Dynamic Behavior of

Closed-Loop Control
Systems
C
h
a
p
t
e
r

1
1

C
h
a
p
t
e
r

1
1

Next, we develop a transfer function for each of the five elements
in the feedback control loop. For the sake of simplicity, flow rate
w
1
is assumed to be constant, and the system is initially operating
at the nominal steady rate.
Process
In section 4.1 the approximate dynamic model of a stirred-tank
blending system was developed:
( ) ( ) ( )
1 2
1 2
(11-1)
1 1
K K
X s X s W s
s s
| | | |
' ' '
= +
| |
+ +
\ . \ .
where
1
1 2
1
, , and (11-2)
w V x
K K
w w w
t

= = =
C
h
a
p
t
e
r

1
1

C
h
a
p
t
e
r

1
1

The symbol denotes the internal set-point composition
expressed as an equivalent electrical current signal. is
related to the actual composition set point by the
composition sensor-transmitter gain K
m
:
( )
sp
x t
'
( )
sp
x t
'
( )
sp
x t
'
( ) ( )
(11-7)
sp m sp
x t K x t
' '
=
C
h
a
p
t
e
r

1
1

Current-to-Pressure (I/P) Transducer
The transducer transfer function merely consists of a steady-state
gain K
IP
:
( )
( )
(11-9)
t
IP
P s
K
P s
'
=
'
Control Valve
As discussed in Section 9.2, control valves are usually designed so
that the flow rate through the valve is a nearly linear function of
the signal to the valve actuator. Therefore, a first-order transfer
function is an adequate model
( )
( )
2
(11-10)
1
v
t v
W s
K
P s s
'
=
'
+
C
h
a
p
t
e
r

1
1

Composition Sensor-Transmitter (Analyzer)
We assume that the dynamic behavior of the composition sensor-
transmitter can be approximated by a first-order transfer function,
but
m
is small so it can be neglected.
( )
( )
m
m
X s
K
X s
'
=
'
Controller
Suppose that an electronic proportional plus integral controller is
used.
( )
( )
1
1 (11-4)

c
I
P s
K
E s s
'
| |
= +
|
\ .
where and E(s) are the Laplace transforms of the controller
output and the error signal e(t). K
c
is dimensionless.
( )
P s
'
( )
p t
'
C
h
a
p
t
e
r

1
1

C
h
a
p
t
e
r

1
1

1. Summer
2. Comparator
3. Block
Blocks in Series
are equivalent to...
G(s)X(s) Y(s) =
C
h
a
p
t
e
r

1
1

C
h
a
p
t
e
r

1
1

Closed-Loop Transfer Functions
Indicate dynamic behavior of the controlled process
(i.e., process plus controller, transmitter, valve etc.)
Set-point Changes (Servo Problem)
Assume Y
sp
= 0 and D = 0 (set-point change while disturbance
change is zero)
(11-26)
Disturbance Changes (Regulator Problem)
Assume D = 0 and Y
sp
= 0 (constant set-point)
(11-29)
*Note same denominator for Y/D, Y/Y
sp
.
( )
( ) 1
m c v p
sp
c v p m
K G G G
Y s
Y s G G G G
=
+
( )
( ) 1
d
c v p m
G Y s
D s G G G G
=
+
C
h
a
p
t
e
r

1
1

C
h
a
p
t
e
r

1
1

C
h
a
p
t
e
r

1
1

C
h
a
p
t
e
r

1
1

Figure 11.16 Block diagram for level control system.
C
h
a
p
t
e
r

1
1

C
h
a
p
t
e
r

1
1

C
h
a
p
t
e
r

1
1

EXAMPLE 1: P.I. control of liquid level
Block Diagram:
C
h
a
p
t
e
r

1
1

Assumptions

1. q
1
, varies with time; q
2
is constant.

2. Constant density and x-sectional area of tank, A.

3. (for uncontrolled process)

4. The transmitter and control valve have negligible dynamics
(compared with dynamics of tank).

5. Ideal PI controller is used (direct-acting).
) h ( f q
3
=
0 K
As
1
) s ( G
As
1
) s ( G
K ) s ( G
K ) s ( G
s
1
1 K ) s ( G
C L
P
V V
M M
I
C C
> =
=
=
=
|
|
.
|

\
|
t
+ =
For these assumptions, the transfer functions are:
C
h
a
p
t
e
r

1
1

1
1
d
C V P M
G Y H
D Q G G G G
'
= =
'
+
M V
I
C
K
As
K
s
K
As
D
Y
|
.
|

\
|

|
|
.
|

\
|
+
=
1 1
1 1
1
t
M P C I M V C I
I
K K K s K K K s A
s
D
Y
+ +
=
t t
t
2
0
2
= + +
M P C I M V C I
K K K s K K K s A t t
1 s 2 s
K
) s ( G
2 2
+ ,t + t
=
The closed-loop transfer function is:
Substitute,
Simplify,
Characteristic Equation:
Recall the standard 2
nd
Order Transfer Function:
(11-68)
(2)
(3)
(4)
(5)
C
h
a
p
t
e
r

1
1

For 0 < , < 1 , closed-loop response is oscillatory. Thus
decreased degree of oscillation by increasing K
c
or t
I
(for constant
K
v
, K
M
, and A).
To place Eqn. (4) in the same form as the denominator of the
T.F. in Eqn. (5), divide by K
c
, K
V
, K
M
:
0 1 s s
K K K
A
I
2
M V C
I
= + t +
t
A
K K K
2
1
I M V C
t
= ,
1 0 < , <
Comparing coefficients (5) and (6) gives:
Substitute,
t
t
= , t = ,t
t
= t
t
= t
2
2
K K K
A
K K K
A
I
I
M V C
I
M V C
I
2
unusual property of PI control of integrating system
better to use P only
C
h
a
p
t
e
r

1
1

Stability of Closed-Loop
Control Systems
C
h
a
p
t
e
r

1
1

Proportional Control of First-Order Process
Set-point change:
M P V C OL
OL OL
OL
sp
M P V C
M P V C
sp
K K K K K
K K
K
K
s
K
Y
Y
s
K K K K
s
K K K K
Y
Y
=
+
=
+
=
+
=
+
+
+
=
1 1
1
1
1
1
1 1
1
1
t
t
t
t
t
C
h
a
p
t
e
r

1
1

1 +
=
s
K
G
P
P
t
) , , (K
gains constant , ,
V m C
M C V
K K
G G G
Set-point change = M
Offset =
See Section 11.3 for tank example
( )
1
1
( ) 1
t
y t K M e
t
=
( ) ( )
1
sp
OL
M
y y
K
=
+
C
h
a
p
t
e
r

1
1

Closed-Loop Transfer function approach:
First-order behavior
closed-loop time constant
(faster, depends on K
c
)
1
1
1
1
+
+
+
=
+ +
=
s
KK
KK
KK
KK s
KK
Y
Y
C
C
C
C
C
sp
t
t
C
KK +
=
1
t
C
h
a
p
t
e
r

1
1

C
h
a
p
t
e
r

1
1

General Stability Criterion
Most industrial processes are stable without feedback control.
Thus, they are said to be open-loop stable or self-regulating. An
open-loop stable process will return to the original steady state
after a transient disturbance (one that is not sustained) occurs. By
contrast there are a few processes, such as exothermic chemical
reactors, that can be open-loop unstable.
Definition of Stability. An unconstrained linear system is said to
be stable if the output response is bounded for all bounded
inputs. Otherwise it is said to be unstable.

C
h
a
p
t
e
r

1
1

Effect of PID Control on a Disturbance Change
For a regulator (disturbance change), we want the
disturbance effects to attenuate when control is applied.
Consider the closed-loop transfer function for proportional
control of a third-order system (disturbance change).
K
c
is the controller function, i.e., .
3 2
8
( ) ( )
6 12 8 8
C
Y s D s
s s s K
=
+ + + +
is unspecified
) (s D
( )
3
8
1 1
2
V M P d
G G G G
s
= = = =
+
C C
K ) s ( G =
C
h
a
p
t
e
r

1
1

Let
If K
c
= 1,
Since all of the factors are positive, ,
the step response will be the sum of negative
exponentials, but will exhibit oscillation.
C
K s s s s 8 8 12 6 ) (
2 3
+ + + + = I
( )( ) ( )( )( ) j s j s s s s s s 3 1 3 1 4 4 2 4 ) (
2
+ + + + = + + + = I
|
.
|

\
|
=
+
at
e
a s
1
C
h
a
p
t
e
r

1
1

If K
c
= 8,
3 2 2
( ) 6 12 72 ( 6)( 12) s s s s s s I = + + + = + +
Corresponds to sine wave (undamped), so this case is
marginally stable.
If K
c
= 27
Since the sign of the real part of the root is negative, we
obtain a positive exponential for the response. Inverse
transformation shows how the controller gain affects the
roots of the system.
Offset with proportional control (disturbance step-
response; D(s) =1/s )
( )( ) 28 2 8 224 12 6 ) (
2 2 3
+ + = + + + = I s s s s s s s
C C
s
C
K K
s sY t y
s K s s s
s Y
+
=
+
= =

+ + + +
=

1
1
8 8
8
) ( lim ) (
1
8 8 12 6
8
) (
0
2 3
( )( )( ) j 3 3 1 s j 3 3 1 s 8 s + + =
C
h
a
p
t
e
r

1
1

Therefore, if K
c
is made very large, y(t) approaches 0,
but does not equal zero. There is some offset with
proportional control, and it can be rather large when
large values of K
c
create instability.
C
h
a
p
t
e
r

1
1

Integral Control:
( )
s
K
) s ( G ) s ( E
s
K
) s ( P t d t e
K
P
I
C
C
I
C
t
0
I
C
t
=
t
= ' '
t
=
}
For a unit step load-change and K
c
=1,
( )
) ( y 0 ) s ( sY lim
s
1
8
2 s s
s 8
) s ( Y
0 s
I
3
= =
t
+ +
=

no offset
(note 4
th
order polynomial)
adjust K
c
and t
I
to obtain satisfactory response (roots of
equation which is 4
th
order).
PI Control:
no offset
PID Control: (pure PID)
0 ) s ( sY lim
s
1
s K 8
K 8
) 2 s ( s
s 8
) s ( Y
s
1
1 K ) s ( G
0 s
C
I
C
3
I
C C
=
+
t
+ +
=
|
|
.
|

\
|
t
+ =

|
|
.
|

\
|
t +
t
+ = s
s
1
1 K ) s ( G
D
I
C C
No offset, adjust K
c
, t
I
, t
D
to obtain satisfactory result
(requires solving for roots of 4
th
order characteristic
equation).
Analysis of roots of characteristic equation is one way to
analyze controller behavior 0 G G G G 1
M P V C
= +

C
h
a
p
t
e
r

1
1

Rule of Thumb:
Closed-loop response becomes less oscillatory and more stable by
decreasing K
c
or increasing t
I
.
General Stability Criterion
Consider the characteristic equation,
Note that the left-hand side is merely the denominator of the
closed-loop transfer function.
The roots (poles) of the characteristic equation (s - p
i
) determine
the type of response that occurs:
Complex roots oscillatory response
All real roots no oscillations
***All roots in left half of complex plane = stable system
0 G G G G 1
M P V C
= +
C
h
a
p
t
e
r

1
1

C
h
a
p
t
e
r

1
1

Figure 11.25 Stability regions in the complex plane for roots of the
characteristic equation.
Stability Considerations
Feedback control can result in oscillatory or even
unstable closed-loop responses.
Typical behavior (for different values of controller
gain, K
c
).
C
h
a
p
t
e
r

1
1

Roots of 1 + G
c
G
v
G
p
G
m

(Each test is for different
value of K
c
)
C
h
a
p
t
e
r

1
1

(Note complex roots
always occur in pairs)
Figure 11.26 Contributions of characteristic equation roots to closed-loop response.
C
h
a
p
t
e
r

1
1

) 3 )( 2 )( 1 (
2
) (
+ + +
=
s s s
K
s G
C
OL
Routh Stability Criterion
Characteristic equation
Where a
n
>0 . According to the Routh criterion, if any of
the coefficients a
0
, a
1
, , a
n-1
are negative or zero, then
at least one root of the characteristic equation lies in the
RHP, and thus the system is unstable. On the other
hand, if all of the coefficients are positive, then one
must construct the Routh Array shown below:
(11-93)
0
0 1
1
1
= + + + +

a s a s a s a
n
n
n
n

C
h
a
p
t
e
r

1
1

For stability, all elements in the first column must be
positive.
C
h
a
p
t
e
r

1
1

The first two rows of the Routh Array are comprised of the
coefficients in the characteristic equation. The elements in the
remaining rows are calculated from coefficients by using the
formulas:
(n+1 rows must be constructed; n = order of the characteristic eqn.)
n-1 n-2 n n-3
1
n-1
a a a a
b
a

=
1 n
5 n n 4 n 1 n
2
a
a a a a
b

=
1
2 1 n 3 n 1
1
b
b a a b
c


=
1
3 1 n 5 n 1
2
b
b a a b
c


=
.
.
(11-94)
(11-95)
(11-96)
(11-97)
C
h
a
p
t
e
r

1
1

The important constraint is K
c
<8. Any K
c
>8 will cause instability.
Application of the Routh Array:
Characteristic Eqn is
We want to know what value of K
c
causes instability, I.e., at least
one root of the above equation is positive. Using the Routh array,
Conditions for Stability
C C M V
3
L P
K G 1 G G
) 2 s (
8
G G = = =
+
= =
0 1 = +
M P V C
G G G G
0 K 8 8 s 12 s 6 s
0 K 8 ) 2 s ( 0
) 2 s (
K 8
1
C
2 3
C
3
3
C
= + + + +
= + + =
+
+
( )( )
0 K 8 8
0
6
K 8 8 1 ) 12 ( 6
K 8 8 6
3 n 12 1
C
C
C
+
+
+
=
( )
1 K 0 K 8 8
8 K 0 K 8 8 72
C C
C C
> > +
< > +
C
h
a
p
t
e
r

1
1

Figure 11.29
Flowchart for
performing a
stability analysis.
C
h
a
p
t
e
r

1
1


1. Bode Stability Criterion
Ch. 14 - can handle time delays

2. Nyquist Stability Criterion
Ch. 14
Additional Stability Criteria
C
h
a
p
t
e
r

1
1

Direct Substitution Method
Imaginary axis is the dividing line between stable and unstable systems.
1. Substitute s = je into characteristic equation

2. Solve for K
cm
and e
c

(a) one equation for real part
(b) one equation for imaginary part

Example (cf. Example 11.11)

characteristic equation: 1 + 5s + 2K
c
e
-s
= 0 (11-101)

set s = je 1 + 5je + 2K
c
e
-je
= 0

1 + 5je + 2K
c
(cos(e) j sin(e)) = 0
C
h
a
p
t
e
r

1
1

Direct Substitution Method (continued)
Re: 1 + 2K
c
cos e = 0 (1)

Im: 5e 2K
c
sin e = 0 (2)


solve for K
c
in (1) and substitute into (2):
e
e e e
e
+ = + =
sin
5 5 tan 0
cos
Solve for e: e
c
= 1.69 rad/min (96.87/min)
from (1) K
cm
= 4.25
(vs. 5.5 using Pade approximation in Example 11.11)
C
h
a
p
t
e
r

1
1

Previous chapter Next chapter
C
h
a
p
t
e
r

1
1