Вы находитесь на странице: 1из 235

HCMC UNIVERSITY OF TECHNOLOGY AND EDUCATION

FACULTY OF ELECTRICAL AND ELECTRONIC ENGINEERING


Automatic Control Department

Hệ thống Điều khiển tự động


(Automatic control systems)

PGS.TS. Trương Đình Nhơn


Bộ môn Tự động Điều khiển
Email: nhontd@hcmute.edu.vn
0903 675 119

1
Content
Introduction to control systems
Mathematical models of systems
Stability of the control systems
Steady state errors of the control systems
Frequency domain of the control systems
Bode plot of the control systems
The Root locus method
Frequency response method
Design PID controller
Design Lead/Lag controller
Design of feed back control systems

2
References
[1] “Lý thuyết Điều khiển tự động”, Nguyễn Thị Phương Hà, Huỳnh
Thái Hoàng, NXB ĐHQG TPHCM.

[2] “Bài tập Điều khiển tự động”, Nguyễn Thị Phương Hà, Huỳnh
Thái Hoàng, NXB ĐHQG TPHCM.

[3] “Automatic Control Systems,” B. C. Kuo.

[4] “Control Systems Engineering,” N. S. Nise.

3
Criteria
1. Process examination (50%)
- Homework via online submission
- Paper test in class

2. Final examination (50%)


- Paper test with open book

4
Introduction
System – An interconnection of elements and devices for a desired
purpose.

Control System – An interconnection of components forming a


system configuration that will provide a desired response.

Process – The device, plant, or system under control. The input and
output relationship represents the cause-and-effect relationship of the
process.

5
Introduction
Open-Loop Control Systems
utilize a controller or control
actuator to obtain the desired
response.

Closed-Loop Control
Systems utilizes feedback
to compare the actual output
to the desired output
response.

Multivariable Control System

6
History
The first automatic feedback controller used in an industrial process is
generally agreed to be James Watt’s Flyball governor, developed in
1769 for controlling the speed of a steam engine.

7
History
James Watt’s

8
History

The first historical feedback system,


claimed by Russia, is the water-level
float regulator said to have been
invented by I. Polzunov in 1765.

The float detects the water level and


controls the valve that covers the
water inlet in the boiler.

9
History

10
History
18th Century James Watt ’ s centrifugal
governor (bộ điều tốc ly tâm) for the speed
control of a steam engine.
1920s Minorsky worked on automatic
controllers for steering ships (hệ thống lái tàu).

11
History
1930s Nyquist developed a method for analyzing the stability of
controlled systems

1940s Frequency response methods made it possible to design linear


closed-loop control systems

1950s Root-locus method due to Evans was fully developed

1960s State space methods, optimal control, adaptive control and

1980s Learning controls are begun to investigated and developed.

Present and on-going research fields. Recent application of modern


control theory includes such non-engineering systems such as
biological, biomedical, economic and socio-economic systems

12
Examples of Control Systems

13
Examples of Modern Control Systems

14
Examples of Modern Control Systems

A three-axis control system for


inspecting individual semiconductor
wafers with a highly sensitive camera.

15
Examples of Modern Control Systems

Coordinated control system for a boiler–generator.

16
Examples of Modern Control Systems

The Utah/MIT Dextrous Robotic Hand: A dextrous robotic hand having


18 degrees of freedom. It is controlled by five Motorola 68000
microprocessors and actuated by 36 high-performance
electropneumatic actuators via high-strength polymeric tendons.
The hand has three fingers and a thumb. It uses touch sensors and
tendons for control.

17
Examples of Modern Control Systems

The ongoing area of research


and development of
unmanned aerial vehicles
(UAVs) is full of potential for
the application of control
The Honda P3 humanoid robot.
systems.
P3 walks, climbs stairs, and turns
corners.

18
Examples of Modern Control Systems

A feedback control system model of the national income.

19
The future evolution of control systems

Future evolution of control systems and robotics.

20
Control system design

21
Design example: turntable speed control

22
Design example: turntable speed control

23
Design example: Electric ship concept

Vision
Electrically
Reconfigurable
All
Integrated Ship
Electric
Power Ship
System  Technology
Insertion Increasing Affordability and Military Capability
 Reduced manning
 Electric Drive  Warfighting
 Automation
 Reduce # of Prime Capabilities
Movers  Eliminate auxiliary
systems (steam,
 Fuel savings hydraulics, compressed Main Power
 Reduced maintenance air) Distribution
Propulsion Motor Prime
Generator
Motor Drive Mover

Power Ship
Conversion Service
Module Power

24
Control systems are found in abundance in all sectors of industry
- Quality control of manufactured product
- Automatic assembly line
- Machine-tool control
- Space technology and weapon systems
- Computer control
- Transportation systems
- Power systems
- Robotic
- And many others

25
Control systems are found in abundance in all sectors of industry

26
Control systems are found in abundance in all sectors of industry

27
What type of control systems ?

The basic ingredients of the system can be described by:

Objectives Results
Control
system

In more scientific terms, these three basic ingredients can be


identified with Inputs, System components and Outputs.

Inputs (u) Output (c)


Control
system

28
Example 1:

There are two basic forms of control system:

 Open-loop control system

 Closed-loop control system

29
Open-loop control systems (Nonfeedback systems)

Reference Actuating Control

Input (r) Signal (u) Controlled Variable (c)


Controller
Process

Elements of an Open-loop control systems.

Input signal or command (r) applied to the controller, whose


output acts as the actuating signal (u); the actuating signal then
controls the controlled process so that the controlled variable (c)
will perform according to some prescribed standards.

30
Open-loop control systems (Nonfeedback systems)

Reference Actuating Control

Input (r) Signal (u) Controlled Variable (c)


Controller
Process

Elements of an Open-loop control systems.

In simple cases, the controller can be an amplifier, mechanical


linkages, or other control means, depending on the nature of the
system. In more sophisticated electronic control, the controller
can be an electronic computer, such as a micro-processor.

31
Example 2:

Ceiling fan

Gearbox Speed
Fan
Five-speed gearbox

32
Closed-loop control systems (Feedback control systems)
To obtain more accurate control, the controlled signal c(t) should be
feedback and compared with the reference input, and an actuating signal
proportional to the difference of the input and the output must be sent
through the systems to correct the error. A system with one or more
feedback paths such as that just discribed is called a closed-loop system
Actuating Control
Error detector
Reference Signal (u) Controlled Variable (c)
Controller
Input (r) Process

Feedback
Elements of an Closed-loop control systems.

33
Example 2:

Air Conditioner

Set Temperature Temperature


Air Conditioner

Measuring temperature device

34
Mathematical models for systems
A mathematical model of a system is a “replica” (mô hình, bản
sao) of the relationships between the inputs and the outputs of
the system.
Example: A motor as a system.
The input is voltage (V)
The output is the speed of the rotation of the shaft ( )

If this is a linear system, then the mathematical model is:

Where G is the constant of proportionality

35
Mathematical models for systems
Thus, we have:

Hence: steady  state value of 


G
steady  state value of V

The constant G is called the Transfer Function

In general, we can define the Transfer Function as the ratio of the


steady-state output to steady-state input for a system or subsystem.

steady  state output


Transfer function G 
steady  state input

36
Mathematical models for open-loop systems
Consider three elements in series:

Input (r) u1 u2 Output (c)


G1 G2 G3

Transfer function G  G1.G2.G3

37
Mathematical models for closed-loop systems
Consider a simple closed-loop system.
r + e c
X G
- f
H

c G
Transfer function : 
r 1  GH

38
System models
Mathematical models are needed of the elements used in
such systems. These are equations representing the
relationship between the input and output of a system.

 Electrical system building blocks:

Inductor: vL
di 1
 i   vdt E
1 2
Li
dt L 2
Where L is the inductance
Capacitor: 1 dv 1 2
C
v idt  i  C E Cv
dt 2

Where C is the capacitance

39
System models

Resistor: v  Ri 1 2
P  v  Ri 2
R
Where R is the resistance
RLC in series circuit:
v = v R + v L + vC

ò i dt
di 1
= i.R + L +
dt C
Where R,L,C are the resistance, inductance and
capacitance, respectively

40
System models
RLC in parallel circuit:
i  iR  iL  iC
v 1 dv
   vdt  C
R L dt
Where R,L,C are the Resistance, Inductance and
Capacitance, respectively

41
System models
Ex 1: RC system:
v  vR  vC
 i.R  vC
dvC
 RC  vC
dt
Ex 2: RL system:
v  i.R  vL
R
  vL dt  vL
L

42
System models

di
v  i.R  L  vC
dt
dvC d 2 vC
 RC  LC 2  vC
dt dt

43
Transfer function
An importance first step in the analysis and design a control
system is the mathematical description and modeling of the
process that is to be control.
The input-output relation of a linear time-invariant system is
described by the following n th-order differential equation
with constant real coefficients:

d n c(t ) d n 1c(t ) d m r (t ) d m1r (t )


n
 an n 1
 ....  a1c(t )  bm1 m
 bm m 1
 ....b1r(t )
dt dt dt dt
c(t): Output response
r(t): Input
ai ( i = 1,…,n) and bJ (j = 1,...,m+1) are real constants (n ³ m)

44
Transfer function

Take the Laplace transform on both sides of the equation,


and assume zero initial conditions. The result is:
(sn + ansn-1 + … + a1)C(s) = (bm+1sm + bmsm-1 + … + b1)R(s)

The transfer function between r(t) and c(t) is given by:

C ( s) bm1sm  bmsm-1    b1
G( s)  
R( s ) sn  a nsn-1    a1

45
System models

di
v  i.R  L  vC
dt
dvC d 2 vC
 RC  LC 2  vC
dt dt

1
G(s) = 2
s LC + sRC +1

46
Block Diagram

Block Diagram is often used by control engineer to portray


systems of all types. A block diagram can be used simply to
represent the composition and interconnection of a system.

Summing Block
Takeoff
point point
+ Transfer
Function
+ or -

47
Block Diagram
R(s) E(s)
+ G(s)
- C(s)
B(s)
H(s)

Basic block diagram of a feedback control systems

()
G s =
C(s)
E(s)
: Open - loop transfer function Or forward - path transfer function

()
M s =
C(s)
=
G(s)
R(s) 1+ G(s).H(s)
:Closed - loop transfer function

H (s) :Feedback path transfer function


G (s) H (s) :Loop transfer function

48
Multi-element systems
In Series

In parallel

49
Equivalent systems

R1
R1 C G
1) G = C
() R2 ()
R2 G

2)

50
Example
Example2:
2

What is the transfer function of the system?

51
Example 2

C G2
Thus  (1  G1 )
R 1  G2 H 1

52
Equivalent systems

3)

4)

53
Example 1

What is the output (C) of the system?

54
Example 1

The output is:

G1G2 R  G2 N
C  WR R  W N N 
1  G2 H 2  G1G2 H 1

55
Signal flow graphs

A signal flow graph may be defined as a graphical means of


portraying the input-output relationships between the
variables of a set of linear algebraic equations.

When constructing a signal flow graph, junction points or


nodes are used to represent the variables, the nodes are
connected together by line segments called branches. The
branches have associated

56
Signal flow graphs
Consider a simple closed-loop system.
r + e c
X G
- f
H
G s  
C( s)
: Open  loop transfer function or forward  path transfer function
E( s)

 
M s 
C ( s)

G( s)
: Closed  loop transfer function
R( s ) 1  G( s).H ( s)
H s  : Feedback path transfer function
Gs H s  : Loop transfer function

57
Mason’s gain formula for Signal flow graph
Mason’s gain formula is given by:
C 1 n
Where: G    M K K
R  K 1
M k : path gain or transmitta nce of Kth forward path

  1   La   Lb Lc  L L L d e f  ...
a b ,c d ,e , f

D = 1- (sum of all individual loop gains) + (sum of gain productsof all


possible combinations of two nontouching loops) - (sum of gain
productsof all possible combinations of three nontouching loops) +...
The factor  K is obtained from  by removing the loops that touch path M K
Samuel Jefferson Mason (1921–1974)
58
Example 2

59
Example 1
Obtain the closed-loop transfer function by using the
Mason’s gain formula.

Ei y1 z2 y3 z4 1 E0

-y1 -z2 -y3

E0 1 y1 z 2 y 3 z 4
M   M 11 
Ei  1  y1 z 2  y 3 z 2  y 3 z 4  y1 z 2 y 3 z 4

60
Example 2
Obtain the closed-loop transfer function by using the
Mason’s gain formula.

61
Example 2

62
Basics of State Space Modeling

63
Basics of State Space Modeling

The differential equations of a lumped linear network can be written in


the form:

the state equation of the system


the output equation

The vector x(t) is the state vector, and u(t) is the input vector. A is
called the state matrix, B the input matrix, C the output
matrix, and D is the direct transition matrix.

64
Basics of State Space Modeling

d n c(t ) d n 1c(t ) d m r (t ) d m1r (t )


n
 an n 1
 ....  a1c(t )  bm1 m
 bm m 1
 ....b1r(t )
dt dt dt dt

65
Basics of State Space Modeling
consider an 1st order linear plant model described by the differential
equation

Where y(t) is the plant output and u(t) is the plant input

A useful set of state variables, referred to as phase variables, is


defined as:

66
Basics of State Space Modeling

In matrix form

67
68
Basics of State Space Modeling

Obtain the state equation of RLC circuit:

di
v  i.R  L  vC
dt
dvC d 2 vC
 RC  LC 2  vC
dt dt

69
Basics of State Space Modeling

Obtain the state equation in phase variable form for the


following differential equation:

70
Basics of State Space Modeling

Write the state equation for the circuit

71
Basics of State Space Modeling

In general, if there are nC loops of all capacitors and


voltages sources, and nL cut sets of all inductors and
current sources, the number of state variables is

where eC and eC are the numbers of inductors and


capacitors, respectively.

72
Basics of State Space Modeling

73
Basics of State Space Modeling

74
Transfer Function to State Space Conversion

Consider the transfer function of a third-order system

It is decomposed into two (frequency domain) blocks

75
Transfer Function to State Space Conversion

First block

Second block

76
Transfer Function to State Space Conversion

Assigned the state variables as

77
State Space to Transfer Function Conversion

X2 X1
+
R
3 s2 C

- s4 s5

X3
s 1
s6

78
State Space to Transfer Function Conversion
Take Laplace transform and rearrange as follows:

79
Transfer Function

 When order of the denominator polynomial is greater


than the numerator polynomial the transfer function is
said to be ‘proper’.

 Otherwise ‘improper’

80
Transfer Function

Transfer function helps us to check

The stability of the system

Time domain and frequency domain characteristics of

the system

Response of the system for any given input

81
Stability of Control System

There are several meanings of stability, in general


there are two kinds of stability definitions in control
system study.

Absolute Stability

Relative Stability

82
Stability of Control System

Roots of denominator polynomial of a transfer function are


called ‘poles’.

And the roots of numerator polynomials of a transfer


function are called ‘zeros’.

83
Stability of Control System

Poles of the system are represented by ‘x’ and zeros of the


system are represented by ‘o’.
System order is always equal to number of poles of the
transfer function.
Following transfer function represents nth order plant.

84
Stability of Control System
Poles is also defined as “it is the frequency at
which system becomes infinite”. Hence the name
pole where field is infinite.

And zero is the frequency at which system


becomes 0.

85
Stability of Control System
Poles is also defined as “it is the frequency at which
system becomes infinite”.
Like a magnetic pole or black hole.

86
Relation b/w poles and zeros and frequency response of
the system

The relationship between poles and zeros and the frequency


response of a system comes alive with this 3D pole-zero plot.

Single pole system

87 87
Relation b/w poles and zeros and frequency response of
the system

3D pole-zero plot


System has 1 ‘zero’ and 2 ‘poles’.

88
Relation b/w poles and zeros and frequency response of
the system

89
Example

Consider the Transfer function calculated in


previous slides.
X (s) C
G( s )  
Y ( s ) As  B

the denominato
r polynomialis As  B  0

The only pole of the system is

90
Example
Consider the following transfer functions.
Determine
– Whether the transfer function is proper or improper
– Poles of the system
– zeros of the system
– Order of the system

s3 G( s ) 
s
i) G( s )  ii)
s( s  2) ( s  1)( s  2)( s  3)

( s  3)2 s 2 ( s  1)
iii) G( s )  iv) G( s ) 
s( s 2  10) s( s  10)

91
Finding Zeros and Poles with Matlab

The command: [z,p,k] = tf2zp(num,den) to obtain the


Zeros, Poles, and gain K of transfer function

For example:
C ( s) 4s2  16s  12
 4
R( s ) s  12s3  44s2  48s

92
Finding Zeros and Poles with Matlab

93
The following Matlab program will yield the original num/den

94
Partial-fraction expansion with Matlab

Consider the following function:

C ( s ) num bm1sm  bmsm-1    b1


 
R( s ) den sn  a nsn-1    a1
Where some of ai and bj may be zero
In Matlab row vectors num and den specify the coefficients of the
numerator and denominator of the transfer function. That is:
num  [bm1 bm  b1 ]
den  [1 aa a n-1  a1 ]

95
Partial-fraction expansion with Matlab

The command: [r,p,k]=residue(num,den)

Finds the residues (r), poles (p), and direct terms (k) of partial-
fraction expansion of the ratio of two polynomials C(s)/R(s).

The partial-fraction expansion of C(s)/R(s) is given by:

r 1 r 2  r n 
 k s 
C ( s)
   ...
R( s ) s - p1 s - p2  s - pn 

96
Example 1:

Consider the following transfer function:

C ( s) 2s3  5s2  3s  6
 3
R( s) s  6s2  11s  6

97
Example 1:

Then:

C ( s ) 2s3  5s2  3s  6
 3
R( s ) s  6s2  11s  6
6 4 3
   2
s  3 s  2 s 1

98
Partial-fraction expansion with Matlab
To convert the partial-fraction expansion back to the
polyminal ratio C(s)/R(s), we have the command:

Note: the command printsys(num,den,’s’) prints the


num/den in terms of the ratio of polynominal in s

99
Example 2:
Expand this polynomial into partial-fraction with matlab:

C ( s) s2  2 s  3
 3
R( s) s  3s2  3s  1

100
Poles, zero and Stability.
An important requirement for a control system
is that should be stable.

A system can be defined as stable if every bounded.

Poles are the roots of a transfer function denominator and


zeros the roots of the numerator.

Example: s 1
G s  
( s  2)( s 2  1)
The pole: s = -2; s = +j; and s = -j.
The zero: s = 1

101
Pole-zero
Pole-zero plot:
plot

Poles and zeros can be represented on a


diagram, call the pole-zero plot.
The position of a pole is marked with a cross “x” and
The position of a zero with a small circle “o”.

Example: j
Pole Zero
x1
s 1
G s  
( s  2)( s 2  1)
x
-2 0
o
1 
x -1

102
Stability of Control Systems

The poles and zeros of the system are plotted in s-


plane to check the stability of the system.

j

LHP RHP

Recall s    j

s-plane

103
Stability of Control Systems

If all the poles of the system lie in left half plane the
system is said to be Stable.
If any of the poles lie in right half plane the system is
said to be unstable.
If pole(s) lie on imaginary axis the system is said to be
marginally stable. j

LHP RHP

s-plane

104
Stability of Control Systems
For example
C
G( s )  , if A  1, B  3 and C  10
As  B
Then the only pole of the system lie at
j
pole  3
LHP RHP

X 
-3

s-plane

105
Examples

Consider the following transfer functions.


 Determine whether the transfer function is proper or improper
 Calculate the Poles and zeros of the system
 Determine the order of the system
 Draw the pole-zero map
 Determine the Stability of the system

s3 G( s ) 
s
i) G( s )  ii)
s( s  2) ( s  1)( s  2)( s  3)

( s  3)2 s 2 ( s  1)
iii) G( s )  iv) G( s ) 
s( s 2  10) s( s  10)

106
_________________Routh-Hurwitz stability criterion

107
Routh stability criterion
The denominator of the transfer function is called as
Characteristic equation:
F(s) = ansn + an-1sn-1 + an-2sn-2 + … + a1s + a0 = 0

 If all the coefficients ai (i=1..n) are positive and none


are zero the system can be stable.

 If any coefficients is negative then the system is


unstable.
If all the elements in the first column of Routh Array
are positive, the system is thus Stable

108
Routh stability criterion
ansn + an-1sn-1 + an-2sn-2 + … + a1s + a0 = 0
Routh Array: Where:
sn an an 2 an 4 ... an 1an 2  an 3an
b1 
s n 1 an 1 a n 3 an 5 ... an 1
an 1an 4  an 5 an
s n 2 b1 b2 b3 ... b2 
an 1
s n 3 c1 c2 c3 ...
b1an 3  b2 an 1
... c1 
b1
s1 y1 y2 b1an5  b3an1
c2 
s0 z1 b1

109
Routh stability criterion

Example 1
F(s) = s4 + 10s3 + 3s2 + 19s + 1 = 0

110
Routh stability criterion

Example 2
F(s) = s5 + s4 + 4s3 + 4s2 + 2s + 1 = 0

111
Routh stability criterion

Example 3
F(s) = s5 + 10s4 + 16s3 + 160s2 + s + 10 = 0

112
Stability

Example 4
Given a system.

K
R(s) C(s)
s( s  2) 2

What range can K be if the system is to be stable?

113
Hurwitz stability criterion
The denominator of the transfer function is called as
Characteristic equation:
F(s) = ansn + an-1sn-1 + an-2sn-2 + … + a1s + a0 = 0

 If all the coefficients ai (i=1..n) are positive and none


are zero the system can be stable.

 If any coefficients is negative then the system is


unstable.
If all the Dk, k = 0. . n are positive, the system is thus
Stable

114
Hurwitz stability criterion

If all the Dk, k = 0. . n are positive, the system is thus


Stable

an-1 an-3 an-5 . . . . 0


an an-2 an-4 . . . . 0
0 an-1 an-3 . . . . 0
D0 = an, D1 = an-1, . . ., Dn =
0 an an-2 . . . . 0
. . .. . .. . . . . . . . . . .
0 0 a0

115
_________________________Relative stability

116
Relative stability
Relative stability is amount for a stable system how close it
is to being unstable
j
a

 0 

To obtain relative stability of a system, we shift left the plane by


replace s by (w - a) in the denominator of the TF, and using Routh-
Hurwitz test for stability.

117
Relative stability

Example 5

Given a system.

K
R(s) C(s)
s( s  2) 2

What range can K will result in relative stability = 1/2

118
_________Frequency-Response Analysis

119
Frequency-Response Analysis
By the term of Frequency-Response, we mean the steady-
state response of a system to a sinusoidal input. In
Frequency-Response methods, we vary the frequency of
the input signal over a certain range and study the resulting
response.
Frequency-Response methods were developed in 1930s
and 1940s by Nyquist, Bode, Nichols, and many others.
Obtaining Steady-State Outputs to Sinusoidal Inputs:

The Steady-State output of a transfer function system can


be obtained directly from the sinusoidal transfer function,
that is, the transfer function in which s is replaced by j ,
where  is frequency.

120
Frequency-Response Analysis

Let us assume that the input signal is given by:

r t   X sin t
Suppose that the TF can be written as a ratio of two
polynomials in s, that is:
C ( s ) ( s  z1 )( s  z2 )...( s  zm )
G s   
R( s ) ( s  s1 )( s  s2 )...( s  sn )
The Laplace Transform output C(s) is then:
C s   Gs Rs 

121
Frequency-Response Analysis

After waiting until steady-state conditions are reached, the


frequency response can be calculated by replace s in the
transfer function by j

G j   Me j  M
Where:
M is amplitude ratio of the output and input si nusoids
 is the phase shift between in the input si nusoid and the output si nusoid

122
Frequency-Response Analysis

The steady-state response of a stable, linear, time-


invariant system to a sinusoidal input doesn’t depend on
the initial conditions.
X
C ( s )  G ( s ) R ( s )  G s  2
s  2
a a b1 b2 bn
     ... 
s  j s  j s  s1 s  s2 s  sn

The invert Laplace transform is:

ct   ae jt  ae jt  b1e s1t  b2e s2t  ...  bn e snt (t  0)

123
124
Frequency-Response Analysis

For stable system, -s1, -s2, …,-sn have negative real parts,
so that the steady-state response become:

css t   ae jt  ae jt (t  0)

X XG  j 
a  G s  2 ( s  j ) s  j  
s  2
2j
X XG  j 
a  G s  2 ( s  j ) 
s  2 s  j
2j

125
Frequency-Response Analysis

G( j)  G( j) e j and G( j)  G( j) e j  G( j) e j

X G  j  e j X G  j  e j
a a
2j 2j
Then the output is:
e j (t  )  e  j (t  )
css t   X G  j 
2j
 X G  j  sint     C sint   

126
Frequency-Response Analysis

rt   X sint 

css t   C sint   

127
_________________Bode diagram

128
Bode diagram
A Bode diagram consists of two graphs:

- Plot of the logarithm of the magnitude of a sinusoidal transfer


function.

- Plot of the phase angle.

129
Bode diagram

Basic factors of G( j) H ( j)

The basic factors that very frequently occur in any arbitrary


transfer function are:

1.Gain K
2.Integral and derivative factor  j 1
3.First  order factors 1  jT 
1


4.Quadractic factors 1  2 ( j / n )  ( j / n ) 
2 1

130
1. The Gain K
R2
Electrical circuit R1
-
+
R
R
0

Transfer function G(s) = K dB


20lgK
Bode diagrams
lg

L( ) dB  20 lg G( j )  20 lg K

131
( )
±1
2. Integral and derivative factor jw
The logarithmic of integral and derivative in decibels are:
1
20 lg  20 lg  dB
j
The phase angle of integral and derivative are:  900

Bode diagrams of integral and derivative 132


( )
±1
2. Integral and derivative factor jw
The logarithmic of integral and derivative in decibels are:

20 lg j  20 lg  dB

The phase angle of integral and derivative are: 900

Bode diagrams of integral and derivative 133


( )
±1
3. First - order factors 1+ jwT
The log of magnitude of the first-order factor 1 / 1  jT  is:
1
20 lg  20 lg 1   2T 2 dB
1  jT
1
*    20 lg 1   2T 2  20 lg 1  0dB
T
1
*    20 lg 1   2T 2  20 lg T dB
T
The phase angle of the first-order factor 1 / 1  jT  is:
*  0    0
   arctan(T ) *   1    450
T
*       900

134
( )
±1
3. First - order factors 1+ jwT

The maximum error occurs at the corner frequency and


approximately equal to -3dB since:  20 lg 1  1  10 lg 2  3,03dB

135
( )
±1
3. First - order factors 1+ jwT

The log of magnitude of the first-order factor 1  jT  is:

20 lg 1  jT  20 lg 1   2T 2 dB

1
*    20 lg 1   2T 2  20 lg 1  0dB
T
1
*    20 lg 1   2T 2  20 lg T dB
T
The phase angle of the first-order factor 1  jT  is:
*  0    0
  arctan(T ) *   1    450
T
*       900

136
( )
±1
3. First - order factors 1+ jwT

Log-Magnitude curve, together with the


asymptotes, and phase angle curve for 1  jT 

137
Example

10s  1
G( s) 
( s  10)( s  100)

138
Example

139
Example

140
4.Quadratic factors
Control systems open possess quadratic factors of the form:

n2
G  j   2
s  2n s  n2

G  j  
1
1  2 ( j / n )  ( j / n ) 2

if   1 : G( j ) can be ex pessed as a product of two


first  order factors with real poles.

if 0    1 : G( j ) is the product of two


complex  conjugate factors

141
4.Quadratic factors
Control systems open possess quadratic factors of the form:

G  j   2
1
s  2 ( j / n ) s  ( j / n ) 2

The asymptotic frequency response curve may be obtained as follows:

20 lg G  j   20 lg
1

1  2 ( j / n )  ( j / n ) 2

1  jT   20 lg [1  ( / n )2 ]2  (2 / n )2

142
( ) ( )
2
4. Quadratic factors [1+ 2z jw / wn + jw / wn ]±1
if   n the log magnitude become :  20 lg(1)  0 dB

if   n the log magnitude become :



 20 lg(  / n )  40 lg
2
dB
n
And the phase-angle curves:

if 0    1 : G( j ) is the product of two complex  conjugate


factors

1  2 ( / n ) 
   arctan  2
1  2jT( j / n )  ( j / n ) 2
1  ( / n ) 

143
( ) ( )
2
4. Quadratic factors [1+ 2z jw / wn + jw / wn ]±1

At   0    00
At   n    900
At       1800

144
( ) ( )
2
4. Quadratic factors [1+ 2z jw / wn + jw / wn ]±1

The Resonant frequency and Resonant peak value:

The magnitude of: G  j  


1
1  2 ( j / n )  ( j / n )2
Is: G  j  
1
[1  ( / n )2 ]2  (2 / n )2

145
( ) ( )
2
4. Quadratic factors [1+ 2z jw / wn + jw / wn ]±1

If G  j  has a peak value at some frequency, this frequency is


called the resonant frequency. Since the numerator of G  j  is
constant, a peak value of G  j  will occur when:

g   [1  ( / n )2 ]2  (2 / n )2 is a mi nimum


2
   (1  2 ) 
2 2 2
g     n
  4 2
(1   2
)
 n2

146
( ) ( )
2
4. Quadratic factors [1+ 2z jw / wn + jw / wn ]±1

The minimum value of g   occurs at   n 1  2  . Thus


2

the resonant frequency is:

r  n 1  2 2  for 0    0.707

The magnitude of the resonant peak is:

M r  G  jr  
1
2 1   2

147
Plot Bode diagrams using Matlab
bode(num,den)
grid on
title(‘ title…’)

Example:

148
Manual plot

Steps for Bode plot


1.
2.
3.

149
Example

10s  1
G( s) 
( s  10)( s  100)

150
Example

151
Stability via Bode plots

• Gain margin is the magnitude value when


the phase is -1800.

• Phase margin is the phase difference from


-1800 when the magnitude is zero.

152
Homework

Given a system.

K
R(s) C(s)
s( s  2) 2

a. Plot Bode diagram when K = 1

b. Plot Bode diagram when K = 16

153
Stability via Bode plots

𝜔𝑏𝑖𝑒𝑛
Gain m arg in

Phase m arg in
𝜔𝑝ℎ𝑎

154
Stability via Bode plots

Using Matlab

G = tf(num,den)
Bode(G)
Margin(G)

155
Stability via Bode plots
The system is stable if the Gain margin >0
and the Phase margin >0

156
Example

Find Phase margin?

100(𝑠 + 10)
𝐺(𝑠) =
𝑠(𝑠 + 1)(𝑠 + 100)

157
158
159
𝑙𝑔𝜔

160
1
20

161
Dynamic system models

162
Dynamic system models
First-order and Second-order elements:
 The order of an element or system can be defined as being
the highest power of s in the denominator.
First-order element:

G s  
K
s  1 where K and  are const ants
Second-order element:
where : b0 is const ants
bon2 n is freely oscillate angular frequency
G s   2
s  2 n s  n2  is the damping ratio

163
Dynamic system models
Step input
 Rs  
1
r(t) = 1(t)
s
Ramp input:

 Rs   2
1
r(t) = t(t)
s
Parabolic input:
t2
()
r t =
2
()
t  Rs   3
s
1

Impulse input:
() ()
r t =d t ()
Þ R s =1

164
Dynamic system models
Step input to a first-order system:
C s 
G s  
K

s  1 R s 
 C s   .R s  
K K 1
.
s  1 s  1 s
(1 /  )
 K.
s[ s  (1 /  )]

Hence, for a unit step input, the output is:


ct   K 1  e  t / 

165
166
Dynamic system models
Ramp input to a first-order system:
C s 
G s  
K

s  1 R s 
 C s   .R s  
K K 1
. 2
s  1 s  1 s
(1 /  )
 K. 2
s [ s  (1 /  )]
Thus the output for a unit slope ramp is given by:

ct   K [t   1  e  t / ]

167
Dynamic system models
Impulse input to a first-order system:
K s 
G s  
K

s  1 R s 
 C s   .R s  
K K
.1
s  1 s  1
(1 /  )
 K.
[ s  (1 /  )]
Thus the output is:
ct   K 1 /  e  t /

168
Step input to a second-order system

169
Step input to a second-order system
bow2n (
C s )
()
G s =
s 2 + 2zw n s + w2n
=
R ( s)
bow 2n bow 2n
()
ÞC s = 2
s + 2zw ns + w n
2
.R s = 2 () .
(s + 2zw n s + w n ) s
2
1

bow2n
=
(s - m1 )(s - m 2 )s
where : m1 = -zw n + wn z 2 -1
m 2 = -zwn - wn z 2 -1
()
The type of response depends on the value of the damping factor z

170
Step input to a second-order system
when z > 1Þ m1 and m 2 are real
The system is said to be overdamped
Using partial fraction we have: Vc

()
1
Cs = +
A
+
B
s s - m1 s - m2
V

()
c t =1+ Aexp(m1t) + Bexp(m2 t)
0 t

b0 z b0 b0 z b0
where : A = - - B= -
2 x2 -1 2 2 x2 -1 2

171
Step input to a second-order system

when z = 1 Þ m1 =m 2 = -zwn
The system is said to be Critically damped
The output then become:
bow2n
()
Cs =
(s + wn ) 2 s Vc
V
The inverse transformation of this is:

()
c t = b0 [1- exp(-wn t) - wn t exp(-wn t)]
0 t

172
Step input to a second-order system
when z < 1Þ m1 and m 2 are complex
The system is said to be Underdamped
So we have: Vc
m1 = -zw n + jw n 1- z 2 V

m 2 = -zw n - jw n 1- z 2
Thus, the output is given: 0 t

é ù
()
c t = b0 ê1-
êë
1
exp(-zwn t)sin[wn (1- z 2 ).t + F]ú
úû
1- z 2
where : cosF = z

173
Step input to a second-order system

when z = 0 there is no damping


The system is said to be undamped
Thus, the output is given:

()
c t = b0 (1- sin wn t) Vc

The output thus oscillates with V


the under damped frequency w
n
0 t

174
Ramp input to a second-order system

175
Ramp input to a second-order system

bow2n C s ()
()
G s = 2 =
s + 2zw n s + w n R s
2
()
bow 2n bow2n
()
ÞC s = 2
s + 2zw n s + w n
2
.R s = 2 () 1
. 2
(s + 2zw ns + w n ) s
2

bow 2n
=
(s - m1 )(s - m 2 )s 2
where : m1 = -zw n + wn z 2 -1
m 2 = -zwn - wn z 2 -1

176
Ramp input to a second-order system
æA B D ö
()
C s = b0 çç 2 + +
C
+ ÷÷
è s s s - m1 s - m 2 ø
Using partial fraction we have:
()
c t = b0 [At + B+ Cexp(m1t) + Dexp(m2 t)]
2z
where : A = 1 B=-
wn
z 2z 2 -1 z 2z 2 -1
C= + D= -
w n 2w x2 -1 w n 2w x2 -1
n n
Ex. p.139

177
Impulse input to a second-order system

178
Impulse input to a second-order system
bow2n ( )
C s
()
G s =
s 2 + 2zw n s + w 2n
=
R ( s)
bow 2n bow2n
ÞC s = () s + 2zw n s + w
2 2 ()
.R s =
(s + 2zw n s + w )
2 2
.1
n n

bow2n
=
(s - m1 )(s - m 2 )

where : m1 = -zw n + wn z 2 -1
m 2 = -zwn - wn z 2 -1

179
Impulse input to a second-order system

Using partial fraction we have:

()
Cs =
A
+
B
s - m1 s - m 2
()
c t = Aexp(m1t) + Bexp(m2 t)
b0w n
where : A = -B =
2 x2 -1
Thus, we have:
b0w n
c t =() [exp(m1t) - exp(m 2 t)]
Ex. p.142
2 x2 -1

180
Impulse input to a second-order system
Vc
1

0 t

Response of a second-order system to a unit impulse at t=0

Ex. p.142

181
______________________Steady-state error

182
Steady-state error

The steady-state error is a measure of the


accuracy of a control system to track a command
input and is the error after all transient responses
to the input have decayed.

183
Steady-state error
The error is the difference between the
required output and the actual output signal.
For a closed loop system:
R(s) E(s) C(s)
G(s)
(-)
B(s)

H(s)

1
Es   Rs   Bs  E ( s)  R( s )
1  GH ( s)

184
Steady-state error

ess  lim e(t )


t 
sR( s)
lim e(t )  lim sE( s)  lim
t  s 0 s 0 1  GH ( s )

Example 1, p.168

185
Steady-state error

Steady-state error for a step input:

 Rs  
1
r(t) = 1(t)
s
1 1 1
ess _ p = lim s . =
s®o s 1+ G(s)H(s) 1+ K p

where: K p  lim G( s) H (s) : Positional error


s 0
constant
Example 3, p.172

186
Steady-state error

Steady-state error for a ramp input:

 Rs   2
1
r(t) = t(t)
s
1 1 1
ess _ v = lim s 2 . =
s®o s 1+ G(s)H(s) Kv

where: K v  lim sG(s) H ( s) : Velocity error constant


s 0

Example 4, p.173

187
Steady-state error

Steady-state error for a parabolic input:

t2
r t   t   Rs   3
1
2 s
1 1 1
ess _ a = lim s 3 . =
s®o s 1+ G(s)H(s) Ka

K
where: a  lim s 2
G( s) H ( s) : Acceleration error constant
s 0

Example 5, p.175

188
Given a system.

K
R(s) C(s)
s( s  2) 2

where K = 5
Find steady-state errors

189
Given a system.

K
R(s) C(s)
s( s  2) 2

If ess-v = 0.1, find K

190
Nyquist diagrams_________________________

191
Nyquist diagrams
The Nyquist diagram is a polar plot of the frequency response
of the system
On the Nyquist diagram the output, for a multi-amplitude sinusoidal input at
particular angular frequency, is specified by drawing a line of length equal
to the magnitude G( j ) at the phase angle  with the real axis.

Imaginary
R esponse at 
for G ( j )
G( j)

 Input
0 Real

192
Nyquist diagrams

In plotting Nyquist diagram there are four key points to be


represented on the diagrams:

1. The start of the plot where 0


2. The end of the plot where 
3. Where the plot crosses the real axis, i.e.   00 or  1800
4. Where it crosses the imaginary axes, i.e.    900

193
Nyquist diagrams
First-order systems:
Transfer functions is: G s   1
where  is const ant
s  1
 G j  
1
1  j

Magnitude:
G  j  
1
1   
2

Phase: j = 0 - arctan(wt)

194
Nyquist diagrams

( )
G jw =
1
j = 0 - arctan(wt)
( )
2
1+ wt
Nyquist Diagram

Where :   0  G( j)  1,   00 0.1

Where :     G( j)  0,   900 0

-0.1

Imaginary Axis
1 1
Where :    G( j)  ,   450 -0.2

 2 -0.3

-0.4

-0.5

-0.2 0 0.2 0.4 0.6 0.8 1 1.2


Real Axis

195
Nyquist diagrams
Second-order systems:
n2
Transfer functions is: G s   2
s  2 n s  n2
where : n is freely oscillate angular frequency
 is the damping ratio
Magnitude:
G  j  
1
1   /     2  /   
n
2 2
n
2

Phase:  2  / n  
   arctan  2
1   / n  

196
Nyquist diagrams

G  j  
1
1   /     2  /   
n
2 2
n
2

 2  / n  
Nyquist Diagram
1.5

   arctan  2
1

1   / n   0.5

-0.5

Imaginary Axis
-1

-1.5

-2

-2.5

-3

-3.5
-1 0 1 2 3 4 5
Real Axis

197
Given a system.

K
R(s) C(s)
s( s  2) 2

where K = 5
Plot Nyquist diagram

198
Nyquist diagrams

Nyquist diagrams using Matlab

nyquist(num,den)
grid on
title(‘ title…’)

199
Nyquist stability criterion

If the Open-loop system is stable then the Closed-loop


system will be stable

The system is stable If the


magnitude of phase angle of
1800 not exceed 1 then the
polar plot must not enclose 0
the -1 point on the real axis. -1   1800

Stable
Unstable

200
Nyquist stability criterion

Example 12: Page 280

Nyquist diagram

201
Nyquist stability criterion

Example 13: Page 282

202
Gain margin and Phase margin
Gain margin is defined as being the factor by which the
system gain, i.e. magnitude, can be increased before
instability occurs.
It is thus the amount by which the magnitude at 1800 has to
be increased to reach the critical value of 1.

1  Gain m arg in * G j 1800


In decibels
Gain m arg in  20 lg1  20 lg G j 1800  20 lg G j 1800

203
Gain margin and Phase margin

G j 1800

Gain m arg in  20 lg G j 1800

 If the Nyquist plot never crosses the negative real axis


then the gain margin is infinite.

 If the Nyquist plot passes through the negative real axis at


value less than 1 then the gain margin is positive.

204
Gain margin and Phase margin
G j 1800

 If the Nyquist plot passes through the axis at 1 then the gain
margin is zero.

 If the Nyquist plot passes through the negative real axis at


value greater than 1, i.e. the plot encloses the -1 point, then the
gain margin is negative.

205
Gain margin and Phase margin

Phase margin is defined


as the angle through
which the Nyquist plot
must be rotated in order Phase
margin
that the unity magnitude
point on the plot passes
G j  1
through the -1 point on the
real axis.

Example 14, page 283

206
Given a system.

K
R(s) C(s)
s( s  2) 2

where K = 5
Plot Nyquist diagram and find GM and PM

207
208
209
_____________________________________ Root locus analysis

210
Root locus analysis_______________________
The properties of the root locus can be derived from the
general closed-loop control system:

R(s) E(s)
+ KG(s)
- C(s)
B(s)
H(s)

Transfer function : Ms 


C(s) KG(s)

R(s) 1  KG(s).H(s)

211
________________________________________ Root locus analysis

The denominator becomes zero, so we call:


Characteristic equation:

Fs  1  KG(s).H(s)  0

 KG(s).H(s)  1  1(2n  1).1800


 KG(s).H(s)  1
Thus:
 KG(s).H(s)  (2n  1).1800

The roots of denominator of a transfer function of a system,


the so-called poles, determine the general form of the
transient response of that system.
Root locus is a technique study relationship between
behaviour of systems and the position of their poles
212
________________________________________ Root locus analysis

10 Steps for plotting of a root locus:

1. Obtain the characteristic equation of the open-loop transfer


function for the system.

2. Determine the positions of the poles (n) and zeros. (m)

3. Determine the number of loci.


=> Root locus branch = max(poles, zeros)
=> Root loci start at the n_poles, where K=0
=> Root loci end at the m_zeros, where K   , if there are
more poles than zeros, the usual case, then m loci end at the m_zeros
and the remaining (n-m) loci end at infinity

4. Plot the loci symmetry on the real axis.

213
________________________________________ Root locus analysis

10 Steps for plotting of a root locus:

5. Determine the angles of the asymtotes:


(2l  1)
k  
nm

6. Obtain the intersection of the asymptotes with the real axis.


Center of gravity or Centriod of the asymtotes:


p1  p2  ...  pn   z1  z2  ...  z m 
nm
7. Obtain the intersection of the asymptotes with the imaginary axis.

putting s  j in characteristic equation and solve equation


or u sin g Routh array to find the lim iting of K that give stability

214
________________________________________ Root locus analysis

10 Steps for plotting of a root locus:

8. Determine the breakaway points:

- Breakaway points are used for where two or more loci meet at a point
and subsequently “break away” from that point along separate paths.
- The breakaway points occur at these point of which, for the
characteristic equation, dK/ds = 0

9. Obtain the angles of departures from complex points and the


angle of arrival at complex zeros:
  180  ( s  z1 )  ( s  z2 )  ...  ( s  zm ) 
 ( s  p1 )  ( s  p2 )  ...  ( s  pn )

10. Sketch the root loci

215
________________________________________ Root locus analysis

Example 1: Sketch the root loci for the system with open-
loop transfer function:

Gs 
K
s  1s  2s  3

216
________________________________________ Root locus analysis

Completed root loci plot

-3 -2 -1
0

-
1,42

217
________________________________________ Root locus analysis

Example 2: Sketch the root loci for the system with open-
loop transfer function:

K(s  1)
Gs 
s  2  3 js  2  3 j

218
________________________________________ Root locus analysis

Completed root loci plot

1990

-2+3j

-2 -1 0
-4,16

-2-3j

219
________________________________________ Root locus analysis

Root locus using Matlab

rlocus(num,den)

220
________________________________________ Root locus analysis

Homework:
Sketch the root loci for the system with open-loop
transfer function:

()
G s =
K
(
s s2 + 8s + 20 )

221
________________________________________ Root locus analysis

222
________________________________________ Root locus analysis

Example 9, page 222

223
Compensation according to the desired frequency response

224
Series compensation

225
Compensation according to the desired frequency response
Example:
The desired frequency response is shown in Fig. 1.
Given

Determine Gc(s)

R(s) + Gc(s) G(s) C(s) -20dB/dec


100
-
1 10

-40dB/dec
Fig.1

226
Compensation according to the desired frequency response
Example: 1
GH(s) 
s( s  1)

-20dB/dec
100 1005
GH(s) 
10
s( s  100)
-40dB/dec
Fig.1

227
Compensation according to the desired frequency response

Solution:
In terms of the desired frequency response we have:
10
G H d esi re ( s )  G c ( s ) G H ( s )  -20dB/dec
s ( 0 .0 1 s  1 ) 100
10

1 Fig.1 -40dB/dec
G H (s) 
s ( s  1)

GH desire ( s ) 10 1
Gc ( s)  
GH ( s ) s ( 0 .01s  1) s ( s  1)
( s  1)
 10 
( 0 .01s  1)

228
Parallel compensation

229
The configuration of the Parallel compensation

G’0(s) G’20(s)
R(s) C(s) R(s) C(s)
G 0 (s) G10 G 20
- - - -
GC GC

The basic Parallel compensators


G c (s) = a ® proportional(position) parallel compensation;
G c (s) = as[1 (ts +1)] ® differential(speed) parallel compensation;
G c (s) = as 2 [1 (ts +1)]
® 2th - differential(acceleration) parallel compensation;

230
Function of the feedback:
- Decrease the time constant of the encircled elements →
Quicken the response of the encircled elements-may be;
G’20(s)
R(s) C(s)
G10 G 20
For example: - -
GC

'
K 20 K K
G 20 ( s )  , G c ( s )    G 20
'
(s)  20
 ' 20
Ts  1 Ts  1  K 20 T s  1
T
T 
'
 , but also K 20
'
 K 20 /( 1  K 20 ) 
1  K 20

231
Function of the feedback:
- Decrease the time constant of the encircled elements →
Quicken the response of the encircled elements-may be;
G’20(s)
R(s) C(s)
G10 G 20
For example: - -
GC

K 20 K 20
But if G c ( s )   s  '
G 20 (s)   '
Ts  1  K 20 s T s  1
T '  (T  K 20 )  , K 20  K 20

232
Function of the feedback:
- Decrease the time constant of the encircled elements →
Quicken the response of the encircled elements-may be;
'
K 20 K 20 K 20
For example: G 20 ( s )  , G c ( s )    G 20 ( s ) 
'
 '
Ts  1 Ts  1  K 20 T s  1
T
T 
'
 , but also K 20'
 K 20 /( 1  K 20 ) 
1  K 20
K 20 K 20
But if G c ( s )   s  G 20 ( s ) 
'
 '
Ts  1  K 20 s T s  1
T '  (T  K 20 )  , K 20  K 20

- Impair (weaken) the influences of the disturbance to the


encircled elements.
- Make the performance of the encircled elements to be desired .

233
The design procedure of the parallel compensator
- Design the desired characteristics, such as the desired Bode
diagram, of the encircled elements in terms of the system’s
analysis.
- Choose the appropriate feedback compensators to get the
desired characteristics. G’20(s)
R(s) C(s)
G10 G 20
- -
Example: For the system shown GC
in Fig.1, G20=10/s2, the desired
G’20(jω) shown in Fig.2. Fig.1
determine the Gc. -40dB/dec
-20dB/dec 10
0.1 1
-40dB/dec
Fig.2

234
-40dB/dec
-20dB/dec 10
0.1 1
-40dB/dec
Fig.2

Solution:

235

Оценить