Scribd
Upload
9 views

Chapter 2

The document defines various terms related to feedback control systems including open-loop and closed-loop transfer functions, types of systems, zeros and poles, and sensitivity. It discusses how feedback can improve accuracy and reduce sensitivity to changes in the plant transfer function compared to an open-loop system.

Uploaded by

Firas
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
9 views

Chapter 2

The document defines various terms related to feedback control systems including open-loop and closed-loop transfer functions, types of systems, zeros and poles, and sensitivity. It discusses how feedback can improve accuracy and reduce sensitivity to changes in the plant transfer function compared to an open-loop system.

Uploaded by

Firas
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

Definition:

G(s) is called Direct or Forward Transfer


Function. It is not the Open-Loop Transfer
Function unless the feedback is unity.
GH(s) is called the Open-Loop Transfer
Function. It is a common mistake to refer to
G(s)as the Open-Loop Transfer Function.
Cs Gs
R s 1 GH s is called the Closed-Loop
Transfer Function.
Ea s
Rs is called the Actuating Signal Ratio.
In the previous lectures, we talked about
Control Action, also known as Actuating
Signal, or Error Signal for the negative unity
feedback system. For a negative unity
feedback system, H(s) = 1. Then, Ea(s) = E(s)
= C(s) – R(s). In this case (especially for the
Es
regulators), R s is called the Error Ratio.

Bs
Rs is called the Primary Feedback Ratio.
1 GH s 0 is called the Characteristics
Equation. It is the denominator of the
closed-loop transfer function.
Note: In the first half of the semester, we will
analyse and design the feedback control system
using the closed-loop transfer function. However,
in the second half of the semester, we will go one
step further and we will analyse and design the
closed-loop system using the open-loop transfer
function because in reality the open-loop transfer
function is available, and the performance of the
system is to be improved by making it a
closed-loop system. Therefore, GH(s) will be
considered to understand the characteristics of
1 GH s 0.
We will get to see why a positive feedback is
more unstable than a negative feedback or in
other words, a positive feedback is less stable

than a negative feedback. In most cases, we will


use negative feedback to improve the
performance and the characteristics of the control
system.
In general, the open-loop transfer function GH(s)
can be written in canonical form of
m
K s zi
i 0
GH s n
s s pj ,
j 0
where + n > m.
Definition:
Given the open-loop transfer function in the
canonical form of
m
K s zi
i 0
GH s n
s s pj ,
j 0

the system is called type system.

For example,
s 2 s 2
GH s
s 4 s0 s 4 ,
is a type 0 system, because the exponent of s, i.e.
, is zero.
s 2 s 2
GH s
ss 4 s1 s 4 ,
is a type 1 system because the exponent of s, i.e.
, is one.
Definition:
Roots of the numerator of the open-loop
transfer function are called the zeros of the
open-loop system.
Roots of the denominator of the open-loop
transfer function are called the poles of the
open-loop system.

The closed-loop system can be represented as:


closed-loop system.
Roots of the denominator of the closed-loop
transfer function are called the poles of the
closed-loop system.
Also, we can define the zeros of the system as
those values of s that force T(s) 0 and the poles
as those values of s that force T(s) . Based on
this definition the number of poles is always
equal to the number of zeros.
Characteristic of Feedback Control System:
1. Effects of Feedback on Accuracy:
First we are going to look at the effects of
feedback on accuracy. By accuracy we mean
the ability to reproduce the input. For example,
if we set the thermostat at 75o F, we want to
see how close the room temperature, i.e. the
output, gets to 75o F. We want to show that the
closed-loop system is more capable than the
open-loop system to achieve this goal.

Furthermore, when a plant transfer function


G(s) or more generally the open-loop transfer
function GH(s) has certain undesirable
characteristics, do not expect to be able to
change the undesirable characteristics by
building a new plant because those undesirable
characteristics are part of the nature of the
plant. For example, if the steady-state error of
the open-loop system is found to be too large,
we cannot build a system that its steady-state
error is small because the large steady-state
error is part of the nature of the system.
Therefore, the nature of a system cannot be
changed, but its performance can be improved
by converting it to a closed-loop system.
For the open-loop system, since G(s) is not
equal to unity, it is impossible to rebuild the
output from a given input. In general,
C(s) = G(s) R(s).
However, for the closed-loop system, it is

possible, under certain conditions, to rebuild


the output from the input. In general,
Gs
C (s) R( s)
1 GH s .
Therefore, if the feedback is unity and
G(s)>> 1,
then C(s) R(s), meaning that the output will
closely follow the input.
2. Effects of Feedback on Sensitivity:
It will be shown that mathematically the open-
loop system is more sensitive to the changes in
the plant than the closed-loop system, or at
least the changes in the plant can be better
controlled in the closed-loop system than in the
open-loop system.
Any system, regardless of its nature, will
eventually change over time, and we have to
eliminate or reduce the effects of these changes

on the performance of the aging system. Note


that some systems are economically
irreplaceable. Suppose over some time,
G(s) G(s) + G(s).
Then, for the open-loop system,
C(s) + C(s) =[G(s) + G(s)]R(s). Therefore,
C(s) = G(s)R(s), and cannot be eliminated,
meaning that there will be a change in the
system output no matter what.
For the closed-loop system, it is possible to
eliminate or at least minimize the impacts of
the changes on the performance of the system.
In general,
Gs Gs
C s Cs Rs
1 Gs Gs H s
or
Gs Gs
Cs Rs Cs
1 Gs Gs H s

Gs Gs
Rs Continued on
Next Page

1 Gs Gs H s

Gs
Rs
1 GH s .
Simplifying the expression, we get
Gs
C s Rs
1 Gs G s H s 1 GH s .
Since G(s) >> G(s),
Gs
Cs 2
Rs
1 GH s .
2
Notice that we have 1 GH s factor in the
denominator of C(s). Therefore, we should be
able to at least reduce the impact of the
changes on the system output.
Definition: Sensitivity of T with respect to
T
( is any system parameter) denoted as S is
defined as
T s
T ( s) T (s) ( s)
ST
s ( s) T ( s)
(s) ,

or

LnT s
ST
Ln s .
For the open-loop system,
Gs
Gs
SGTO 1
Gs
Gs .
TO
Nothing can be done to reduce S G since it is
a constant.

For the closed-loop system,


G s
T ( s) G ( s) 1 GH s G ( s)
S GTC
G (s) T (s) G(s) G ( s)
1 GH s
1 GH s GH s 1
1 GH s
1 GH s
2
1 GH s .
Once again, since we have a factor of
T
1 GH s in the denominator of S GC , we can
reduce the impact of the changes on the
system.

Note that ideally a sensitivity of zero is


desirable. We can also determine the sensitivity
with respect to a certain parameter of the
system. For example, we can run the
sensitivity test as a cost analysis to determine
which parameter is more critical, thus by
making more investment in the implementation
of that parameter we make it more resistant to
changes.
For the open-loop system, since there is no
TO
S
H(s), H 0.
However, for the closed-loop system, we need
to investigate the impact of the feedback
element H(s) on the system sensitivity. Thus,
G(s)
T ( s) H ( s) 1 GH s H (s)
S HTC
H (s) T ( s) H (s) G ( s)
1 GH s ,

or

TC G2 s H ( s) 1 GH s GH s
S
1 GH s .
H 2
1 GH s Gs

Notice that if GH(s)>> 1, we get


S HTC 1 1.

Therefore, if we want to improve the


performance or characteristics of a system by
making it a closed-loop system, we must be
very concerned about the changes in H(s) over
a long period of time since the sensitivity
analysis indicates that it is very critical

You might also like