarni SV (2014) Coupled electromagnetic-structural analysis of the spiralling
phenomenon in a helical winding of a power transformer. IEEE Trans Power Deliv 29(1):235–
240
11. Bhalla D, Bansal RK, Gupta HO (2018) Analysing short circuit forces in transformer with
double layer helical LV winding using FEM. https://doi.org/10.1109/RAECS.20157453274
12. Prakash T, Singh VP, Singh SP, Mohanty SR (2018) Economic load dispatch problem: quasi-
oppositional self-learning TLBO algorithm. Energy Syst 9(2):415–438
13. Rathore NS, Singh VP, Kumar B (2018) Controller design for DOHA water treatment plant
using grey wolf optimization. J Intell Fuzzy Syst 35(5):5329–5336
14. Prakash T, Singh VP, Mohanty SR, Singh SP (2017) Binary Jaya algorithm based optimal
placement of phasor measurement units for power system observability. Int J Control Theory
Appl 10(5):515–523
15. Faiz J, Ebrahimi BM, Noori T (2008) Three- and two-dimensional finite-element computation
of inrush current and short-circuit electromagnetic forces on windings of a three-phase core-
type power transformer. IEEE Trans Magn 44(5):590–597
16. Ahn H-M, Lee J, Hahn S-C (2011) Finite-element analysis of short-circuit electromagnetic
force in power transformer. In: IEEE Transactions on Industry Applications, pp 1267–1272
Thermal Separation in 2D Vortex Tube
for a Different Tube Length and Cold
Mass Flow Ratio

Ashish Kumar Gupt, Deepak Kumar, and M. K. Paswan

Abstract Vortex tube (VT) is used to disperse the pressurized gas (air) into hot and
cold streams. Further, the gas is inserted tangentially into the vortex tube with extreme
pressure and with the high degree of the swirl. The outer most part of the fluid flow
within the vortex tube is allowed to escape from the hot outlet, and the inner fluid
part is to escape from the cold outlet. In this paper, the computational fluid dynamics
(CFD) model is designed as a 2D axisymmetric model with an axial annular inlet
and a radial annular outlet which provides required rotational fluid flow inside the
vortex tube. Model was analysed for thermal separation for a counter-flow vortex
tube in commercially available software ANSYS Fluent 19.0 by taking working fluid
as air. The standard k-ε model is used for turbulent flow within the vortex tube. The
CFD model is used to check the VT boundary conditions on the thermal separation
and to provide detailed information on the flow velocities, temperature and pressure
distributions, as well as the cooling/heating effect. It is clearly observed that the
obtained cooling effect is maximum for a vortex tube of length 100 mm. Also,
cooling effect and heating effect with variation in cold mass flow ratio are validated
with already reported research data (Guo and Zhang in Int J Refrig 85:42–57, 2018
(Aljuwayhel et al. in Int J Refrig 28:442–450, 2005)).

Keywords Thermal separation · Cooling effect · Heating effect · Vortex tube ·

CFD

1 Introduction

Vortex tube (VT) is used to generate two streams at different temperatures with a
single inlet. It has no moving part which separates the pressurized inlet fluid in two
streams, cold stream and hot stream. Thermal separation takes place within the vortex
tube which separates the fluid into two cores, outer core and the inner core. The outer

A. K. Gupt · D. Kumar (B) · M. K. Paswan

Department of Mechanical Engineering, National Institute of Technology Jamshedpur,
Jamshedpur 831014, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 367

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_35
368 A. K. Gupt et al.

Hot Air out

Cold Air out

Hot Air out

Air in

Fig. 1 Schematic diagram of counter-flow vortex tube

core has a higher temperature and the inner core has lower temperature, both relative
to inlet fluid temperature. A powerful turbulent fluid is created within the main tube
due to the tangential flow of pressurized fluid inside the tube. It is used effectively
in refrigeration without using moving part. The other applications include cooling
during machining operations, waste energy recovery, air suits and masks, cryogenic
cooling in high speed machining, in-micro-power generation, etc. Manimaran [1]
studied flow characteristics that affects the thermal separations between the internal
core and external core inside the vortex tube and analysed with varying number of
the nozzle from 1 to 6 with trapezoidal shaped inlet while the other geometrical
parameters and mass flow rate kept at a constant. In their analysis, air is taken as
working fluid with the assumption of the ideal gas (Fig. 1). Sadeghiazad [2] had
investigated different convergent nozzles for improvement of performance of vortex
tube by experimental and numerical method. He observed that the maximum cooling
effect is obtained for a convergence ratio K = 2.22 (Fig. 1).
In the vortex tube (VT), the thermal separation characteristics had been investi-
gated and the best parameters are selected for the design in which the temperature
difference of fluid is maximum. The simulation results show that as the diameter of
cold outlet increased, the mass flow rate is also increased from the cold outlet. Further,
the temperatures at both the outlets decrease, as the number of inlets increases [3].
Rafiee and Sadeghiazad [4] had compared the experimental and simulated results
for the different control valve diameter, nozzle number and injection pressure for a
convergent vortex tube. Further, they observed that the optimum diameter (5.5 mm) of
a control valve is achieved for higher pressure drop, higher cooling effect and higher
separation efficiency. Further, they also found that as the number of nozzles increases
to 4, temperature effectiveness (cooling) improved. In their research paper, they also
analysed the fluid separation process inside the vortex tube using different oper-
ating gas. In their simulation, they used the standard k–ε turbulence model to solve
nonlinear equations. They concluded that the maximum cooling or heating effect is
obtained for the gas NO2 [5]. They also simulated the vortex tube by varying the
different hot tube length and found the optimum vortex tube length for the cooling
effect and heating effect. For the tube length 115 and 110 mm, the maximum thermal
heating and cooling are obtained, respectively [6]. Manimaran [7] studied the energy
separation by CFD analysis in a counter-flow vortex tube with different inlet-shaped
nozzle such as rectangular- and trapezoidal-shaped nozzle and with different aspect
ratio. They found that the trapezoidal-shaped inlet gives the higher cooling effect
as compared to the rectangular-shaped inlet and higher aspect ratio at inlet gives a
Thermal Separation in 2D Vortex Tube for a Different Tube Length … 369

higher cooling effect. Rafiee and Sadeghiazad [8, 9] had designed the various shapes
of hot control valve such as spherical, cone, plate and truncated cone inside a vortex
tube. Further, they also analysed the different model for the heat transfer and energy
separation by experimentally and numerically. Their experimental result shows that
the truncated valve in the vortex tube produces the maximum temperature differ-
ence and minimum with plate valve (cooling capacity also minimum). They also
examined various types of turbulence model inside the vortex tube for predicting the
separation phenomena by the 3D computational analysis. They found that among all
the tested method RSM method gives accurate thermal separation results. Dutta et al.
[10] numerically tested different turbulence models like standard k–ε, standard k–ω,
RNG k–ε and SST k–ω for predicting the temperature separation in an axisymmetric
CFD model of the vortex tube. The performance analysis shows that the temperature
difference is nearer to the experimental results predicted by the standard K-ε turbu-
lent model. Guo et al. [11] investigated the energy separation and flow characteristics
inside the vortex tube with CFD on the basis of the different cold mass fraction. The
velocity distributions and pressure fields for nine different cold mass fractions are
considered for simulation. For a better understanding of vortex structure, the different
type of vortex breakdowns was also discussed. Their research shows that when the
cold mass fraction is increasing, then the adverse pressure gradient enhanced along
the axis of the vortex tube which affects the vortex break down in the fluid domain
and its increase to the growth of the inner core of fluid domain. Aljuwayhel et al. [12]
investigated a counter-flow vortex tube for the flow behaviour and the energy sepa-
ration mechanism with the help of a 2D axisymmetry computational fluid dynamics
(CFD) model. There are three regions in the model of vortex tube flow field which
are hot flow region, cold flow region and recirculating region. The results of this
simulation are compared to previous experimental results from experiments of the
vortex tube by compressed air. A parametric study is also done in this paper for
effective length and diameter of the vortex tube. Upendra et al. [13] had optimized
the parameters of the vortex tube by CFD analysis through experimental investiga-
tion. Further, they maximized the COP of the vortex tube by optimization of tube
diameter (D), cold outlet diameter (d c ) and d c /D. Their analysis shows that length-
to-diameter (L/D) ratio has very small effect on COP of the vortex tube which is
negligible. Bazgir and Nabhani [14] investigated the effects of convergent and diver-
gent angles on the flow behaviour by the help of computational fluid dynamic (CFD)
using the RNG k–ε turbulence model. They computed the isentropic efficiency (ηis )
and coefficient of performance (COP) of the vortex tube on different divergent angles
and different convergent angles. Further, their finding on efficiency included some
geometrical and operating parameter including length of the tube, cold exit diameter
and different inlet pressure and mass flow rate. Their computed results show that the
COP and the isentropic efficiency of the vortex tube increase when the divergence of
the tube is increased for the most value of the inlet pressures. Also, divergent-shaped
vortex tubes are more efficient than the straight and convergent shapes of the vortex
tubes.
In this paper, a 2D axisymmetric model is analysed for thermal separation
behaviour for a counter-flow vortex tube in commercially available CFD package
370 A. K. Gupt et al.

ANSYS 19.0. The standard k-ε model is used for turbulent flow within the vortex
tube. The thermal separation in the vortex tube is analysed with the flow veloci-
ties, temperature and pressure distributions, as well as the cooling/heating effect.
Also, for the different cold mass flow ratio, effect of cooling and heating is plotted.
The optimum length of the vortex tube has been computed for optimum cooling and
heating effects. However, the effect of number and shape of the inlet nozzle for vortex
tube are not realized in 2D simulation.

2 Computational Fluid Dynamics (CFD) Governing

Equations

The following governing equations have been used in 2D vortex tube (VT) analysis
as:

2.1 Mass Conservation Equation

The continuity equation is written as follows [1]:

∂ρ
+ ∇ · ρ v = 0 (1)
∂t

2.2 Momentum Conservation Equation

The momentum equation is expressed as follows [1]:

∂  
(ρ v) + ∇.(ρ vv) = −∇ p + ∇. τ + ρ g (2)
∂t

where p is the static pressure of the fluid, τ is the stress tensor (Eq. 3), and ρ g is
the gravitational body force of fluid.
The stress tensor τ is expressed as
  
−
→ 2
τ = μ ∇ v + ∇ v − ∇ · v I
T
(3)
3

where m is the molecular viscosity and I is the unit tensor.

Thermal Separation in 2D Vortex Tube for a Different Tube Length … 371

2.3 Energy Conservation Equation

Energy conservation equation is expressed as [1]:

∂   
v (ρ E + p)) = ∇ · p + ∇. keff ∇T + τ eff · v
(ρ E) + ∇( (4)
∂t
where k eff is the effective conductivity (k + k t ), and k t is the turbulent thermal
conductivity. E is the energy and expressed by (Eq. 5):

p v2
E =h− + (5)
ρ 2
T
Here, h = Tref Cp dT and T ref is 298.15 K.

2.4 Ideal Gas Law for Compressible Flows

For compressible flows, the ideal gas law is expressed as, [1]

Pop + p
ρ= R
(6)
Mw
T

2.5 Viscosity

The equation of viscosity by Sutherland’s law can be written as follows [1]:

 3/
T 2 T0 + S
μ = μref (7)
Tref T +S

where S is the Sutherland constant and μref = 1.716 × 10−5 kg

m
− s, Tref =
273.11 K, and S = 110.56 K.

2.6 Equations of Turbulence Modelling

The mean flow characteristics in the standard k-ε turbulence model are simulated
for turbulent flow conditions. This is described by a two-equation model (two
transport equations) which gives a overall description of turbulence [1].
372 A. K. Gupt et al.
  
∂ ∂ ∂ μt ∂ε
(ρε) + (ρεu i ) = μ+
∂t ∂ xi ∂x j σε ∂ x j
ε ε2
+ C1ε (G k + C3ε G b ) − C2ε ρ (8)
k k
  
∂ ∂ ∂ μt ∂k
(ρk) + (ρku i ) = μ+ + G k + G b − ρε − Y M (9)
∂t ∂ xi ∂x j σk ∂ x j

The constants used in Eqs. 8 and 9 are C 1ε = 1.44, C 2ε = 1.92, C μ = 0.09 and
σ ε = 1.3.
Turbulent viscosity μt is given by

k2
μt = ρCμ (10)
ε
where C μ is a constant value and equal to 0.09.

2.7 Cold Mass Flow Ratio (µc )

The ratio of cold mass flow rate ṁc and the inlet fluid mass flow rate ṁin is known
as cold mass flow ratio [13] i.e.

μc = ṁ c /ṁ in (11)

3 CFD Analysis of 2D Axisymmetric Vortex Tube

The chosen geometry is based on the primary VT dimensions which is taken from
already published literature [12]. It provides a comprehensive set of measured data
suitable for validation, and at the same time, their hot throttle is set up as an annular
exit, thus allowing the cold mass flow ratio to be adjusted easily by changing the
pressure.

3.1 Vortex Tube Geometry (2D)

Vortex tube is modelled in ANSYS Workbench [15] as 2D axisymmetric with an

axial annular inlet and a radial annular outlet which provides required rotational
fluid flow. The direction of cold end is opposite to the inlet direction in counter-flow
Thermal Separation in 2D Vortex Tube for a Different Tube Length … 373

Fig. 2. 2D geometry of vortex tube [12]

Fig. 3. 3D view of 2D axisymmetric geometry of vortex tube

vortex tube. The geometry of the vortex tube is shown in Fig. 2, and its 3D view is
shown in Fig. 3.

3.2 Material Properties

The working fluid is taken air as an ideal fluid, and the wall of vortex tube is made
of steel. The properties of air and steel are listed in Table 1.

3.3 Assumptions

The flow behaviour in the vortex tube is highly complex. Therefore, the following
assumptions have been made to study the thermal separations in the vortex tube.
374 A. K. Gupt et al.

Table 1 Material properties

Air Steel
of air and steel [12]
Density (ρ) Ideal gas Density (ρ, 8030
kg/m3 )
Specific heat 10,006.43 Specific heat 502.48
(C g , J/kg K) (C p , J/kg K)
Thermal 0.0242 Thermal 16.27
conductivity conductivity
(w/m K) (w/m K)
Viscosity 1.7894 × 10–5
(kg/m s)
Molecular 28.966
weight
(kg/K mol)

• The working flow is considered to be an ideal gas.

• Vortex tube walls are assumed to be adiabatic.
• No-slip condition is used at the solid boundary.
• Flow is considered to be highly turbulent.
• Flow is steady and therefore the system properties are invariant with time.
• Flow is compressible and therefore the density of the fluid changes significantly.
• The body force is considered to be negligible.
• The flow is considered to be three dimensional and subsonic with uniform fluid
properties at the inlet.

3.4 Selection of Turbulence Model

Distinctive turbulence models like standard k-ε model, RNG k-ε model with swirl,
RNG k-ε model without swirl, RSM turbulence, realizable k-ε model, SST k-ω
model and standard k-ω disturbance model are analysed for count accuracy of the
vortex tube. The RSM turbulence and realizable k-ε model are in nearer concurrence
with estimations than different models. However, realizable k-ε turbulence model is
increasingly appropriate for numerical simulation of the vortex tube stream whenever
considered the expanded computational expenses of RSM turbulence model.
From Table 2, it can be seen that the standard k-ε model and the SST k-ω model
can better match with the available literature results [12]. However, the computation
time for the standard k-ε model is three times less than that of the SST k-ω model.
Hence, “standard k-ε turbulence model” is selected for this present study.
Thermal Separation in 2D Vortex Tube for a Different Tube Length … 375

Table 2 Variations of CFD

Selected model T c T h T
simulated cooling and heating
effect with meshing numbers k-ε standard 8.50 3.50 5.00
for the different k-ε standard k-ε RNG 5.20 1.50 3.70
turbulence model
k-ε RNG (swirl) 4.30 1.20 3.10
k-ω SST 7.70 2.25 5.45
k-ω standard 4.2 1.25 2.95

3.5 The Boundary Conditions and Solver Factors

The boundary conditions are taken as:

• Inlet air mass flow rate 4.0 g/s and inlet total temperature 295 K
• Cold end at 1 atm static pressure (absolute)
• Hot end static pressure: varied
• Cold mass flow ratio μc : 0.3 is used for meshing elements number investigation
and turbulence model selection.
• No-slip condition and adiabatic conditions are applied on walls of the vortex tube.
The primary iteration using the second-order upwind method is used to obtain the
final results. Lower default values of under-relax factors are used to achieve a smaller
convergent target.

3.6 Meshing

The simulation accuracy and time influenced the number of the meshing elements.
The results of the simulated VT parameters are compared using different mesh
element lengths. The number was increased from 30,000 to 150,000 at an increment
of 20,000.
Quadrilateral meshing is used and inflation was applied at the boundaries as shown
in Fig. 4. The number of elements is increased by sizing option. A total number of
elements and nodes are 88,902 and 81,948, respectively. The inlet of tube is named
as “inlet”, cold outlet is named as “cold_outlet”, hot outlet is named as “hot_outlet”
and axis of tube is named as “axis”.

3.7 Mesh Sensitivity Test

Mesh sensitivity is performed using the quadrilateral element under the same condi-
tions (air, μc = 0.3, T in = 295 K, pin = 0.17 MPa, pc = 0.10 MPa). The cooling
effect and heating effect using the k-ε standard turbulence model are shown in Fig. 5.
376 A. K. Gupt et al.

Fig. 4 Meshing of the vortex tube (VT)

Fig. 5 Mesh sensitivity test

When the number of meshing elements is increased from 30,000 to 150,000 with an
approximate increment of 20,000, it is observed that the incremental changes ( ) of
the cooling effect ( T c ) and heating effect (ΔT h ) are within 0.1 °C.
It is evident from mesh sensitivity tests that 90,000 meshing elements provide
compromise between accuracy and computational time for further analysis.

4 Results and Discussions

Pressure, velocity and temperature fields are the characteristics of the vortex tube,
and these characteristics have been studied using the standard k-ε turbulence model
which is discussed in the last section. The strong swirling flow of air inside the vortex
tube results in expansion of air moving from wall of the tube to its axis. Consequently,
the expansion of the ideal gas causes decrease in its temperature at its axis. Total and
static temperature, pressure at inlet, hot outlet and cold outlet for air, μc = 0.3, T in
= 295 K, Pin = 0.17 MPa, Pc = 0.101 MPa are computed and shown below.
Thermal Separation in 2D Vortex Tube for a Different Tube Length … 377

Fig. 6 Velocity vector profile of the vortex tube (2D)

Primary Flow

Secondary Flow

Fig. 7 Velocity streamline profile of the vortex tube (2D)

4.1 Velocity Vector and Velocity Streamline

If we consider two adjacent fluid elements inside the vortex tube, one being an
outward movement while the other being an inward movement, the inertia force
attempts to maintain their angular momentum. At the same time, the shear force
between these two elements slow down their angular speed, as indicated by the
decreasing tangential velocities in the radial direction. Therefore, there must be a
transfer of kinetic energy to the internal energy by some kind of internal friction,
leading to a temperature rise of these fluid elements in the primary flow towards the
hot end. Figure 6 shows the velocity vector profile in the vortex tube.
Figure 7 demonstrates the streamlines flow in the vortex tube with the standard
k–ε model. There exist two flows in the vortex tube chamber which have peripheral
flow (primary flow) and an internal vortex flow (secondary flow). The primary flow
turns out from the hot outlet and the secondary flow, which passes through the entire
length of the tube, reverses its direction of flow and turns out from the cold outlet.

4.2 Pressure Profile of the Vortex Tube (2D)

Along the axial direction of the vortex chamber and the hot tube, the fluid expands
and moves, from higher pressures at inlet towards a lower pressure at the hot end.
378 A. K. Gupt et al.

Fig. 8 Static pressure contour profile of the vortex tube (2D)

Fig. 9 Total pressure contour profile of the vortex tube (2D)

Due to the pressure difference near the hot end, inner layers of primary flow give rise
to secondary flow as it moves towards the cold end. Figure 8 shows that the variation
of static pressure within the vortex tube for air at μc = 0.3, T in = 295 K, Pin =
0.17 MPa, Pc = 0.101 MPa.
Figure 9 shows that the variation of total pressure within the vortex tube for air at
μc = 0.3, T in = 295 K, Pin = 0.17 MPa, Pc = 0.101 MPa.

4.3 Temperature Profile of the Vortex Tube (2D)

The cooling effect appears mainly due to the expansion through the nozzle and the
adiabatic expansion in the spiral motion in the primary and secondary flows. Further,
the heating effect mainly comes from the internal friction due to shear friction.
Figure 10 shows the temperature contour profile of the vortex tube.
In the vortex tube, the fluid undergoes an isentropic expansion when passing
through the nozzle, resulting in a considerable temperature drop. The result (Fig. 10)
shows that the air static temperature at the nozzle outlet area is around 245 K with a
nozzle inlet temperature set at 295 K; this represents a drop of around 50 K through
Thermal Separation in 2D Vortex Tube for a Different Tube Length … 379

Fig. 10 Temperature contour profile of the vortex tube (2D)

Fig. 11 Total temperature contour profile of the vortex tube (2D)

the nozzle. Figure 11 shows the variation of total temperature within the vortex tube
for air, μc = 0.3, T in = 295 K, Pin = 0.17 MPa, Pc = 0.101 MPa.

4.4 Variation of Cooling Effect and Heating Effect with Cold

Mass Flow Ratio (µc ) for Air

Figure 12 shows the comparison of the cooling effect by the simulation and already
published research paper [12]. It is clear that the cooling effect obtained from the
simulation is comparable to the available previous research paper [12]. An important
conclusion can be drawn from Fig. 12 that as the cold mass flow ratio increases,
the cooling effect decreases. Figure 13 presents the variation of heating effect corre-
sponding to different cold mass flow ratio by the simulation and already published
research paper [12]. It is evident that the heating effect obtained from the simulation
is comparable to the available previous research paper [12]. For a higher cold mass
flow or fraction ratio, cooling effect and heating effect obtained are minimal.
380 A. K. Gupt et al.

Fig. 12 Variation of cooling effect ( T c ) with cold mass flow ratios (μc ) for air

Fig. 13 Variations of heating effect ( T h ) with cold mass flow ratios (μc ) for air

4.5 Variation of Cooling Effect and Heating Effect

with Length of the Vortex Tube for Air

Figure 14 shows the variation of cooling and heating effect with length of the vortex
tube for air (μc = 0.3, T in = 295 K, Pin = 0.17 MPa, Pc = 0.101 MPa). It is
observed that the cooling effect and heating effect are maximum, i.e. 8.50 and 3.50 K,
respectively, for the length of 100 mm of the vortex tube.
Thermal Separation in 2D Vortex Tube for a Different Tube Length … 381

Fig. 14 Variations of cooling and heating effects with length of the vortex tube for air

5 Conclusions

A vortex tube CFD model is successfully simulated for thermal separation effect.
The optimal meshing element numbers for providing a good compromise between
the simulation accuracy and the calculation time is chosen. Having assessed various
choices of turbulence models, the standard k-ε turbulence model is selected. The
magnitude of the energy separation increases as the length of the vortex tube increases
to a critical length of 100 mm. However, a further increase or decrease of the vortex
tube length beyond the critical length does not improve the energy separation within
the vortex tube. The vortex tube inlet total temperature is set at 295 K in the simulation,
and the hot end and cold end pressures are varied to fine-tune the cold mass flow ratio
μc from 0.1 to 0.9 and the results of simulation of 2D vortex tube are comparable to
the published research data [12]. The cooling effect and heating effect at μc = 0.3,
T in = 295 K, pin = 0.17 MPa, pc = 0.101 MPa are found to be 8.50 K and 3.50 K,
respectively, for the optimum vortex tube length.
Notations

ΔT c Cooling effect (°C) R Gas constant (kJ/kg K)

ΔT h Heating effect (°C) Pr Prandtl number
μc Cold mass flow ratio (ṁc /ṁin ) γ Specific heat exponent
ṁc Mass flow rate at cold outlet T in Inlet temperature
ṁin Mass inlet at nozzle Tc Cold outlet temperature
v Velocity (m/s) Th Hot outlet temperature
Cp Specific heat capacity (kJ/kg K) Pin VT inlet pressure
h Specific enthalpy (kJ/kg) L Tube length
k Turbulence kinetic energy (m2 /s2 ) τ Stress tensor
(continued)
382 A. K. Gupt et al.

(continued)
ΔT c Cooling effect (°C) R Gas constant (kJ/kg K)
ε Dissipation rate (m2 /s3 ) k eff Effective conductivity
α Thermal diffusivity (cm2 /s) kt Turbulent thermal conductivity
μ Viscosity (kg/m s) E Energy
ν Kinematic viscosity (cm2 /s) μref Reference viscosity (kg/m s)
ρ Density (kg/m3 ) T ref Reference temperature (K)
Pc Pressure at cold outlet S Sutherland constant
Pa Atmospheric pressure σ ε and σ k Prandtl numbers (turbulent)

References

1. Manimaran R (2017) Computational analysis of flow features and energy separation in a

counter- flow vortex tube based on number of inlets. Energy 123:564–578
2. Sadeghiazad M (2017) Experimental and numerical study on the effect of the convergence
angle, injection pressure and injection number on thermal performance of straight vortex tube.
Int J Heat Technol 35(3):651–656
3. Moraveji A, Toghraie D (2017) Computational fluid dynamics simulation of heat transfer and
fluid flow characteristics in a vortex tube by considering the various parameters. Int J Heat
Mass Transf 113:432–443
4. Rafiee SE, Sadeghiazad MM (2016) Experimental and 3D-CFD study on optimization of
control valve diameter for a convergent vortex tube. Front Heat Mass Transf 7(1)
5. Rafiee SE, Sadeghiazad MM (2016a) Three-dimensional numerical investigation of the
separation process in a vortex tube at different operating conditions. J Mar Sci Appl 15(2):1–10
6. Rafiee SE, Sadeghiazad MM (2016b) Heat and mass transfer between cold and hot vortex
cores inside Ranque-Hilsch vortex tube-optimization of hot tube length. Int J Heat Technol
34(1):31–38
7. Manimaran R (2016) Computational analysis of energy separation in a counter-flow vortex
tube based on inlet shape and aspect ratio. Energy 107:17–28
8. Rafiee SE, Sadeghiazad MM (2017) Experimental and 3D CFD investigation on heat transfer
and energy separation inside a counter flow vortex tube using different shapes of hot control
valves. Appl Therm Eng 110:648–664
9. Rafiee SE, Sadeghiazad M (2016c) Three-dimensional computational prediction of vortex
separation phenomenon inside the Ranque-Hilsch vortex tube. Aviation 20(1):21–31
10. Dutta T, Sinhamahapatra KP, Bandyopdhyay SS (2010) Comparison of different turbulence
models in predicting the temperature separation in a Ranque-Hilsch vortex tube. Int J Refrig
33(4):783–792
11. Guo X, Zhang B (2018) Computational investigation of precessing vortex breakdown and
energy separation in a Ranque–Hilsch vortex tube Étude numérique de la rupture du vortex et
de la séparation énergétique dans un tube vortex de Ranque-Hilsch. Int J Refrig 85:42–57
12. Aljuwayhel NF, Nellis GF, Klein SA (2005) Parametric and internal study of the vortex tube
using a CFD model ‘tres d’ e ‘coulement a “l” inte’ rieur d ‘un tube vortex a Etude des parame’
sur la dynamique des fluides nume ‘rise’ e l ‘aide d’ un modele fonde. Int J Refrig 28:442–450
13. Behera U, Paul PJ, Kasthurirengan S, Ram SN, Dinesh K, Jacob S (2005) CFD analysis and
experimental investigations towards optimizing the parameters of Ranque-Hilsch vortex tube.
Int J Heat Mass Transf 48(10):1961–1973
Thermal Separation in 2D Vortex Tube for a Different Tube Length … 383

14. Bazgir A, Nabhani N (2018) Investigation of temperature separation inside various models of
Ranque-Hilsch vortex tube: convergent, straight, and divergent with the help of computational
fluid dynamic approach. J Therm Sci Eng Appl 10(5):051013
15. ANSYS 19.0 Documentation, ANSYS India Private Limited. https://www.ansys.com
GA and PSO Based Optimization
for Benchmark Thermal System Using
Wavelet-Based MRPID Controller

Abhas Kanungo, Monika Mittal, and Lillie Dewan

Abstract A high performance thermal control system that can handle tempera-
ture uncertainties is significant for various applications. The characteristics of such
systems can be improved by tuning the control parameters of multi-resolution
proportional, integral, and derivative (MRPID) controller, which is still a challenge.
Therefore, this paper presents the two optimization techniques for optimizing the
parameters of a thermal system MRPID controller. First is particle swarm optimiza-
tion (PSO), and the second is genetic algorithm (GA). This controller controls the
temperature uncertainty error between the reference and measured temperature of the
thermal system by decomposing it into different frequency segments at non-identical
sizes of multi-wavelet transform (MWT). Different wavelet coefficients of this error
decomposition are used to obtain the control signal. The controller is wavelet-based
and is simulated in MATLAB 2015a. Analysis of the proposed techniques has been
verified with the help of simulations using SIMULINK followed by conclusion.

Keywords Thermal system · MRPID controller · GA and PSO · MWT

algorithm · Wavelet coefficients

1 Introduction

Thermal system is very commonly used in power industry. Hence, it becomes very
important to control its thermal response in stable input conditions [1]. In thermal
power plant, it gives optimal performance with respect to overall cost of installation
due to low cost of fuel making it very important in power generation industry [2].
If temperature is not regulated in a thermal system properly, the performance of
the associated controller degrades causing overall system performance degradation

A. Kanungo (B)
KIET Group of Institutions, Delhi-NCR, Ghaziabad, India
e-mail: [email protected]
M. Mittal · L. Dewan
National Institute of Technology Kurukshetra, Kurukshetra, India

© Springer Nature Singapore Pte Ltd. 2021 385

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_36
386 A. Kanungo et al.

and more systemic errors [3]. Most commonly used controller for a typical thermal
system is PID controller having combination of three gain parameters K P , K I , and K D
[4]. Application of PID controller requires proper tuning of these gain parameters,
which is still a challenge. Zeigler–Nichols method is most commonly used method to
tune parameters of PID controller [5]. Sometimes, a PID controller may be referred
for nonlinear systems too due to its simplicity and structure [6]. But, performance of
PID controller degrades in the presence of uncertainty because it is very sensitive to
noise and disturbances. The fixed structure PID controller with constant parameters’
values is generally tuned for single operating condition that may cause nonlinearity in
the characteristics of dependent systems. Also, characteristics may differ in different
operating conditions, and they may deflect from the desired value [7]. Additionally,
it is not a better choice to be used for higher-order systems due to increase in number
of pole locations. Intelligent control techniques, referred to as automated tuning of
the PID controller, can overcome the above shortcomings. Optimization of PID gain
parameters is one such approach. One of the important such optimization approach
is based on using fuzzy controller. Discrete version of PID controller can be referred
to as fuzzy PID controller. It follows the linear structure similar to PID controller to
control proportional, derivative, and integral gain values with the difference that the
fuzzy logic has self-tuned control gains. Fuzzy logic controller is implemented with
the help of fuzzy rule base by following conventional PID controller characteristics
increasing complexity of the system [8]. For instance, speed of knife roller and
paper roller of paper cutter can be controlled by a fuzzy PID controller in which
the control algorithm is realized a PI regulator, and fuzzy rule base is realized by a
programmable logic controller (PLC). As a result, both the problems of instability
and accuracy are resolved [9]. There are various other intelligent control techniques
such as neural network, GA, firefly algorithm, and Ant colony optimization, for
optimizing a PID controller. But, these techniques are very sensitive to disturbances
[10]. All these techniques have limited applications due to shortcomings viz slow
control, lack of efficiency, and poor handling of system nonlinearities. Hence, use
of wavelet-based MRPID controller is motivated which also overcomes the effect of
uncertainty [11]. In this controller, multi-resolution property of wavelet is utilized,
which allows to give resolution in both frequency and time domain. Also, wavelets
are very useful to separate fine details of the signal [12, 13]. Wavelet-based MRPID
controller has been widely used for thermal system as reported in the recent literature
because it gives good approximation of function by using only few coefficients [14].
Generally, it breakdowns the error signal in to various frequency components with
appropriate gain values that are further optimized by optimization techniques. It can
also compress and denoise the signal without degradation in signal amplitude [15]. In
this paper, tuning of existing MRPID controller is proposed to be optimized with GA
and PSO. The biggest advantage of MRPID controller is that it is also very effective
in presence of noise [16]. Therefore, uncertainty in the dynamic systems motivated
this research work.
The paper is organized as follows; structure of a benchmark thermal system is
explained in Sect. 2, proposed methodology is presented in Sect. 3 along with MRPID
GA and PSO Based Optimization for Benchmark Thermal System … 387

controller and MWT algorithm. Finally, simulation results have been explained,
followed by conclusion at the end.

2 Benchmark Thermal System

Feedback structure of thermal system, shown in Fig. 1, has been considered as the
benchmark problem. In this system, surrounding air is sucked into the electric heater
which then is made to flow through air flow pipe. A bead thermistor is used as
a temperature sensor to measure the temperature of the heated outgoing air. To
get better performance of the system, difference between reference temperature and
measured temperature should ideally be zero. A nonzero value of difference between
the two temperatures results into an error signal. This error signal is decomposed
using wavelet transform, coefficients of which are then used to obtain MRPID control
signal. The control signal is given to the thermal system which further takes necessary
action to minimize the error. Optimization of MRPID controller further enhances its
performance. In this paper, two optimization techniques (GA and PSO) are proposed
to be applied for this purpose. The schematic diagram of thermal control system has
been presented in Fig. 1 [17].
The proposed MRPID controller is applied in thermal system having transfer
function G(s) given in [7]. Different optimization techniques are also implemented
using MATLAB/Simulink. The performance of proposed techniques is compared by
performance curve.
The input output relationship of thermal system is given by the transfer function
as in Eq. (1),

(0.0882s + 1.26)e−sTd
G(s) = (1)
0.00073s 3 + 0.0153s 2 + 0.585s + 1

Air flow
Bead
Electric heater
Thermistor

Air Inlet
Control
Signal System
output
MRPID CONTROLLER

Optimization
technique

Fig. 1 General control structure of thermal control system

388 A. Kanungo et al.

Here, Td is denoted as time delay. The thermal systems is a third-order system

with this time delay.

3 Proposed Methodology

This part describes MWT algorithm, MRPID controller, the proposed optimiza-
tion techniques, and their comparison. Multi-resolution property of wavelet has
been utilized to decompose the error signal which is estimated from reference and
measured value of temperature.

3.1 MWT Algorithm

There are generally two types of wavelet viz scalar wavelet and multi-wavelet. Scalar
wavelet is a combination of single wavelet and single scaling function. However,
multi-wavelet may have two or more scaling and wavelet functions. MWT is a gener-
alization of scalar wavelet, and bases of MWT are objective-based on multi-resolution
analysis (MRA). MWT decomposes the error signal into two high-pass and low-pass
sub-bands for each level of decomposition. With MRA structure, wavelet coefficient
can be computed by Mallat’s algorithm [18]. Digital filtering technique has been used
in MWT for time scale representation. The set of scaling function can be written as

(t) = [φ1 (t), φ1 (t), . . . , φr (t)]T (2)

Here, multi-scaling function is represented by (t), and its transpose is denoted

by T, and wavelet function is presented by (t) as

(t) = [ψ1 (t), ψ1 (t), . . . , ψr (t)]T (3)

The dilation and wavelet equation can be described as

√ 
(t) = 2 G k (2t − k) (4)
k
√ 
(t) = 2 Hk (2t − k) (5)
k

Multi-wavelet filter bank is described by pair of low-pass filter matrix (G k ) and

high-pass filter matrix (Hk ). More, the matrix elements more is the degree of freedom.
In multi-wavelet function, filter banks contain two channel. So for each level of
decomposition, two set of wavelet coefficients and two set of scalar coefficient are
obtained as shown in Fig. 2.
GA and PSO Based Optimization for Benchmark Thermal System … 389

Fig. 2 Multi-wavelet transforms decomposition

The multi-wavelets high-pass and low-pass filter banks are described as H and
G, and their dual filter banks are presented as H ∗ and G ∗ . Decimation and zero-
padding are mentioned as 2 ↓ and 2 ↑. The MWT decomposition algorithm is given
by Eqs. (6) and (7),

c j−1,k = G n c j,2k+n (6)
n

d j−1,k = G n d j,2k+n (7)
n

 T
The input vector is described as c1,0,k , c2,0,k . Low-pass sub-band signal is given
 T
as vector c1,−1,k , c2,−1,k , known as multi-scaling coefficients of the input signal.
 T
The high-pass sub-band signal is represented as vector d1,−1,k , d2,−1,k , known as
multi-wavelet coefficients of the input signal. Same process will be repeated for the
successive decomposition to get second-level multi-scaling coefficient from first-
level coefficients. For specific application, choice of optimum wavelet function is
dependent on type of the input signal. In MRPID controller, MWT has been applied
on error signal, and feedback structure of the controller has been followed. Mother
wavelet selection and levels of decomposition is dependent on temperature error
signal. Optimal level of decomposition is obtained at an appropriate chosen resolution
of MWT at each level. MRPID controller is presented next.

3.2 The MRPID Controller

Multi-resolution property of wavelet has been utilized to decompose the error signal
as a linear combination of various compact waves known as wavelets. Wavelet trans-
form provides greater advantage as compared to conventional PID controller in terms
390 A. Kanungo et al.

Fig. 3 Proposed structure of controlled thermal system

of compensating the error [19]. So, control output of PID controller combining three
terms K p , K i and K d (proportional, derivative and integral gains) is given as,

d
u PID = K p e + K i ∫ edt + K d e (8)
dt
In frequency domain, K p and K I terms are utilized to capture low-frequency
and high-frequency uncertainty data, respectively. Similarly, error signal has been
decomposed by MRPID controller to obtain different scale components as

f (x) = f H (x) + f M1 (x) + · · · + f M(n−1) (x) + f L (x) (9)

Each component is scaled or weighted by separate gains. These are then combined
to generate controller output signal u by relating wavelet decomposition to the error
signal just as aggregate impact of numerous fundamental wanderers [19]. Every
component is scaled by individual gain at the chosen operating point to produce
control signal (u) as,

u MRPID = K H eH + K M eM1 + · · · + K M(n−1) eM(n−1) + K L eL (10)

Figure 3 represented the proposed structure of controlled thermal system.

The MRPID controller decomposes the uncertainly error in temperature into the
detail and approximation components. A MRPID controller is a combination of three
parameters that is k h , k m , and k l . Decomposition of error signal is done up to the
second level of decomposition of MWT. As mentioned earlier, gain parameters K H
and K M are used for optimizing higher and medium frequencies, and gain K L is used
for lower frequencies.

3.3 GA Optimization

Genetic algorithm (GA) is inquired to explore about search space comparable to

evolution in nature in certain respects. GA changes the potential solution in search
space by the application of probabilistic rules using fitness function to find out the
GA and PSO Based Optimization for Benchmark Thermal System … 391

Fig. 4 Steps involved in GA optimization

solution. GA depends on set of universal solution called as global population infor-

mation. In GA, fitness function is defined according to the specified problem. Then,
the selection of best solution takes place by combining the selected solution or parent
classes through which child classes or new solutions have been found. It is done by
the genetic operator crossover and mutation. This process will be continued till the
optimal solution of specified problem is achieved according to the fitness function
[20]. Various steps involved in genetic algorithm are explained in Fig. 4.

3.4 PSO Optimization

PSO is an artificial intelligence-based computational method to optimize the given

problem to improve the possible solution with respect to a chosen cost function.
It is stochastic evolutionary computation optimization technique. But unlike GA,
PSO do not use operators like crossover and mutation for recombination. PSO yields
solutions known as particles. Each particle keeps track of its location in the problem
domain. These locations correspond to the solutions expected to be best for the
specific problem. The velocity of each particle is changed on the basis of Pbest and
Ibest locations. The stored fitness value is called as Pbest and best value which is
obtained by PSO optimizer is known as Ibest. Gain parameter optimization by PSO
technique can be referred from [21]. Various steps involved in PSO optimization are
represented in Fig. 5 [22].
392 A. Kanungo et al.

Fig. 5 Steps involved in

PSO optimization Initialize the

Evaluate fitness of
individual particle

Modify velocities
based on previous

Terminate when
condition is reached

End

3.5 Comparison Between GA and PSO Optimization

Particle swarm optimization (PSO) is heuristically search method which is influ-

enced by swarming and collective nature of biological solutions. PSO and GA both
are biologically inspired methods. Both follows heuristic population based search
algorithms. By mixing of probabilistic and deterministic rules, different new solu-
tions have been found from one set of point to another set of point in repetitive
iterations. These set of points can be called as solutions or population. Intuitive-
ness, easy implementation and effective to solve nonlinear problems property of GA
utilizes in academia and industry. Expensive computational cost is biggest disad-
vantage of GA. PSO has better computational efficiency by the application of better
statistical analysis and hypothesis testing. In this paper, comparison of both these
optimization techniques has been shown for proposed MRPID controller. Benchmark
thermal system has been taken in this paper to review the comparison of obtained
results.

4 Results and Discussion

In this section, analysis of thermal system with proposed MRPID controller has
been discussed. GA and PSO optimization techniques have been used to optimize
the gain values of MRPID controller. Simulated results are evaluated with the help of
MATLAB@2015a software. Reference value of temperature is taken as 268 °C. The
GA and PSO Based Optimization for Benchmark Thermal System … 393

Fig. 6 Comparative analysis of thermal system in stable input condition

comparative results of two optimization techniques under stable input conditions has
been shown in Fig. 6.
Different transient- and steady-state characteristics, considered in this paper, are
explained below.
Rise time is defined as the time taken by the output to reach from 10 to 90% of
its final value, peak time is defined as the time taken in reaching first peak of the
response also known as peak overshoot, settling time is defined as time taken by the
output to settle down within 2% of its final value, overshoot is the output value over
and above its final value during the first peak, and peak value is the highest value
attained by the output.
The performance of a system is better when rise time is less, peak time is earlier,
and settling time is minimum. Also, overshoot should be within the specified limit,
and peak value should not exceed the chosen reference value.
Comparative analysis of these two optimization techniques with respect to
transient- and steady-state characteristics has been given in Table 1.
As given in Table 1, the specifications of rise time, peak time, and settling time
are less for GA as compared to PSO. Peak value is equal to the reference temperature
in both the cases. Settling time is less in GA. Hence, it can be clearly seen that GA
optimization technique performs better in terms of time domain characteristics as
compared to PSO optimization.

Table 1 Time domain

Characteristics GA PSO
specifications for GA and
PSO optimization Rise time (s) 0.8 1.65
Peak time (s) 1.35 2.5
Peak (°C) 268 267.99
Settling time (s) 1.102 1.9950
Overshoot (%) 0.05 0.04
394 A. Kanungo et al.

5 Conclusion

Thermal response of a benchmark thermal system is evaluated using optimized

MRPID controller. Two optimization techniques viz GA and PSO are used for the
purpose. In MRPID controller, MWT is used for decomposing the error signal, and
multi-resolution property of the wavelet has been utilized. Further, gain values have
been optimized by using GA and PSO to enhance thermal performance of the system.
Simulations are carried out, and results obtained with two optimization techniques
are compared in terms of system transient- and steady-state characteristics. It has
been shown that GA gives better system performance. In future, hybrid optimization
techniques may be used to improve performance of this thermal system further.

References

1. Jino Y, Matsumoto S, Kamito A (2006) Risk based model predictive control with hybrid
system structure and its application to thermal power plants. In: SICE-ICASE international
joint conference. https://doi.org/10.1109/sice.2006.315571
2. Emami-Naeini A, Ebert JL, de Roover D, Kosut RL, Dettori M, Porter LML, Ghosal S
(2003) Modeling and control of distributed thermal systems. IEEE Trans Control Syst Technol
11(5):668–683. https://doi.org/10.1109/tcst.2003.816411
3. Jackson ME, Shtessel YB (1998) Sliding mode thermal control system for space station furnace
facility. IEEE Trans Control Syst Technol 6(5):612–622. https://doi.org/10.1109/87.709496
4. Dong Z, Su Y, Yan X (2009) Temperature control system of the thermal analyzer based on
fuzzy PID controller. In: Ninth international conference on hybrid intelligent systems. https://
doi.org/10.1109/.123
5. Hang CC, Astrom KJ, Ho WK (1991) Refinements of the Ziegler–Nichols tuning formula.
Proc Inst Electr Eng Control Theory Appl 138(2):111–118
6. Ziegler JG, Nichols NB (1942) Optimal settings for automatic controllers. Trans ASME
64(11):759–768
7. Kanungo A, Dewan L, Mittal M (2015) Performance improvement of a thermal system PID
controller using Haar wavelet based denoising. In: International conference on industrial
instrumentation and control (ICIC). https://doi.org/10.1109/iic.2015.7150773
8. Tang KS, Man KF, Chen G, Kwong S (2001) An optimal fuzzy PID controller. IEEE Trans Ind
Electron 48(4):757–765. https://doi.org/10.1109/41.937407
9. Liu X, Song L, Wang F (2011) The fuzzy-PID application in speed control system of the
paper cutter. In: IEEE 10th international conference on cognitive informatics and cognitive
computing (ICCI-CC’11). https://doi.org/10.1109/coginf.2011.6016169
10. Om BH, Rajamayyoor S, Shreeraman PR (2012) PID controller tuning techniques: a review. J
Control Eng Technol (JCET) 2:168–176
11. Khan MASK, Azizur Rahman M (2011) Implementation of wavelet-based controller for battery
storage system of hybrid electric vehicles. IEEE Trans Ind Appl 47(5):2241–2249
12. Khan MASK, Azizur Rahman M (2008) Implementation of a new wavelet controller for interior
permanent-magnet motor drives. IEEE Trans Ind Appl 44(6):1957–1965
13. Dewan L, Garg M (2012) Non-recursive Haar connection coefficients based approach for linear
optimal control. J Optim Theory Appl (JOTA) 153(2)
14. Parvez S, Zhiqiang G (2005) A wavelet-based multiresolution PID controller. IEEE Trans Ind
Appl 41(2):537–543
15. Vajpayee M, Vineet S, Tiwari AP (2016) A multi resolution wavelet based subspace
identification. Int J IFAC 49(1):247–253
GA and PSO Based Optimization for Benchmark Thermal System … 395

16. Khan M, Abdesh SK, Azizur Rahman M (2010) A novel neuro-wavelet-based self-tuned
wavelet controller for IPM motor drives. IEEE Trans Ind Appl 46(3):1194–1203
17. Kanungo A, Mittal M, Lillie D (2018) Comparison of Haar and Daubechies wavelet denoising
for PID controlled thermal system. J Adv Res Dyn Control Syst 9:2405–2411
18. Guo L, Rivero D, Pazos A (2010) Epileptic seizure detection using multiwavelet transform
based approximate entropy and artificial neural networks. J Neurosci Methods 193:156–163
19. Parvez S (2003) International advanced control techniques for motion control problem. Ph.D.
dissertation, Cleveland State University, Cleveland
20. Fan L, Joo EM (2009) Design for auto-tuning PID controller based on genetic algorithms. In:
4th IEEE conference on industrial electronics and applications. https://doi.org/10.1109/iciea.
2009.5138538
21. Seekuka J, Rattanawaorahirunkul R, Sansri S, Sangsuriyan S, Prakonsant A (2016) AGC using
particle swarm optimization based PID controller design for two area power system. In: Inter-
national computer science and engineering conference (ICSEC). https://doi.org/10.1109/icsec.
2016.7859951
22. Hassan R, Cohanim B, de Weck O, Venter G (2005) A comparison of particle swarm optimiza-
tion and the genetic algorithm. In: 46th AIAA/ASME/ASCE/AHS/ASC structures, structural
dynamics and materials conference. https://doi.org/10.2514/6.2005-1897
The Stablization of Soil Using Treated
Bagasse Ash

Manish Chaudhary, Mohit Rathore, Manoj Bhatt, and Deepak Verma

Abstract In today’s era, the problematic soils (swelling soil, soft soil, etc.) cause
the civil engineering structures to fail due to the excessive settlement and low bearing
capacity. The main function of the soil is to give satisfactory support to the civil struc-
tures, and for this, the soil should own enough stability under unfavorable climatic
conditions. We replaced problematic soil with some other strong materials like hard
soil, rocks, etc. and treated with any other material. Agricultural waste from the
industries is increasing. So we have to work for effective utilization or do investiga-
tion on the industrial agricultural waste. In our research, we have found that when we
mix the treated sugarcane bagasse ash with the soil in different proportions, we get
the positive result. The laboratory tests are performed at various proportions (5, 10,
15, 20% of total solid mass) with 10% of NaOH, and the CBR value is increasing.
By the use of sugarcane bagasse ash, we not only stabilize the soil, but we also have
to reduce the agricultural industrial waste (sugarcane bagasse ash).

Keywords CBR · Sugarcane bagasse ash · Agricultural industrial waste ·

Stabilization · Adverse

1 Introduction

The soil properties are enhanced by some techniques, and these techniques are known
as ground improvement techniques. GIT can be done by many types like adding

M. Chaudhary (B) · M. Rathore · M. Bhatt

Department of Civil Engineering, Graphic Era Hill University, Dehradun, Uttarakhand, India
e-mail: [email protected]
M. Rathore
e-mail: [email protected]
D. Verma
Department of Mechanical Engineering, Graphic Era Hill University, Dehradun, Uttarakhand,
India
e-mail: [email protected]; [email protected]

© Springer Nature Singapore Pte Ltd. 2021 397

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_37
398 M. Chaudhary et al.

admixture, reinforcement or mechanical methods. The soft soil is defined as the soil
which will have low shear strength, low bearing capacity or high moisture content.
In India, the CO2 emission from the industries is more (almost 2 millions tonnes
per year) which increases the green house gas and possesses the drastic change in the
global warming [1]. The sugarcane bagasse ash also possesses serious environmental
problem, but it is more effective for soil stabilization. It is produced by burning the
bagasse in the sugar mills. SBA is very effective in soil stabilization, or it also reduces
the lime or cement content in the process of soil stabilization [2].
According to our result, we have investigated that soil properties are improved
when SBA is mixing to the soil. Alkalization of SBA makes it alternative binder to
the traditional binder. Alkalization of SBA is done with the help of NaOH in this
research.

1.1 Soil

It consists mostly of mineral matter formed by the breakdown of rocks, by the oper-
ation of water, temperature, ice crystals, pressure or by animal and plant life. Based
on the grain size of soil particles, soils have been classified as silt, clay, sand and
gravel. The attribute of soil particles depends on the shape surface texture, chemical
composition and size. Dry density and moisture affect the engineering department
of a soil mass [3]. In our research, the soil sample was collected near around the
Gehu Campus. The sample was sealed immediately so the soil characteristics do not
change or there is no loss in the natural moisture content. Then, the soil was dried in
the air, and the cramps break down. Then, the soil sample is passed through 2.36-mm
sieve.

1.2 Sugarcane Bagasse Ash (SBA)

SBA is a sugarcane waste product. The discarding of this causes the environmental
hazards near around the sugar mills or industries. Sugarcane is the prime crop grown
in many countries, and its total production is about 1600 tonnes. The production
of sugarcane in India is about 350 million tonnes per year [4]. After extraction of
all sugar, the 45% (approximately) fibrous remnant was obtained, which is further
reused in the same sugar mill/industries as a fuel in boilers for the purpose of heat
generation leaving behind 7–10% ash as a waste, known as SBA (Figs. 1, 2 and 3;
Table 1).
The Stablization of Soil Using Treated Bagasse Ash 399

Fig. 1 Bagasse ash

Fig. 2 Production of bagasse ash

1.3 Sodium Hydroxide

It is an inorganic compound which is also known as lye and caustic soda. It contains
sodium cations Na+ and hydroxide anions OH− which is generally white solid ionic
compound (Fig. 4).
NaOH is a highly corrosive base and alkali that decompose protein at ordinary.
It is soluble in water and readily absorbs the CO2 and moisture from the air. It
crystallizes from water solution between 12.5 and 62 °C. Commercially available
NaOH is monohydrate [5]. It crystallizes water solution between 12.5 and 62 °C.

NaOH + H2 O = Na + and OH− ions.

This reaction is known as exothermic reaction, in which the heat is released.

NaOH is considered as a strong base which is fully or completely disassociated in
400 M. Chaudhary et al.

Fig. 3 Bagasse ash production in Doiwala Sugar company LTD, Dehradun

Table 1 Component of
Component Weight fraction
bagasse ash
SiO2 (silicon dioxide) 78.34
Al2 O3 (aluminum oxide) 8.55
Fe2 O3 (iron III oxide) 3.61
CaO (calcium oxide) 2.15
Na2 O (sodium oxide) 0.12
K2 O (potassium oxide) 3.46
MnO (manganese oxide) 0.13
TiO2 (titanium dioxide) 0.5
MgO (magnesium oxide) 1.65

aqueous solution. It is corrosive and causes chemical burns. In contact to your skin,
immediately wash it with a large quantity of water. Add sodium hydroxide to water;
do not add water to solid sodium hydroxide [6].
Sodium hydroxide reaction with water:

NaOH + H2 O → Na+ + OH− + H2 O + Heat

The Stablization of Soil Using Treated Bagasse Ash 401

Fig. 4 NaOH in hard crystal form

1.3.1 Potential Health Effect of NaOH

• Inhalation: Can cause irritation of the nose and throat

• Contact to skin: Cause blistering, pain, redness and burns
• Contact to eye: Burns with redness, swelling, blurred vision and pain.
• Ingestion: Burns the tongue, lips and stomach (Figs. 5 and 6) [7].

2 Literature Survey

According to authors, the swelling and shrinkage of the soil become a challenge
to the civil engineers to do construction because it causes settlement resulting in
damage to the constructed work. Authors’ aim is to study the effect of utilization
of industrial waste (by product of sugar factory) on some engineering properties of
soil. The tests are performed at different percentages by dry weight of soil. Their
results showed a decrease in plasticity, swelling pressure and swelling potential of
soil. The main purpose of soil stabilization is to increase strength of an existing soil
to enhance its load bearing capacity, permeability improvement and enhancement of
soil resistance to the process of weathering [4, 8].

3 Methodology

To improve soil stability and to find index property of soil, we perform these tests.
402 M. Chaudhary et al.

Fig. 5 Before use of NaOH

Fig. 6 After use of NaOH

The Stablization of Soil Using Treated Bagasse Ash 403

3.1 Method of Oven Dry

According to IS: 2720 (part-II)-1973. This test is performed to calculate the water
content in the soil.

3.1.1 Apparatus Required

• Thermostatically controlled oven maintained at a temperature of 110 ± 5 °C

• Weighing balance with accuracy of 0.04% of the weight of soil taken
• Air tight container
• Gloves

3.1.2 Formula Used

Water content, w = (weight of water/weight of dry soil)
× 100%w = [W2 − W3 ]/[W3 − W1 ] ∗ 100%

3.2 Sieve Analysis

A sieve analysis also known as gradation test is used to assess the particle size
distribution of a granular material by allowing material to pass through a series of
sieve.

3.2.1 Apparatus Used

• Sieve Shaker
• Different types of sieve arranged in ascending order - 0.075 mm < 0.150 mm <
0.300 mm < 0.425 mm < 0.600 mm < 1.7 mm < 2.36 mm < 4.75 mm

= = = = = = = = = = = = = = = = = = = = → Increasing order.

3.2.2 Formula Used

Coefficient of uniformity (U) = D60/D10 Coefficient of curvature (Cc) =

(D30)2 /D10*D60.
404 M. Chaudhary et al.

3.3 Specific Gravity by Pycnometer

It is used to define the density or a weight of a liquid as compared to density of equal

volume of water at a specify temperature. Usually, we take 4 °C because it allows
water to assume its maxi. density.

3.3.1 Apparatus Used

• Pycnometer of about 1 L capacity

• Weighing balance, with an accuracy of 1 g
• Glass rod
• Vacuum pump
• Oven.

3.3.2 Formula Used

G = Weight of soil solids/Weight of water of volume equal to that of soil solids
.
G = (M2 −M1 )/((M2 −M1 )−(M3 −M4 )).

3.4 Liquid Limit, Plastic Limit Test

The Atterberg limits are basic measures of the critical water content of a fine-grained
soil, its shrinkage limit, plastic limit and liquid limit. It is used to distinguish between
clay and silt. Consistency or Atterberg’s limit is defined as when water content
changes from one state to the other.

3.4.1 Apparatus Used

• Casagrande’s apparatus
• Grooving tools
• Electric oven
• Glass sheet
• Spatula
• IS sieve (425 µm)
• Weighing machine
• Wash bottle.
The Stablization of Soil Using Treated Bagasse Ash 405

3.4.2 Formula Used

Water content = weight of water/weight of dry soil Plasticity Index (Ip) = Liquid
limit − Plastic limit.
Liquidity Index (IL) = (Natural water content − plastic limit)/plastic index.

3.5 Standard Proctor Test

This test is used to find out the optimum moisture content (OMC) for the soils.
It is the test which is especially used to determine the relationship between water
content and dry unit weight of soil to establish the maximum density of a soil
needed for filling the area. We apply 25 blows from a rammer dropped from a
height of a 300 mm above the soil [6].

3.5.1 Apparatus Used

• Compaction mold (1000 ml)

• Rammer (2.6 kg)
• Base plate (detachable)
• Collar (60 mm high)
• IS sieve (4.75 mm)
• Electric oven
• Weighing machine
• Mixing pan
• Straight edge
• Graduated glass jar
• Spoons, trowels and mixing tools.

3.5.2 Formula Used

Bulk density = Weight of soil/volume of mold Dry density = Bulk density/(1 +

water content).

3.6 California Bearing Ratio (CBR) Test

It is developed by the California Division of Highway as a technique of classifying

and evaluating soil base course and subgrade material for the pavement. This test is
useful to determine the material properties of the soil.
406 M. Chaudhary et al.

CBR test is a penetration test meant for the evaluation of subgrade strength of
roads and pavements. The results obtained by these tests are used with the empirical
curves to determine the thickness of the pavement and its component layer [7, 8].

3.6.1 Apparatus Used

• Mold
• Spatula
• Surcharge weight
• Dial gauges
• IS sieves
• Penetration plunger
• Loading machine
• Weighing machine.

3.6.2 Formula Used

CBR% = (Load taken by soil at a penetration of 2.5 mm or 5 mm/Standard load

taken by aggregate at same penetration)*100%.
CBR% for 2.5 mm = (P1/1370)*100% CBR% for 5 mm = (P2/2055)*100%.

4 Tools Used

See Figs. 7, 8, 9, 10, 11, 12, 13, 14, 15 and 16.

5 Result and Discussion

After the determination of basic properties of well-graded coarse sand, soil stabilizes
with bagasse ash and the strength parameters maximum dry density (MDD) and
California bearing ratio (CBR) were determined by conducting compaction, CBR
(Table 2).
By sieve analysis, we found soil comes under grading zone I and soil is well-graded
coarse sand (Figs. 17 and 18).
For 5%,10%,15% and 20% treated (with 5% NaOH) bagasse ash, the value
of optimum moisture content (OMC) decreases by 1%,3%,3%,6%, respec-
tively, and the values of maximum dry density (MDD) increases by <
1%(0.18),3%,10.10%,18.46%, respectively, as compared to untreated bagasse ash
(Figs. 19, 20 and 21).
The Stablization of Soil Using Treated Bagasse Ash 407

Fig. 7 Weighing machine,

heavy and light compaction
rammer

Fig. 8 Tray, light compaction rammer, trowel

For 5%,10%,15% and 20% treated (10% NaOH) bagasse ash, the value of OMC
decreases by 1%,0%,2%,5%, respectively, and the values of MDD increased by
1.8%,1.96%,9.35%,20.16%, respectively, as compared to untreated bagasse ash
(Figs. 22 and 23).
CBR values for treated (with 5% NaOH) bagasse ash increased by 18.98% for
5% treated bagasse ash after as we increased concentration of bagasse ash value of
CBR continuously decreases (Fig. 24).
CBR values for treated (10% NaOH) bagasse ash decreases by 17.72% for 5%
treated bagasse ash after as we increased concentration of bagasse ash value of CBR
continuously decreases.
408 M. Chaudhary et al.

Fig. 9 Weighing machine

Fig. 10 Sieve

6 Conclusion

• For well-graded coarse sand, the value of optimum moisture content is continu-
ously increased as we increase the bagasse ash concentration & the value of MDD
continuously decreases.
The Stablization of Soil Using Treated Bagasse Ash 409

Fig. 11 Container used for oven

Fig. 12 Assembling of standard proctor test mold

• For standard proctor test (SPT), the values of MDD & OMC are similar of both (
with 5% & 10% NaOH) treated bagasse ash.
• We obtained maximum CBR % at 5% concentration of treated (with 5% NaOH)
bagasse ash.
• For compaction, we obtained best results at treated bagasse ash with 5% NaOH.
410 M. Chaudhary et al.

Fig. 13 Leveling of top

surface of soil

Fig. 14 Mixing of soil with water

The Stablization of Soil Using Treated Bagasse Ash 411

Fig. 15 Deassembling of
CBR mold

Fig. 16 During compaction

of soil in CBR test
412 M. Chaudhary et al.

Table 2 Soil index properties

Soil properties Values
Water content 33.36%
Liquid limit 50%
Plastic limit 35%
Plasticity index 14.96
Consistency index 1.109
Liquidity index 12%
Pycnometer test 2.23
Sieve analysis Grading zone I (coarse sand)
Coefficient of uniformity 11.56 (Well graded soil)
Coefficient of curvature 1.0 (Well graded soil)
Standard proctor test (OMC & 14% & 1.88 g/cc
MDD)
CBR% (California bearing 11.32%
ratio)

Graph 1 Liquid limit Liquid Limit

100
W/C(%)

10

1
1 10 100
Number of Blows

7 Future Scope

Bagasse is a waste product and having disposal problem. Due to continuously high
demand of sugar, the production of bagasse ash is continuously increasing, so we
need some reliable use of bagasse ash in nearby future. Bagasse is a cheap material,
and it can be used as a alternative of cement in construction and can be used for
filling of embankment.
The Stablization of Soil Using Treated Bagasse Ash 413

Graph 2 Gradation curve Grain Size Distribution Curve

Percent finer by weight

10
Grains size in mm

Parent Soil
Maximum Dry Density

10 15 20
Optimum Moisture Content (%)

Graph 3 Standard proctor test for parent soil

Treated Bagasse Ash with 5% NaOH

Maximum Dry Density

5% Bagasse Ash
10% Bagasse Ash
15% Bagasse Ash
20% Bagasse Ash

10 20 30 40
Optimum Moisture Content (%)

Graph 4 Standard proctor test for soil with treated (5%NaOH) bagasse ash
414 M. Chaudhary et al.

Treated Bagasse Ash with 10% NaOH

Maximum Dry Density

5% Bagasse Ash
10% Bagasse Ash
15% Bagasse Ash
20% Bagasse Ash

10 20 30 40
Optimum Moisture Content (%)

Graph 5 Standard proctor test for soil with treated (10%NaOH) bagasse ash

Graph 6 CBR for parent Parent soil

soil (CBR % = 11.32%) 250
Load (value x 2.77 kg)

50

10 15
Penetration (mm)

CBR for Bagasse with 5% NaOH

300
Load (value x 2.77 kg)

250

200
5% Bagasse Ash
150 10% Bagasse Ash
100 15% Bagasse Ash
50 20% Bagasse Ash

10 15
Penetration (mm)

Graph 7 CBR for soil with treated (5% NaOH) bagasse ash
The Stablization of Soil Using Treated Bagasse Ash 415

CBR for Bagasse at 10% NaOH

90

Load (value x 2.77 kg)

80
70
60 5% Bagasse Ash
50 10% Bagasse Ash
40
15% Bagasse Ash
30
20 20% Bagasse Ash
10

10 12 14
Penetration (mm)

Graph 8 CBR for soil with treated (10% NaOH) bagasse ash

References

Chavan P, Nagakumar MS (2014) Study on soil stabilization by using Bagasse Ash, pp 89–95.
IJSRET ISSN: 2278-0882
Srinivasa Reddy K, Sai Vivek P, Syam Chambrelin K (2017) Stabilization Of expansive soil using
Bagasse Ash. Int J Civil Eng Technol 8(4):1730–1736
Mangi, Ali S, Jamaluddin N, Ibrahim MH, Wan, Abdullah H, Abd AA, Abdul SM, Samiullah S,
Nizakat A (2017) IOP Conf Ser Mater Sci Eng 271(1):012001
Shimola K (2018) A study on soil stabilization using sugarcane Bagasse Ash. Int J Innov Technol
Expl Eng (IJITEE) 8(2S). ISSN: 2278-3075
Dixit P (2019) Gate tutor, pp 389 – 511
https://en.wikipedia.org/wiki/Soil
https://en.wikipedia.org/wiki/Bagasse
https://en.wikipedia.org/wiki/Sodium_hydroxide
Status of Water Quality in Various Ponds
and Lakes in India

Narender Singh and S. K. Patidar

Abstract Water is one of the most important natural resource because existence of
living organisms depends on water. Total amount of water on the earth is approxi-
mately 1.4 billion km3 out of which 97% is saline water and 3% is freshwater. Nowa-
days, water is polluted due to rapid industrialization, urbanization, use of pesticides,
and fertilizers and wastewater disposal. Due to use of polluted water, human beings
suffer from various waterborne diseases, so availability of good quality water is an
essential requirement for prevention of diseases. Almost all lakes and ponds in India
are adversely affected by water pollution. The main adverse effects of water pollution
are death of aquatic organisms, waterborne diseases, eutrophication, bioaccumulation
of contaminants, etc. In the present paper, water pollution, source of water pollution,
and adverse effects of water pollution are discussed, and a summary of studies on
water quality assessment is presented.

Keywords Physico-chemical parameters · Surface water · Water pollution

1 Introduction

Earth is the only planet with about 70% of its surface covered with water. Water
is one of the essential needs for all living organisms for their survival on the earth,
and it is equally important for all. Good quality of water is indispensable for all
organisms. Because of fast-growing population and carelessness of human being, the
quality of water is deteriorating. Most of freshwater bodies are polluted today. Good
quality water is required for drinking, industries, agriculture, and other purposes.
It is a serious problem to provide safe drinking water. The water is non-potable in
many sources because of presence of heavy metal, fluorides, hardness, and arsenic
in excess, and situation is worsened in summer season due to scarcity of water. Due
to use of contaminated drinking water, human beings suffer from various type of

N. Singh (B) · S. K. Patidar

Department of Civil Engineering, National Institute of Technology Kurukshetra, Kurukshetra,
India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 417

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_38
418 N. Singh and S. K. Patidar

waterborne diseases such as cholera, dysentery, and typhoid. According to World

Health Organization (WHO), about 80% of all the diseases to human being are
caused by water. Once the water is polluted, its quality cannot be restored easily.
Due to rapid industrialization, agricultural activity, urbanization, and anthropogenic
activities, water pollution reaches at very high level and affects the aquatic ecosystem.
In India, about 75–80% of water pollution is due to domestic sewage discharge.
Water quality of pond or stagnated water is deteriorated due to eutrophication which
arise due to over supply of nutrient such as nitrogen and phosphorous. Ultimately,
it results in high algal bloom. Extant of environmental pollution and assessment
of water quality are essential to develop management plan for water resources. In
present paper, water pollution, sources of water pollution, and adverse effects of
water pollution are discussed, and a summary of studies on water quality assessment
is presented.

2 Water Pollution

Water pollution is any change in the physical, chemical, and biological proprieties
of water that have detrimental effects on aquatic organisms, plants, and benefi-
cial uses of water. Water pollution occurs when contaminants are directly or indi-
rectly discharged into water bodies without adequate treatment to remove harmful
compounds. Contaminants may include detergents, food-processing waste, insecti-
cides and herbicides, volatile organic compounds, petroleum hydrocarbons heavy
metals, silt, garbage, plastic, etc. Water pollution is the leading cause of deaths and
diseases. In India, polluted water is one of the major factor behind the general low
level of health especially in the rural areas. Polluted water can lead to diseases such
as cholera, dysentery, diarrhea, and jaundice. Around 80% stomach ailment in India
happen due to consuming polluted water. 80% of India’s surface water is polluted.
Based on the latest data from the Ministry of Urban Development, census 2011,
and the Central Pollution Control Board, 80% of India’s surface water is polluted.
The report holds inadequate sanitation facilities, domestic sewerage, and wastew-
ater policy frameworks responsible for this. Major causes of water pollution in India
include urbanization, industries, agriculture, religious, and social practices.

3 Source of Water Pollution

The most common sources of water pollution are industrial waste discharge and
municipal sewage disposal. Source of surface water pollution are classified into two
groups
Status of Water Quality in Various Ponds and Lakes in India 419

1. Point sources
When contamination originates from a single source, it is called point source of
pollution. It includes wastewater discharge by a particular industry, oil refinery,
wastewater treatment plant, chemical and oil spills, and city storm drains. While
point source pollution originates from a specific place, it can affect miles of
waterways and ocean.
2. Non-point sources
Non-point source pollution is contamination derived from diffuse sources. These
may include agricultural or stormwater runoff into waterways from land. This
type of pollution is often the cumulative effect of small amounts of contaminants
gathered from a large area.

4 Major Water Pollutants

The various infectious agents such as bacteria, parasites, and viruses that contaminate
the water are contributed by sewage, human waste, and animal excreta. The chemical
can be either organic or inorganic coming from domestic, industrial, or agricultural
waste. Radioactive waste such as uranium and thorium may contributed from power
plant, mining activities, or natural sources. Nutrients like nitrogen and phosphorous
are contributed from various sources like chemical waste, fertilizers, and manure.
Sediments from soil erosion and heated waters from industries and power plants also
affect water bodies.

5 Adverse Effects of Water Pollution

Water pollution affects humans, animals, and aquatic life. The extent of adverse effect
due to water pollution depends on the kind and concentration of pollutants. Many
water bodies near urban areas are highly polluted. The main adverse effect of water
pollution are as following.
1. Death of Aquatic Organisms
Due to high pollution in lakes, i.e., high organic content, the available DO content
reduces to a very low level, at which many aquatic organisms cannot survive.
2. Bioaccumulation of Contaminants
Pollutants such as lead and cadmium are eaten by tiny animals. Later, these
animals are consumed by fish, and contaminants accumulate higher levels.
3. Waterborne Diseases
In many poor nations, there is always outbreak of cholera and other diseases
as a result of poor drinking water treatment from contaminated sources. The
waterborne diseases put an economic burden on both the households and the
national economy.
420 N. Singh and S. K. Patidar

4. Eutrophication
Nutrient in the water bodies encourage the growth of algae. These algae form a
layer on top of the pond or lake. Main problem due to algae decreases the esthetic
appearance and provides the adsorption site for sun rays which inhibit the aquatic
life in water bodies by reducing photosynthesis.
5. Effect on Recreational Activities
Due to high pollution in lakes, the esthetics of lake gets worse and odor start
coming, due to which tourism gets affected and tourist do not come for recreation
activities. Because of this, local economy and employment are also affected.

6 Studies on Water Quality Assessment

Studies on water quality assessment of surface water sources are presented in Table 1.

Table 1 Summary of studies on water quality assessment of surface sources

Sr. no Detail of water source Detail of study Observations References
1 Powai lake, Mumbai Water samples were DO content ranges Koliyar and
Area of lake is about collected from six from 0.2 to 4.4 mg/L; Rokade [4]
370 acres, and depth is stations. Sample were nitrogen content varied
varied from 10 to 40 ft. analyzed every month from April to
Powai lake is a from April to September
artificial lake located in September 2007 Alkalinity was low
Mumbai Sample were analyzed during monsoon and
for pH, conductivity, high during summer;
dissolved oxygen, pH was higher during
COD, salinity, nitrogen winter and low during
content monsoon
2 Chandlodia lake, Sample were collected Turbidity was observed Verma et al. [13]
Ahmedabad from five different in the range of 16–22
Area of lake is about location and mixed NTU, (maximum
22,940 km2 together to form a during monsoon season
Chandlodia lake is an integrated sample. and minimum during
artificial lake located in Samples were analyzed summer season); pH
Ahmedabad every month from was found in range of
march 2010 to 8.7–9.1 (maximum
February 2011 during monsoon and
Sample were analyzed minimum during
for temperature, pH, summer season);
DO, electrical Alkalinity, DO, and
conductivity, alkalinity, BOD were observed in
total hardness, the range of 192–214,
turbidity, nitrate, 2.14–5.08, and
phosphate, and BOD 1.96–3.38 mg/L,
respectively
(continued)
Status of Water Quality in Various Ponds and Lakes in India 421

Table 1 (continued)
Sr. no Detail of water source Detail of study Observations References
3 Water pond of Orai, Study was carried out pH was in range of Yadav et al. [14]
U.P for a period of one year 8.88–9.4, peak in April
Surface area of pond is (July 2009 to June and least in October;
about 10,000 m2 , and 2010), and sample DO is in range of 5.15
average depth of pond were analyzed every (in June) to 10.5 (in
is about 2.0 m month January); BOD and
Pond is located in orai, Sample were analyzed total alkalinity are in
U.P for pH, electrical range of 1.55–3.26 and
conductivity, DO, 418.25–583.01 mg/L,
BOD, total alkalinity, respectively; nitrate
total hardness, nitrate, and phosphate were
phosphate observed in the range
of 0.69–3.08 and
0.02–0.06 mg/L,
respectively
4 Six different ponds of Samples were collected pH ranged from Mahobe and
Rajnandgaon Town, from five spots in each 7.1–7.9 in summer Mishra [6]
Chhattisgarh pond and mixed 6.8–7.3 in winter; DO
Area of all ponds lies together to prepare a was in range of
between 1 m2 to 2 ha integrated sample. 2.5–4.5 mg/L in
All pond is located in Study was carried out summer and in winter
Rajnandgaon town for 1 year (January to 3.4–5.4 mg/L; turbidity
December 2012) in summer was 2.6–24
Sample were analyzed NTU and in winter
for temperature, color, 1.1–10 NTU; alkalinity
turbidity, pH, total is in range of
alkalinity, electrical 225–360 mg/L in
conductivity, total summer and in winter
hardness, nitrate, 185–341 mg/L
dissolved oxygen, and
BOD
5 Village ponds of Water samples were pH was found in the Shrivastava and
Surguja district, collected and analyzed range of 6.93–7.55; Kanungo [10]
Chhattisgarh every month for a alkalinity, DO, and
period of 10 month BOD were in range of
(August 2009 to May 134.7–205, 2.43–4.4,
2010) and 4.11–7.23 mg/L,
Sample were analyzed respectively
for pH, temperature,
electrical conductivity,
alkalinity, DO, BOD,
total hardness, and
TDS
(continued)
422 N. Singh and S. K. Patidar

Table 1 (continued)
Sr. no Detail of water source Detail of study Observations References
6 Ponds at Varanasi Water samples were pH was observed in the Mishra et al. [7]
collected from eight range of 6.14–8.7;
different ponds during T.S.S, DO, and BOD
the period of were found in the range
2012–2013 of 650–1640, 3.2–5.6,
Samples were analyzed and 1.0–2.5 mg/L,
for pH, TSS, DO, respectively; total
BOD, total alkalinity, alkalinity was found in
nitrate, and phosphate the range of
210–400 mg/L; nitrate
and phosphate varied
from 12.5 to 50.50 and
2 to 7 mg/L,
respectively
7 Surajkund pond at Samples were collected pH was found in range Gupta and
Gorakhpur city from two different sites of 8–8.5; alkalinity, Pandey [2]
in Surajkund pond. DO, and BOD were
And analysis was done observed in range of
in the month of April, 86.2–88.4, 9.4–9.8,
2013 and 6.4–7.2 mg/L,
Sample were analyzed respectively; total
for temperature, pH, nitrate and total
alkalinity, total phosphate varied from
phosphate, total nitrate, 1.08–1.12 and
DO, and BOD 0.06–0.06 mg/L,
respectively
8 Two pond of Periods—3 different pH was in range of Sinha et al. [11]
Samastipur District, season (rainy, winter, 6.5–8.0; DO was found
Bihar summer) 4–5 in rainy, 3.5–3.6 in
These water bodies are The water samples winter, and 3–3.1 mg/L
artificially constructed were collected at an in summer season;
reservoir interval of 30 days BOD was found 27–28
The parameters pH, in rainy, 28–31 in
total alkalinity, total winter, and
suspended solids, and 33–34 mg/L in summer
BOD were analyzed in season; T.S.S was
the laboratory found 380–385 in
rainy, 380–400 in
winter, and
410–415 mg/L in
summer season
9 Loktak lake, Manipur Samples were pH was observed in Laishram and
The Loktak lake is oval collected every month range of 6.05–9.1; DO, Dey [5]
in shape with length from five different sites BOD, and total
and width of 26 km for a period of one year alkalinity was found to
and 13 km, resp., and (March 2012 to be in range of
depth varied from 0.5 February 2013) 4.05–14.18,
to 4.58 m The samples were 1.51–10.65, and
analyzed for water 35–90 mg/L,
temperature, pH, DO, respectively
BOD, total alkalinity,
and conductivity
(continued)
Status of Water Quality in Various Ponds and Lakes in India 423

Table 1 (continued)
Sr. no Detail of water source Detail of study Observations References
10 27 different ponds of Water samples were pH ranged from 6.5 to Dixit [1]
Bilaspur district, collected from twenty 9.69; TDS and DO
Chhattisgarh seven different sites for were in range of
one year (Dec 165.5–254.8 and
2013–Dec 2014) 4.72–6.13 mg/L,
respectively
11 Kshir Sagar water Water samples were Turbidity ranged from Pradeep and
body at Ujjain collected from four 32 to 110 NTU; pH Dwivedi [8]
Area of this pond is selected sites of pond, was found in the range
about 2 ha. And depth and samples were of 8.3–9.2; DO, BOD,
is 3–4 m analyzed in the first and alkalinity were
Kshir Sagar pond week of each month observed in the range
located in Ujjain, M.P for one year (July 2010 of 8.0–15.1, 10.6–23.7,
to June 2011) and 83–99 mg/L,
The sample were respectively
analyzed for turbidity,
pH, DO, BOD, total
alkalinity, total salinity,
and nitrate
12 Three different Ponds Water samples were pH was observed in Tripathi and
at Varanasi collected at interval of range of 6.75–7.5; total Tripathi [12]
fifteen days in each alkalinity, DO, and
pond for a period of BOD were observed in
three month (January range of 143.12–153.9,
to March 2016) 1.54–3.07, and
Water samples were 139–160 mg/L,
analyzed for pH, respectively; turbidity
turbidity, total was in the range of
hardness, total 4.5–40 NTU;
alkalinity, phosphate, phosphate varied from
BOD, and DO 87 to 128.5 mg/L
13 Three urban ponds in Water samples were Average value of DO Saha et al. [9]
Nadia district, W.B collected at interval of in three ponds are 4.03,
First pond have an area fifteen days from three 5.67, and 3.55 mg/L,
of 7000 m2 and ponds. And samples respectively; average
average depth of 5–6 m were analyzed for value of alkalinity in
Second pond have an six months three ponds was
area of 6500 m2 and (January–June 2017) 226.34, 145.93, and
average depth of The samples were 315 mg/L,
6–7 m, and third pond analyzed for respectively; average
have area of 5600 m2 temperature, pH, DO, value of phosphorous
with average depth of free carbon dioxide, in three ponds is 0.23,
3–4 m total alkalinity, BOD, 0.24, and 0.18 mg/L,
All three pond are total phosphorous, and respectively; average
located in Nabadwip at total nitrogen concentration of total
Nadia district, West nitrogen in three ponds
Bengal was 1.28, 1.46, and
1.48 mg/L,
respectively; average
value of BOD in three
ponds was 0.96, 1.13,
and 3.08 mg/L,
respectively
(continued)
424 N. Singh and S. K. Patidar

Table 1 (continued)
Sr. no Detail of water source Detail of study Observations References
14 Kanjia lake Samples were pH was observed Indresha and
Area of lake is 75 ha collected from four between 6.84 and 8.02; Patra [3]
Kanjia lake is located different site at the DO was recorded in
in Bhubaneswar interval of one month the range of
for period of two years 7.82–8.87 mg/L;
(Nov. 2009–Oct. 2011) turbidity was ranges
Sample were analyzed from 3.36 to 6.0 NTU;
for temperature, pH, total alkalinity and
DO, electrical phosphate observed
conductivity, turbidity, from 87.0 to
total alkalinity, and 122.68 mg/L and 2.5 to
phosphate 2.83 mg/L, respectively

7 Conclusions

Water pollution affects humans beings and aquatic life. Polluted water is unsuit-
able for drinking purposes, agriculture, industries, and recreational purposes. It also
affects the esthetic quality of water body. Various studies have indicated increasing
level of water pollution in various lakes and ponds. There is urgent need to take
measures for protecting pollution of lakes and ponds and conserve these resources
for future use.

References

1. Dixit AK (2015) Study of physico-chemical parameters of different pond water of Bilaspur

district, Chhattisgarh, India. Environ Skept Crit 4(3):89–95
2. Gupta SK, Pandey G (2014) Assessment of physico-chemical water quality parameters of
Surajkund pond in Gorakhpur city. Int J Eng Sci Res Technol 3(1):23–25
3. Indresha GN, Patra AK (2014) Seasonal variations in the physico-chemical parameters of
Kanjia Lake. Life Sci Leafl 47:55–64
4. Koliyar JG, Rokade NS (2008) Water quality in Powai lake: Mumbai, Maharashtra. In: Proceed-
ings of Taal2007: the 12th world lake conference, vol 1655, Jaipur (India), 28 Oct–02 Nov 2007,
p 1659
5. Laishram J, Dey M (2014) Water quality status of Loktak lake, Manipur, Northeast India and
need for conservation measures: a study on five selected villages. Int J Sci Res Publ 4(6):1–6
6. Mahobe H, Mishra P (2013) Study of physico-chemical characteristics of water ponds of
Rajnandgaon town, Chhattisgarh. Int J Sci Eng Res 4(8):738–748
7. Mishra S, Singh AL, Tiwary D (2014) Studies of physico-chemical status of the ponds at
Varanasi Holy City under anthropogenic influences. Int J Environ Res Dev 4(3):261–268
8. Pradeep S, Dwivedi HS (2016) Water quality assessment of Kshir Sagar water body at Ujjain
(MP) India. Int J Adv Res Biol Sci 3(8):28–35
9. Saha S, Mandal A, Sahoo D (2017) Study of physico-chemical parameters of three different
urban pond water of Nadia district, West-Bengal, India. Int J Fish Aquat Stud 5(6):23–27
10. Shrivastava S, Kanungo VK (2013) Physico-chemical analysis of pond water of Surguja district,
Chhattishgarh, India. Int J Herb Med 1(4):35–43
Status of Water Quality in Various Ponds and Lakes in India 425

11. Sinha A, Kumar B, Singh T (2014) Water quality assessment of two ponds of Samastipur
district (India). Int J Environ Sci 4(4):567
12. Tripathi R, Tripathi SK (2016) Impact of anthropogenic activities on water quality of prominent
water bodies of Varanasi, UP, India. ESSENCE-Int J Environ Rehabil Conserv 7(2):53–60
13. Verma PU, Purohit AR, Patel NJ (2012) Pollution status of Chandlodia lake located in
Ahmedabad Gujarat. Int J Eng Res Appl 2(4):1600–1606
14. Yadav P, Yadav VK, Yadav AK, Khare PK (2013) Physico-chemical characteristics of a fresh
water pond of Orai, UP, Central India. Octa J Biosci 1(2):177–184
Water Quality Assessment of Brahma
Sarover, Sannihit Sarover and Saraswati
Tirth

Narender Singh and S. K. Patidar

Abstract Water is a vital naturally existing resource because existence of living

organisms depends on water. Total amount of water on the earth is approximately
1.4 billion km3 out of which saline water is 97% and freshwater is 3%. These
days, water is polluted due to rapid industrialization, urbanization, use of pesticides
and fertilizers and wastewater disposal. Usage of polluted water cause waterborne
diseases in humans, so availability of good quality water is an essential requirement
for the prevention of diseases. Almost all lakes and Sarovers in India are adversely
affected by water pollution. The main adverse effects of water pollution are the
death of aquatic organisms, waterborne diseases, eutrophication, bioaccumulation
of contaminants, etc. The issue of water pollution is a serious cause of concern in
the whole world. In India also, it is becoming a serious problem. The main objec-
tive of the present study was to assess status of water quality in three Sarovers,
namely Brahma Sarover and Sannihit Sarover in Kurukshetra and Saraswati Tirth
in Pehowa. Water samples were collected from three Sarovers and analysed for pH,
total dissolved solids, turbidity DO, BOD and total suspended solids during 15 March
2019–30 June 2019. The results indicated that values of pH, total dissolved solids,
turbidity, DO, BOD, total suspended solids were 7.14–9.41, 100–290 mg/l, 1.09–
4.92 NTU, 2.9–5.8 mg/l, 2.0–4.0 mg/l, 40–150 mg/l, respectively, for East Brahma
Sarover and 7.24–9.89, 90–260 mg/l, 1.85–7.5 NTU, 3.0–5.2 mg/l, 2.0–4.0 mg/l, 30–
90 mg/l, respectively, for West Brahma Sarover. The resulted value of pH, DO, BOD,
turbidity, total suspended solids, total dissolved solids were 6.62–9.41, 2.5–4.0 mg/l,
2.0–6.0 mg/l, 2.21–6.61 NTU, 40–140 mg/l, 80–260 mg/l, respectively, for Sannihit
Sarover and 6.8–9.7, 2.1–3.35 mg/l, 3.0–8.0 mg/l, 1.33–8.94 NTU, 50–180 mg/l,
210–530 mg/l, respectively, for Saraswati Tirth.

Keywords Physico-chemical parameters · Surface water · Water pollution

N. Singh (B) · S. K. Patidar

Department of Civil Engineering, National Institute of Technology Kurukshetra, Kurukshetra,
India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2021 427

L. Dewan et al. (eds.), Advances in Renewable Energy and Sustainable
Environment, Lecture Notes in Electrical Engineering 667,
https://doi.org/10.1007/978-981-15-5313-4_39
428 N. Singh and S. K. Patidar

1 Introduction

A Sarover is a body of standing water, either natural or artificial. In the present

study, I have three artificial Sarovers, namely Brahma Sarover, Sannihit Sarover and
Saraswati Tirth. Brahma Sarover is divided into two parts, i.e. East Brahma Sarover
and West Brahma Sarover. All three water Sarovers are prominent religious water
bodies of India. Brahma Sarover is an ancient Sarover with historical and religious
importance, and it is also one of the largest manmade bathing tanks in Asia. The size
of the large tank is about 3600 × 1500 ft. with separate bathing space for men and
women along with periphery of Sarover. As per religious beliefs of people, a single
dip in the Sarover is equal to performing ashvamedha yajya. Water of Brahma Sarover
is polluted due to anthropogenic activities and mass bathing during International Gita
Mahotsav, Amavasya (new moon day), and on solar eclipse. Sannihit Sarover, a sacred
water Sarover in the holy city of Kurukshetra, is believed to be the meeting point of
seven sacred sarasvatis. Bathing in the sacred waters of this Sarover is believed to offer
peace to wandering and unhappy souls. Prayers and pinddaan, a memorial service for
the dead, are performed here. Water quality is also deteriorated due to mass bathing on
solar eclipse and on the day of Amavasya. Saraswati Tirth is a sacred water Sarover in
Pehowa City in Kurukshetra District. The Tirth is considered particularly auspicious
for offering prayers and pinddaan to the souls of the dear departed forefathers or
other family members. The Tirth, therefore, remains crowded with people who come
here to perform Shraddha and offer pinddaan every year.
The main objective of the present study was to assess the water quality status
of three Sarovers, viz. Brahma Sarover, Sannihit Sarover and Saraswati Tirth in
Kurukshetra District, Haryana, India.

2 Materials and Methods

2.1 Materials

2.1.1 Detail of Sarovers

1. Brahma Sarover: Brahma Sarover is located in Kurukshetra on Pehowa Road

behind Birla mandir (Fig. 1). It is one of the largest man-made water tanks in
Asia. The size of the largest tank is about 3600 × 1500 ft., and depth of Sarover
is about 15 ft. However, for the water level on bathing platform is just 4 ft.
Thereafter RCC railing has been provided. Twenty numbers of ladies bathing
ghat have been provided.

2. Sannihit Sarover: Believed to be the permanent abode of Lord Vishnu, Sannihitu

Sarover is situated on Pehowa Road, Kurukshetra. The length of Sannihit Sarover
Water Quality Assessment of Brahma Sarover, Sannihit Sarover … 429

Fig. 1 Brahma Sarover

is about 1500 ft. and breadth 500 ft. The Sarover is linked with the Brahma Sarover
for supply of water. The Sannihit Sarover is shown in Fig. 2.
3. Saraswati Tirth: Saraswati Tirth is situated in Pehowa Town about 30 km from
Kurukshetra. There is an ancient temple dedicated to goddess/river Saraswati.
The Saraswati Tirth is shown in Fig. 3.

Fig. 2 Sannihit Sarover

Fig. 3 Saraswati Tirth

430 N. Singh and S. K. Patidar

2.2 Methods

2.2.1 Schedule of Study

The present study involved water quality assessment of 3 Sarovers. Samples were
collected and analysed during 15 March 2019–30 June 2019 for a period of
3.5 months. Samples were collected and analysed at an interval of 15 days. The
samples were collected from each corners of three Sarovers and then transported
immediately to Environmental Engineering Laboratory, National Institute of Tech-
nology, Kurukshetra, for determination of water quality parameters. Samples were
collected in 4 plastic bottles of 2 L capacity from four corners.

2.2.2 Sampling and Analysis

Water samples were collected after an interval of fifteen days from each Sarover for
3.5 months and analysed for pH, turbidity, DO, total suspended solids, BOD and
total dissolved solids at Environmental Engineering Laboratory, NIT Kurukshetra.
The physiochemical parameters were determined as per the standard method [1].

2.3 Analytical Techniques

pH was measured using ORION 5 star analyser. Turbidity was measured using
ORION AQ4500 turbidimeter. Dissolved oxygen concentration was measured by
Winkler (azide modification) method. BOD, TSS and TDS were measured as per
standard methods [1].
The details of instruments and equipment used in the present study are given in
Table 1.

Table 1 List of instruments

Instrument/equipment Make and model
and equipments used
Turbidimeter ORION AQ4500
pH metre Thermo Scientific, Orion 5 Star
Weighing balance Adair, Dutt, ADGR 200
BOD incubator Microsil
Hot air oven –
Refrigerator –
Water Quality Assessment of Brahma Sarover, Sannihit Sarover … 431

Fig. 4 Variation in pH of Sarover water

3 Results and Discussion

3.1 pH

The pH has an impact on almost all the biological processes and biochemical reactions
in water body [2]. According to CPCB, the permissible limit of pH of water is 6.5–8.5
for Classes A, B, C, D and E. Figure 4 displays the variation in pH of the water of
Sarover. The pH values of East Brahma Sarover lied between 7.14 and 9.41. The pH
values of West Brahma Sarover were found to vary between 7.24 and 9.89. The pH
value of the water of Sannihit Sarover varied between 6.62 and 9.41. The pH value
of Saraswati Tirth was observed to be between 6.8 and 9.7.

3.2 Turbidity

Turbidity is defined as measure of extent to which light is either absorbed or scattered

by suspended and colloidal impurities in water. Turbidity may be caused by both
organic and inorganic particles, such as decomposed plant, algae, silt and clay. The
presence of turbidity affects the aesthetic quality of water. Turbidity interferes with
the penetration of sunlight into water for green aquatic plants and hence interferes
with the photosynthetic reaction. Figure 5 represents the variation in turbidity of the
water of Sarover. The value of turbidity for the East Brahma Sarover ranged between
1.09 and 4.92 NTU. The turbidity of West Brahma Sarover ranged from 1.85 to 7.5
NTU. The turbidity of Sannihit Sarover ranged between 2.21 and 6.61 NTU. The
turbidity of Saraswati Tirth was observed to be between 1.33 and 8.94 NTU.
432 N. Singh and S. K. Patidar

Fig. 5 Variation in turbidity of Sarover water

3.3 Dissolved Oxygen (DO)

DO is generally consumed in various chemical, biological reactions in surface or

groundwater. DO is important for aquatic life. Adequate dissolved oxygen (5–6 mg/L)
is essential for potable water [4]. The dissolved oxygen (DO) is necessary for aerobic
bacteria to break down organic pollutants. The presence of DO is important to take
care of the higher forms of biological life and to maintain proper balance in a healthy
water body. The chemical and biochemical changes undergoing in water body are
mostly dependent upon the presence of oxygen. According to CPCB, the DO in
water should be more than 4 for the propagation of wildlife and fisheries. Banerjea
[3] had reported that if the concentration of DO is about 5 mg/l, throughout the
year, the water body will be productive for fish culture. DO level fluctuates due to
photosynthesis, respiration, aeration and diffusion. Excessive algae growth in surface
water is responsible for supersaturation in oxygen. The DO plays a crucial role in
the existence of aquatic life [6]. Figure 6 represents the variation in DO of the water
of Sarover. The DO of East Brahma Sarover was observed to be between 2.9 and
5.8 mg/l. The DO of West Brahma Sarover varied between 3.0 and 5.2 mg/l. The DO
of Sannihit Sarover varied between 2.5 and 4.0 mg/l. The DO of Saraswati Tirth was
observed to vary between 2.1 and 3.35 mg/l.

3.4 Total Suspended Solids (TSS)

Total suspended solids come from the inorganic particles like silt, clay and organic
particles. The effect of the presence of TSS is increased turbidity in water due to
the presence of silt and organic matters [7]. Figure 7 represents the variation in the
Water Quality Assessment of Brahma Sarover, Sannihit Sarover … 433

Fig. 6 Variation in DO of Sarover water

Fig. 7 Variation in TSS of Sarover water

values of TSS of Sarover water. The TSS of East Brahma Sarover varied between
40 and 150 mg/l. The TSS of West Brahma Sarover varied between 30 and 90 mg/l.
The TSS of Sannihit Sarover varied between 40 and 140 mg/l. The TSS of Saraswati
Tirth varied between 50 and 180 mg/l.
434 N. Singh and S. K. Patidar

Fig. 8 Variation in TDS of Sarover water

3.5 Total Dissolved Solids (TDS)

The total dissolved solids (TDS) present in water comprise of chlorides, biocarbonate,
phosphates, carbonate and nitrates of potassium, calcium, organic matter, sodium,
salt, magnesium [7]. The sewage waste, soap and detergent released due to various
activities of humans contribute to TDS. Figure 8 shows the variation in TDS of
Sarover water. The TDS of East Brahma Sarover was observed to be varying between
100 and 290 mg/l. The TDS of West Brahma Sarover varied between 90 and 260
mg/l. The TDS of Sannihit Sarover varied between 80 and 260 mg/l. The TDS of
Saraswati Tirth varied between 210 and 530 mg/l.

3.6 Biochemical Oxygen Demand (BOD)

BOD represents the amount of oxygen required by microorganisms for decomposi-

tion of biodegradable organic matter present in the water sample at certain tempera-
ture in a specific period. BOD depends on temperature, extent of biological activities
and concentration of organic matter and microbial population such as bacteria and
fungi [8]. According to CPCB, the BOD of water should be less than 2 mg/l for
Class A and less than 3 mg/l for Classes B and C. Figure 9 shows the variation in
BOD of water. The BOD of East Brahma Sarover varied between 2 and 4 mg/l. The
BOD of West Brahma Sarover varied between 2 and 4 mg/l. The BOD of Sannihit
Sarover was observed to be ranging between 2 and 6 mg/l. The BOD of Saraswati
Tirth ranged between 3 and 8 mg/l.
Water Quality Assessment of Brahma Sarover, Sannihit Sarover … 435

Fig. 9 Variation in BOD of Sarover water

Table 2 Summary of overall water quality status

S. No. Parameters Brahma Sarover Sannihit Sarover Saraswati Tirth
1 pH 8.4 ± 0.60 8.20 ± 0.60 7.93 ± 0.53
2 Turbidity 3.62 ± 1.07 4.76 ± 0.61 5.30 ± 0.91
3 DO 4.12 ± 0.76 3.26 ± 0.46 2.75 ± 0.24
4 TSS 75.42 ± 19.71 94.28 ± 20.44 125.35 ± 24.04
5 TDS 177.85 ± 39.38 162.85 ± 44.45 350 ± 67.82
6 BOD 2.76 ± 0.55 3.85 ± 1.06 5.75 ± 1.13

3.7 Overall Water Quality Status

The summary of overall water quality [5] status of Brahma Sarover, Sannihit Sarover
and Saraswati Tirth is given in Table 2.
pH of all three Sarovers was within the range of 6.5–8.5. Saraswati Tirth was
highly turbid as compared to Brahma Sarover. DO of Brahma Sarover was more than
4.0 mg/l, and DO of Sannihit Sarover and Saraswati Tirth was less than 4.0 mg/l. TSS
and TDS of Saraswati Tirth were greater than Brahma Sarover and Sannihit Sarover,
respectively. BOD of Brahma Sarover was less than 3.0 mg/l. BOD of Sannihit
Sarover and Saraswati Tirth was less than 3.0 mg/l.

4 Conclusion

The present study was carried out on assessment of water quality status of Brahma
Sarover, Sannihit Sarover and Saraswati Tirth. Based on the result of present studies,
following finding/conclusion have emerged.
436 N. Singh and S. K. Patidar

• The pH of East Brahma Sarover at three sampling stations was within range of
6.5–8.5, and at one sampling station exceed 8.5. pH of West Brahma Sarover was
within range of permissible limit at all four sampling stations. pH of Sannihit
Sarover was 6.62–9.41 and pH of Saraswati Tirth was 6.8–9.7.
• DO of Brahma Sarover was more than 4.0 mg/l which is suitable for propagation of
wildlife and fisheries. DO of Sannihit Sarover and Saraswati Tirth was found to be
less than 4.0 mg/l, and therefore, they are not suitable for aquatic life propagation.
• Turbidity of east and west Brahma Sarover was 1.09–4.92 NTU, 1.85–7.5 NTU,
respectively. Turbidity of Sannihit Sarover and Saraswati Tirth was varied 2.21–
6.61, 1.33–8.94 mg/l, respectively.
• TSS and TDS of East Brahma Sarover were 40–150 mg/l and 100–290 mg/l,
respectively, and for West Brahma Sarover varies 30–110 mg/l and 90–260 mg/l,
respectively. TSS of Sannihit Sarover and Saraswati Sarover was 40–140 mg/l,
50–180 mg/l, respectively. TDS of Sannihit Sarover and Saraswati Sarover varies
80–260 mg/l, 210–530 mg/l, respectively.
• BOD of Brahma Sarover was less or equal to 3.0 mg/l, and BOD of sannihit
Sarover and Saraswati Tirth was more than 3.0 mg/l.

References

1. APHA (2005) Standard methods for the examination of water and wastewater, 21st edn.
American Public Health Association, Water Environment Federation, American Water Works
Association, Washington
2. Arya S, Kumar V, Raikwar M, Minakshi AD (2011) Physico-chemical analysis of selected
surface water samples of Laxmi Tal (Sarover) in Jhansi City, UP, Bundelkhand Region, Central
India. J Exp Sci 2:01–06
3. Banerjea SM (1967) Water quality and soil condition of fish Sarovers in states of India in relation
to fish production. Indian J Fish 14:115–144
4. Bhardwaj J, Gupta KK, Gupta R (2015) A review of emerging trends on water quality
measurement sensors. In: 2015 international conference on IEEE technologies for sustainable
development (ICTSD), pp 1–6
5. Central Pollution Control Board. www.cpcb.nic.in/Water_Quality_Criteria.php
6. Gupta AK, Kavita M, Pramod K, Chandrasekhar S, Sumita S (2011) Impact of religious activities
on the water characteristics of prominent Sarovers at Varanasi (UP), India. Plant Arch 11(1):297–
300
7. Mahananda MR, Mohanty BP, Behera NR (2010) Physico-chemical analysis of surface and
ground water of Bargarh district, Orissa, India. Int J Res Rev Appl Sci 2(3):284–295
8. Prasanna MB, Ranjan PC (2010) Physico chemical properties of water collected from Dhamra
estuary. Int J Environ Sci 1(3):334–342