CTRL Sys

CONTROL SYSTEMS
Sisil Kumarawadu, PhD, SMIEEE

Senior Professor in Electrical Engineering
Head, Department of Electrical Engineering
University of Moratuwa, SRI LANKA
Distinguished Professor-Shanghai University of Electric Power
Chairman of the BoG-Arthur C. Clarke Institute for Modern Technologies
I have a RL circuit and a variable dc voltage source. How do I control the
current in the circuit to follow a predefined profile?
a] manually b] automatically
𝑡
Feedback (closed-loop) Control System
¿ Reference Actual current

current profile 𝑒 (𝑡 ) Current 𝑣 𝑡 (𝑡 ) in the circuit
+ Plant
- controller
𝑖(𝑡 )
Power supply + RL circuit
𝑡
Feedback
element
Current Sensor
Block diagram of a standard error feedback control system

Understanding the Plant
 By working with the plant sufficiently
longer
 By mathematically modeling the plant

Eg.
 Plant is a dynamical system. Mathematical

models of dynamical systems are always
differential equations
Proportional (P) Controller
In the proportional control law, the control action is

proportional to the instantaneous value of the error
Here, , is a positive constant to be chosen by the

controller designer.
Linear Time-invariant Dynamical
Systems
• Dynamical system is a system whose states vary with time
• If system dynamics or the equations do not vary with
time, such a system is called a time-invariant system
• Mathematical models of a linear system contain linear

differential equations
• Review “Linear Systems Theory”. Remember here we are
considering dynamical systems and not static systems
Laplace Transform
The Laplace transform of a time domain function or signal, f(t), is defined as

F ( s ) e  st
f (t )dt
0
Laplace Transform
Can analyse linear time-invariant systems
• Born23 March 1749

Beaumont-en-Auge, Normandy,
France
• Died5 March 1827(1827-03-05)
(aged 77) Paris, France
• Nationality French
• FieldsAstronomer and
Mathematician
• InstitutionsÉcole Militaire (1769–
1776)
• Academic advisorsJean d'Alembert
Christophe Gadbled
Pierre Le Canu
• Doctoral students
Siméon Denis Poisson
Pierre-Simon Marquis de Laplace

(1749–1827). Source: WIKIPEDIA
Laplace transform of
some common time
functions in engineering
Transfer Function
Recall the RL circuit
By taking Laplace transform, we get
Assuming initial conditions to be zero, i.e.,

and rearranging the above equation we get
Feedback (closed-loop) Control System
Reference I/P Measured O/P

+ Controller Plant
-
Is a signal or information Controlled system
processing device
Feedback
element
Sensors
Block diagram of a standard error feedback control system

Unit Step Response
Control System—Laplace
Domain
Reference I/P O/P
+
-
GC (s ) GP (s ) GP (s )
R (s ) E (s ) 𝑉 𝑡 (𝑠) I (s )
H (s )
E ( s ) R ( s )  H ( s ) I ( s ) GC (s ) Controller T/F
GC ( s )GP ( s ) E ( s ) I ( s ) GP (s ) Plant T/F
I (s) GC ( s )GP ( s ) H (s ) FB element T/F

M (s)   I (s ) Laplace transform
R ( s ) 1  GC ( s )GP ( s ) H ( s ) of the O/P
Closed-loop Laplace transform
transfer function (CL T/F) R (s ) of the reference I/P
Example
• Assume unity feedback, i.e.,
• As the controller, we try
Here, , is a positive constant to be chosen by the controller designer. This

means
• The closed-loop transfer function is

𝐼 ( 𝑠)
=
𝐾
𝑅 ( 𝑠 ) 𝐿𝑠+( 𝑅+ 𝐾 ) Unit Step Response
Step Response
Try K = 100
0.8
0.7
System: M
Time (sec): 0.00833
Amplitude: 0.712
MATLAB® code is 0.6
0.5
L=0.2; R=40;
Amplitude
0.4
K=100;
M=tf([K],[L R+K]);
0.3
step(M), grid on 0.2
0.1
0
0 1 2 3 4 5 6 7 8 9
Time (sec) -3
x 10
steady state error (off-set)

Steady state error with increasing
gain
K Steady state
error (%)
100 28.8
500 7.4
1000 3.8
Step Response
1
System: K = 1000
0.9 Time (sec): 0.00202 System: K = 500
Amplitude: 0.962 Time (sec): 0.00365
Amplitude: 0.926
0.8 System: K = 100
Time (sec): 0.00878
Amplitude: 0.713
0.7
0.6 K = 100
K = 500
Amplitude
0.5 K = 1000
0.4
0.3
0.2
MATLAB® code is
L=0.2; R=40;
0.1 for K=[100 500 1000], M=tf([K],[L R+K]);
step(M), hold on; end, grid on
0
0 1 2 3 4 5 6 7 8 9
Time (sec) -3
x 10
In the proportional control law, the control
Proportional (P) Controller
action is proportional to the instantaneous value

of the error
By taking the Laplace transform and zeroing the initial

conditions, we get
Here, , is a positive constant to be chosen by the controller

designer.
In the proportional control law, the control
Integral (PI) Controller
action is proportional to the time integral of

the error
Here, , are positive constants to be chosen by the

controller designer. By taking the Laplace transform and
zeroing the initial conditions, we get
Integral (I) Controller
Step Response
1
0.9
0.8
0.7
0.6
Amplitude
0.5
0.4
𝑡
𝑣 𝑡 ( 𝑡 )=𝐾 𝐼∫ 𝑒 ( 𝑡 ) 𝑑𝑡
0.3
0.2
0.1 0
0
0 1 2 3 4 5 6 7 8 9
Time (sec) -3
x 10
Let us try PI control
The closed-loop transfer function is
Let .
Check the step response:
R=40; L=0.2;
Kp=100; Ki=500;
M=tf([Kp Ki],[L R+Kp Ki]); step(M)
Step Response
1
0.9
0.8
0.7
0.6
Amplitude
0.5
0.4
0.3
0.2
0.1
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Time (sec)
PI Controller
 P control action is used to obtain a faster

response
 I control action is used to zero the steady-

state error

CTRL Sys

Uploaded by

Copyright:

Available Formats

CTRL Sys

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CTRL Sys

Uploaded by

Copyright:

Available Formats

CONTROL SYSTEMS

Sisil Kumarawadu, PhD, SMIEEE

¿ Reference Actual current

Block diagram of a standard error feedback control system

 By mathematically modeling the plant

 Plant is a dynamical system. Mathematical

In the proportional control law, the control action is

Here, , is a positive constant to be chosen by the

• Mathematical models of a linear system contain linear

The Laplace transform of a time domain function or signal, f(t), is defined as

• Born23 March 1749

Pierre-Simon Marquis de Laplace

By taking Laplace transform, we get

Assuming initial conditions to be zero, i.e.,

Reference I/P Measured O/P

Block diagram of a standard error feedback control system

GC ( s )GP ( s ) E ( s ) I ( s ) GP (s ) Plant T/F

I (s) GC ( s )GP ( s ) H (s ) FB element T/F

Here, , is a positive constant to be chosen by the controller designer. This

• The closed-loop transfer function is

MATLAB® code is 0.6

step(M), grid on 0.2

steady state error (off-set)

action is proportional to the instantaneous value

By taking the Laplace transform and zeroing the initial

Here, , is a positive constant to be chosen by the controller

action is proportional to the time integral of

Here, , are positive constants to be chosen by the

The closed-loop transfer function is

Check the step response:

 P control action is used to obtain a faster

 I control action is used to zero the steady-

You might also like