Control System

Yared Tadesse - Addis Ababa University - 2021

What is control system?

• System is a set of related or connected things to form entirety.
• System is an arrangement of physical components connected in such manner as to form or act as an
entire unit.
• Control means to direct, regulate or command.

Control system is an arrangement of physical system combined to direct or regulate itself or another system.

Control system is implemented in space shuttle, robotics, industry, aerial or ground vehicle, military machines,
and electronics devices. The main purpose of control system in physical system is to regulate or manipulate
a system dynamic to obtain desired characteristics. The control system is also available by nature such as
balancing, walking, voice regulation of human is a simple example of feedback control systems.

Input of a system is a stimulus or excitation applied by external source to trigger or obtain desired response
from control system. Input of a system is also called setpoint, reference, or desired value.

Output of a system is a response generated by the system in response to applied input.

The purpose of control system is to define desired input-output relation or characteristic of a system. The
control system usually generates desired actuating signal by acting against undesired response. The undesired
response of a system usually caused by noise, disturbance, inherent instability and nonlinear dynamics of a
system to produce desired output. If the input and output of a system is known, the system dynamics can be
estimated.

Control system may be part of larger system and its input-output may be internal variable. In this case the
system under control might be subsystem and a control system might be designed to regulate a subsystem.

Example of control system

• Switch control flow of electric current into a bulb or machine

• Floating valve is used to regulate a water level in a tanker
• Governor is used to regulate flow rate of water hitting a turbine.

Open and closed loop control system

Control systems are categorized into open and closed loop control system. The major difference between
closed and open loop system is that the signal used for triggering the controller to produce desired control
action or effort.

In closed loop control system, output of a system under control is feedback to compute its relative deviation
from setpoint/reference/input. Thus, a closed loop control system always has information about a current
state of a system. An open loop control system produces a control action or effort depending input and lack
information about current state of a system. Thus, unlike closed loop control system, a control action of open
loop system is independent of output of a system.

1
Major feature of open loop control system:

• Accuracy of a system solely depends on certainty of calibration or adjustment.

• Not affected by instability of a system

Examples:

Switch of bulb is a good example of open loop system.

Autopilot system, Segway controller, humanoid robot are examples of a closed loop system.Closed loop system
consists of a feedback signal so called feedback systemf. Feedback is a system characteristic quantitatively
measured and returned back as the input of control system to generate appropriate control action.

Feedback system impart the following properties:

• Increased accuracy
• Tendency to instability
• Reduced sensitivity which is a ratio of change in output and input
• Reduced nonlinearity
• Reduced disturbance
• Increased bandwidth

Schematic diagram of open loop control system

Schematic diagram of closed loop control system

2
Analog, discrete and digital control systems
The input-output signal of control system is a function of independent variable, usually time, t.

A signal dependent on a continuum of value of independent variable t is called a continuous-time signal, more
generally a continuous-data signal or (less frequently) an analog signal.

A signal depends on a discrete value of independent variable, time-t, is called discrete time. A binary
representation of a discrete time value of a signal is called digital signal.

Mathimatical Modeling
Its a method of finding an anlytical function or equation using physical law that reasonabily represents
system characterstics. Once the system model is found, the model will be used to anayze its input output
charactersitics, stability, robustness etc. Most system commonly can be reasonably well described by
differential equation model. The system described by no derivative involved in it is called zero order system; one
derivative, first order; two derivative, second order.

Any control system has input, system and output. Output is a system response to input signal which is known as
excitation signal.

Laplace Transform
Laplace transfrom is used to simplify the system analysis process which is complex in time domain analysis.
In time domain analysis of a system, response to any input signal is obtained by appling convolution operation
which is tedious and time consuming. However the Laplace transform, frequency domain analysis, change
the convolution in time domain to multiplication in time domain. Thus, the system repsonse anlysis is easier in
frequency domain than time domain.

where,

Table of Laplace tansfrom [Ref: Norman Nise-11Ed]

3
Laplace Theorems

Examples

4
Inverse Laplace transform
Inverse Laplace transform is computed using partial fraction expansion of a given transfer function to get
simplified expression and using properties of Laplace Transform.

Zero order system

The input-output characterstics of zero order systems best described by scaling the input adding offset.

Where is input, output, a scale or slope, and b is zero bias or zero crossing point.

Example:

Potentiometer:

The potentiometer can be either linear or rotary type. The linear potentiometers are equiped with slider circuit
and neglecting the dynamics of the friction and motion of slider, its resistance at postion is best
described by equation:

The rotary type potentiometer has a rotating slider.

Find the mathimatical model of a system variable voltage divider circuit shown below. Lets take the maximum
angle or rotation of potentiometer is .

5
The voltage read at center tap of potentiometer in the above circuit can be found as follows.

where p refers to percentatge of rotation of pointer relative to maximum angle of

rotation.

Example-2:

Linear spring with stiffness of

where the opposing force produced by spring is F and displacement is

First order system

Not all system is first order system unless some parameter or characteristics are neglected. For instance in
potentiometer the actual input is force which diplace the center tap but the displacement is considered as input
to simpify anaysis.

Mechanical system
Figure below shows a mass-damper system. The external force is applied to the system. If the mass is

m=0.5kg, viscous friction coefficient , and the system is intially at rest,

a. find a mathimatical model of a system if input is force and output is velocity

6
b. determine an impulse input

b. simulate it using matlab to verify your answer.

Gs =

1
-------------
0.5 s^2 + 3 s

Continuous-time transfer function.

Thermal system
A thermometer is immersed in boiling water to measure a temperature. If the convective heat transfer coefficient
is , and specific heat capacity of the water cof the water and mercury is know. Calculate the mathimatical
model of a system.

Ref: Convection Heat Transfer Coefficient

Unit of convective heat transfer coefficient is .

7
There are two types of convection: natural and forced convection.

Natural convection is produced by density difference in a fluid due to temperature difference. eg. Hot air/liquid
rises

Forced convection is produced when fluid/air mostion is produced by fans or pump. It is more efficient than
natural convection and widely used in many engineering applications.

Type of ConvectionConvective Heat Transfer Coefficient h in W/(m2/K)

• Air, free 5–25

• Air, forced 10–500
• Liquids, forced 100–15,000
• Boiling water 500–25,000
• Condensing water vapor 5,000–100,000

# Time domain analysis

Eg. A Mericury thermometer has the time constant of 0.2 second is immersed into a boiling water from ambient
temperature of 20 C. What is the temperature indicated by thermometer 1 second after immeresed into water
steam, 100C.

Solution:

8
 Tm1s = 100

Initial condition

9
#Timed domain

Elecctrical system
Voltage accross an inductor :

Voltage accross the capacitor:

Voltage accross the resistor:

Compute the mathimatical model of the circuit shown below

gs =

5e-06 s
------------------------
5e-08 s^2 + 0.0005 s + 1

10
 Continuous-time transfer function.

#Time domain analysis

11
Sinusoidal response

Compute the mathimatical model of the above circuit taking the voltage as input and current as the output of the
system.

• Step reponse
• Ramp response
• Sinusoidal response

#Time domain analysis

Electromechanical system modeling

DC Motor Modeling - 1

12
Solution:

gse =

1
-----
s + 7

Continuous-time transfer function.

gsm =

1
-------------
0.001 s + 0.1

Continuous-time transfer function.

gsf =

0.1
-------------------------
0.001 s^2 + 0.107 s + 0.7

Continuous-time transfer function.

gs =

0.1
--------------------------
0.001 s^2 + 0.107 s + 0.71

Continuous-time transfer function.

#Time domain analysis

13
Sinusoidal response of DC Motor

14
DC Motor modeling-2 [Ref: Norman Nise - 11Ed]

Example
Calculate the transfer function model of a non-inverting amplifier given in figure below.

15
Solution:

1. Simplify the circuit by finding the equivalent impedance

2.
Determine the voltage gain as

3.
,

4.

5.

gs =

0.01 s^2 + 1.11 s + 1

---------------------
0.01 s^2 + 1.01 s + 1

Continuous-time transfer function.

16
17
Inverse Laplace Transform

18
Eg. Find the a time domain step and ramp response of the system with transfer function model of .

Hint: , and System response and

Creep and Recovery

19
Exercise

Find the Transfer function model of the system and its response to standard test signals [Norman Nise - 11 Ed]

Solution:

gs =

3 s + 1
-----------------------
s^4 + 7 s^3 + 5 s^2 + s

20
Continuous-time transfer function.

21
Signal generation

22
gs =

3
-----
s + 3

Continuous-time transfer function.

23
Linearization of Non-linear system
Linearization is a process of representing a non-linear system with a best approximate of linear model in small
interval near equilibrium point.

In most approximate representation of nonlinear system, Taylor series is used to get linear model near
equilibrium point.

Non-inverted Pendulum
Find linearized transfer function model of a nonlinear system given in figure below near equilibrium point,

24
Solution:

Considering the uniform bar, the centroid of a bar be at mid point, .

For small angle

The transfer function of the system with input and output ,

Given mass M=0.5kg, length L=0.25m,

gs =

1
----------------
0.05 s^2 + 1.226

Continuous-time transfer function.

25
Characteristics of feedback control system

Feedback system is older than humanity. It is widely available in biological control system built in earliest
inhibitants of our planet.

The feedback control system is technically used in the following applications:

• Greeks used to use feedback system around 300BC

• Ktesibios invented a water clock using feedback float valve to keep water tanker at constant level inorder
to keep constant flow rate to indicate time
• Philon of Byzantium applied liquid level control methods invented by Ktesibios.
• Steam pressure regulation began in 1681 within Denis papins invention. The regulator has builtin sacfety
valve to release steam pressure whenever pressure exceed the maximum level.
• Edmund Lee designed wind mill speed control by manipulating its angle of attack.

Characteristics of negative feedback control systems

Open loop system generates control effort independent of a measured feedback signal or output of a system.
Consequently, its performance solely depends on accuracy of its caliberation. However, closed loop system
compute and generate controlling signal or effort based on the deviation of process outputf from input or
reference signal.

Error analysis of negative feedback system

A negative feedback system can alter the perfromance characteristics of a process. It is usually designed to
reduce error between desired output or input signal and actual process output.

26
Super position principle can be used to analyze error or uncertainity of a system

then corresponding output

, then corresponding output

, hen corresponding output

Gross output of a process or total response of a plant under control is the sum of all the three output computed
above.

Error signal is a difference between the input signal and output signal of a process.

If the system feedback gain is different from unity, then the corresponding error signal become

27
The total gain over the loop is called lopo gain.

Sensitivity of closed loop system

System sensitivity is defined as the ratio of change in the system transfer function to small incremental change
in process transfer function or parameter.

Consider the closed loop system transfer function

Sensnitivity of a closed loop system to variation in a process is defined as

It is also possible to calculate closed loop system sensitivity to a parameter variation in a process,

Time Domain Response Analysis of Control System (MATLAB)

• Performance characteristics of feedback system
• Steady state error, error constant, performance Indices
• Stability Analysis of Linear Systems

Poles, Zeros, and System response

The output of a system or response of a system is the sum of forced response and natural response.These
responses can be found by solving by differential equation model of a system or by using inverse Laplace
transform. However, these methods are time consuming and complex than qualitative techniques that are
based on understanding of polse and zeros of a system. The concept of poles and zeros, fundamental to the
analysis and design of control systems, simplifies the evaluation of a system’s response.

Poles of transfer function

28
The poles of a transfer function are

1. the values of the Laplace transform variable, s, that cause the transfer function to become in finite or
2. any roots of the denominator of the transfer function that are common to roots of the numerator.

The common routes of numerator and denominators might be cancelled ant no more leads the transfer function
to infinity. However, the cancelled poles are considered as the pole in control system eventhough it doesn't
drive a system to infinity.

Zeros of transfer function

The zeros of a transfer function are

1. the values of the Laplace transform variable, s, that cause the transfer function to become zero, or
2. any roots of the numerator of the transfer function that are common to roots of the denominator

A strict definition of zeros of a transfer function is mentioned in 1; however, as long as the cancelled zeros are
considered in control system the second definition is also considered.

Example:

Consider the transfer function

29
• A pole of the input function generates the form of the forced response (that is, the pole at the origin
generated a step function at the output).
• A pole of the transfer function generates the form of the natural response (that is, the pole at -5
generated ).

30
 • A pole on the real axis generates an exponential response of the form , where is the pole location
on the real axis. Thus, the farther to the left a pole is on the negative real axis, the faster the exponential
transient response will decay to zero

Common terminologies

Free response or natural response of differential equation is a solution of differential equation for zero input
and non-zero initial condition.

Force response of a differential equation is a solution of a differential equation when all initial condition are set
identically to zero and non-zero input.

Total response of a differential equation is the sum of both free response and forced response of a system.

Steady state response of a system is a part of total response which doesnt approach to infinity as time goest to
infinity.

Transient response of a system

In control system, a transient response is the response of a system to a change from an equilibrium or a
steady state. The transient response is not necessarily tied to abrupt events but to any event that affects the
equilibrium of the system. The transient response of a system is a part of system response which approaches to
zero as time goes to infinity for stable system.

Singularity functions
Singularity functions are widely used to study system characteristics in control system. Each singularity function
are ether integral or differentiation of another signularity function.The singularity functions are also called
standard test function. They includes: unit impulse, unit step and unit ramp functions.

Unit impulse where

Unit step where

Unit ramp where

First order system

31
A first order system without zeros is described by the following transfer function.

Transfer function of first order system G(s) in general can described by the followong equation.

Unit impulse response (

Transfer function G(s) is the ratio of output to input of a system

, rewriting the equation the following is obtained

Inverse Laplace transform of the above equation obtained as follows

Example:

Let , impulse response of G(s) become

32
Unit step response (u(t))

ans = struct with fields:

RiseTime: 1.0985
SettlingTime: 1.9560
SettlingMin: 0.9045
SettlingMax: 1.0000
Overshoot: 0
Undershoot: 0
Peak: 1.0000
PeakTime: 5.2729

Creep and recovery of any system is a result of a system limit to trace reference signal.

Details of step response of first order system

33
Time constant ( can be described as the time for to decay to 37% of its initial value.

The reciprocal of the time constant has the units (1/seconds), or frequency. Thus, we can call the parameter
a the exponential frequency. Since the derivative of is -a when , a is the initial rate of change of the
exponential at . Thus, the time constant can be considered a transient response specification for a first-
order system, since it is related to the speed at which the system responds to a step input.

The time constant is a reciprocal of poles of a system thus the farther the pole from imaginary axis, the faster
the transient response become.

Rise time, , is a time for a system response go from of its final value. It is obtained by computing a
time difference such that .

Settling time is defined as the time for the response to reach, and stay within, 2% of its final value.

for unit step response

Ramp input response

34
Ramp input

35
Inverted Pendulum Problem

36
Mathimatical model

Assumption: small angle deviation from equilibrium point

Freebody diagram of inveted pendulum

System transfer function

Ts =

From input to output...

4.182e-06 s^2 - 0.0001025
1: ---------------------------------------------------------
2.3e-06 s^4 + 4.182e-07 s^3 - 7.172e-05 s^2 - 1.025e-05 s

1.045e-05 s
2: -----------------------------------------------------
2.3e-06 s^3 + 4.182e-07 s^2 - 7.172e-05 s - 1.025e-05

37
 Continuous-time transfer function.

Select a point in the graphics window

selected_point = -0.9526

Inverted pendulum control

Control system topology of inverted pendulum is different from a common controller as it always tend to set
system output, angle, to zero. Thus the set point of a controller is always set to zero; for this reason, it is
called regulator problem. The input force that cause deviation from the equilibrium position is considered as
disturbance.

The main purpose of the controller in this specific problem is to keep system, pendulum, at equilibrium point
Thus, the system output of our interest is angle of deviation from the equilibrium point and the input
disturbance force. The position of a cart in response to disturbance force exerted on inverted pendulum is plot in
figure below.
Warning: Error occurred while executing the listener callback for event POST_REGION defined for class
matlab.internal.language.RegionEvaluator:
Dot indexing is not supported for variables of this type.

Error in alternateGetframe

Error in getframe (line 138)

x = alternateGetframe(parentFig, offsetRect, scaledOffsetRect, includeDecorations);

Error in matlab.internal.editor.FigureManager

Error in matlab.internal.editor.FigureManager.getCurrentFigurePixels

Error in matlab.internal.editor.FigureManager

Error in matlab.internal.editor.FigureManager

Error in matlab.internal.editor.FigureManager.saveSnapshot

Error in matlab.internal.editor.FigureManager.snapshotAllFigures

Error in matlab.internal.editor.OutputsManager/postRegionCallback

Error in matlab.internal.editor.OutputsManager>@(varargin)obj.postRegionCallback(varargin{:})

38
 Error in LiveEditorEvaluationHelperE421203478 (line 307)
kp = 34.8;

Error in matlab.internal.editor.evaluateRegions

Error in matlab.internal.editor.EvaluationOutputsService.evalRegions

The movement of the cart as the pendulum oscillate is uncontrolled as it is not our interest. However, it is
possible to see the movement of a cart to set a pendulum to equilibrium position.

Control System Designer

39
Root locus design example
gs =

s
-----
s + 3

Continuous-time transfer function.

grs =

4
-------------
s^2 + 2 s + 4

Continuous-time transfer function.

40