Chapter 11: Dynamic Behavior & Stability of Closed-loop Control System.

Block Diagram

Next, we develop a transfer function for each of the five elements in the feedback control
loop. For the sake of simplicity, flow rate w1 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:

) = 
K1  2
K
X ′ (s ′X s1+( ) ′ W s2 ( ) (11-1)
 τ s +1
  +
τ s
 1

Where

Vρ w1 −1 x
τ= , K1 = , and K2 = (11-2)
w w w
The symbol x% ′ ( t ) denotes the internal set-point composition expressed as an equivalent
sp
electrical current signal. x%′ ( t ) is related to the actual composition set point x%
sp ′ ( t ) by the
sp
composition sensor-transmitter gain Km:

′ ( t ) = K m xsp
x%
sp ′ ( t) (11-7)

Current-to-Pressure (I/P) Transducer

The transducer transfer function merely consists of a steady-state gain KIP:

Pt′( s )
= K IP (11-9)
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
W2′ ( s ) Kv
= (11-10)
Pt′( s ) τ v s + 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 τm is small so it can be neglected.

X m′ ( s )
= Km
X ′( s )
Controller
Suppose that an electronic proportional plus integral controller is used.
P′ ( s )  1 
= K c 1 +  (11-4)
E ( s)  τI s 

where P′ ( s ) and E(s) are the Laplace transforms of the controller output p′ ( t ) and the error
signal e(t). Kc is dimensionless.
Summer

Comparator

Block

∴Y(s) =G(s)X(s)

• Blocks in Series
“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 Ysp ≠ 0 and D = 0 (set-point change while disturbance
change is zero)

Y ( s) K M GC GV GP
∴ =
Ysp ( s ) 1 + GC GV GP GM

• Disturbance Changes (“Regulator Problem”)

Assume D ≠ 0 and Ysp = 0 (constant set-point)

Y (s) Gd *Note same denominator for Y/D, Y/Ysp.

∴ =
D ( s ) 1 + GC GV GP GM

EXAMPLE 1: P.I. control of liquid level

q1
q2

hm
LT LC I/P

q3

process d (s )

Gd(s)

y sp (s) ε (s ) c (s ) m (s ) + y (s)
+
−
Gc(s) Gf(s) G p(s) +

y m (s )
Gm(s)
Assumptions

1. q1, varies with time; q2 is constant.

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

3. q3 ≠f(h) (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).

For these assumptions, the transfer functions are:

 1 
GC ( s) = −K C 1 +  GM ( s ) = K M
 τI s 
GV ( s ) = K V
1
GP (s) = −
As
1
GL ( s) = KC > 0
As

The closed-loop transfer function is:

Y H′ GL
= =
D Q1′ 1 + GC GV GP GM

Substitute,
1
Y As
=
D  1   1 
1 − K C 1 + KV  − K M
 τ I s   As 

Simplify,
Y τIs
=
D Aτ I s + K C K V K M τ I s + K C K P K M
2
Characteristic Equation:

Aτ I s 2 + K C K V K M τ I s + K C K P K M = 0

Recall the standard 2nd Order Transfer Function:

K
G (s) =
τ 2 s 2 + 2ζ τs +1

To place Eqn. (4) in the same form as the denominator of the

T.F. in Eqn. (5), divide by Kc, KV, KM :
Aτ I
s 2 +τ I s +1 = 0
K C KV K M

Comparing coefficients (5) and (6) gives:

Aτ I Aτ I
τ2 = ⇒ τ=
K CK VK M K CK VK M
τI
2ζ τ= τ I ⇒ ζ=
2τ

Substitute,

1 K C K V K M τI
ζ=
2 A

For 0 < ζ < 1 , closed-loop response is oscillatory. Thus

decreased degree of oscillation by increasing Kc or τ I (for constant
Kv, KM, and A).
• unusual property of PI control of integrating system
• better to use P only
Stability of Closed-Loop Control Systems
process d (s )

Gd(s)

y sp (s) ε (s ) c (s ) m (s ) + y (s)
+
−
Gc(s) Gf(s) G p(s) +

y m (s )
Gm(s)

Proportional Control of First-Order Process

Set-point change:

K C KV K P K M
Y τs + 1
=
Ysp K C KV K P K M
1+
τs + 1
Y K1
=
Ysp τ 1s + 1
K OL τ
K1 = τ1 = K OL = K C K V K P K M
1 + K OL 1 + K OL

Set-point change = M

(
y (t ) = K1M 1 − e −t τ1 )
M
Offset = ysp ( ∞ ) − y ( ∞ ) =
1 + K OL

Closed-Loop Transfer function approach:

KK C
Y KK C 1 + KK C
= =
Ysp τs + 1 + KK C τ
s +1
1 + KK C
First-order behavior τ
=
closed-loop time constant 1 + KK C

(faster, depends on Kc)

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.
 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).

8
Y (s) = D( s)
s + 6 s + 12 s + 8 + 8 K C
3 2
D (s) is unspecified

8
GV = 1 GM = 1 GP = Gd =
( s + 2)
3

Kc is the controller function, i.e., GC ( s ) = K C

Let Γ( s ) = s 3 + 6 s 2 +12 s + 8 + 8 K C

If Kc = 1,

( ) ( )(
Γ( s ) = ( s + 4 ) s 2 + 2 s + 4 = ( s + 4 ) s +1 + 3 j s +1 − 3 j )
 1 
 = e −at 
Since all of the factors are positive,  s + a ,
the step response will be the sum of negative exponentials,
but will exhibit oscillation.

If Kc = 8,

Γ( s ) = s 3 + 6 s 2 + 12 s + 72 = ( s + 6)( s 2 + 12)

Corresponds to sine wave (undamped), so this case is

marginally stable.

If Kc = 27

Γ( s ) = s 3 + 6s 2 +12 s + 224 = ( s + 8) ( s 2 − 2 s + 28 )
 (
= ( s +8) s −1 +3 3 j s −1 −3 3 j )( )
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 )

8 1
Y ( s) = ⋅
s + 6 s +12 s + 8 + 8 K C s
3 2

8 1
y (t → ∞) = lim sY ( s ) = =
s →0 8 + 8K C 1 + K C

Therefore, if Kc 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 Kc
create instability.

Integral Control:
KC KC KC
∫ e( t ′) dt ′
t
P= P( s) = E ( s) GC ( s) =
τI 0 τIs τIs

For a unit step load-change and Kc=1,

8s 1
Y ( s) = ⋅ lim sY ( s) = 0 = y (∞ ) no offset
s( s + 2 ) + 8
3
s s→ 0
τI
(note 4th order polynomial)

PI Control:

 1 
GC ( s ) = K C 1 + 
 τIs
8s 1
Y (s) = ⋅ lim sY ( s ) = 0 no offset
8K
+ 8K C s s
s →0
s ( s + 2) 3 + C
τI

adjust Kc and τ I to obtain satisfactory response (roots of equation which is 4th order).
PID Control: (pure PID)

 1 
1 + τ s +τ D s 
GC ( s ) = K C  
 I 

No offset, adjust Kc, τ I , τ D to obtain satisfactory result (requires solving for roots
of 4th order characteristic equation).
Analysis of roots of characteristic equation is one way to
analyze controller

Rule of Thumb:
Closed-loop response becomes less oscillatory and more stable by
decreasing Kc or increasing tI .

General Stability Criterion

Consider the “characteristic equation,”
1 +GC GV G P G M = 0 Note that the left-hand side is merely the
denominator of the closed-loop transfer
function.

The roots (poles) of the characteristic equation (s - pi) 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
Stability Considerations

• Feedback control can result in oscillatory or even unstable closed-loop responses.

• Typical behavior (for different values of controller gain, Kc).
•
 2K C
GOL ( s ) =
( s +1)( s + 2)( s + 3)

Routh Stability Criterion

Characteristic equation

an s n + an −1s n −1 + + a1s + a0 = 0

Where an >0 . According to the Routh criterion, if any of the coefficients a0, a1, …, an-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:

For stability, all elements in the first column must be positive.

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:

a n-1a n-2 − a n a n-3 b1a n −3 − a n −1b2

b1 = c1 =
a n-1 b1

b1a n −5 − a n −1b 3
a n −1a n −4 − a n a n −5 c2 =
b2 = b1
a n −1

(n+1 rows must be constructed; n = order of the characteristic eqn.)

Application of the Routh Array:

8
GP = GL = GV = G M = 1 GC = K C
( s + 2) 3

Characteristic Eqn is 1 +GC GV GP GM = 0

8K C
1+ =0 (s + 2) 3 + 8K C = 0
(s + 2) 3
s 3 + 6s 2 + 12s + 8 + 8K C = 0

We want to know what value of Kc causes instability, I.e., at least

one root of the above equation is positive. Using the Routh array,

1 12 n=3
6 8 + 8K C
6(12) − (1)( 8 + 8K C )
0
6
8 + 8K C 0
Conditions for Stability

72 − ( 8 + 8 K C ) > 0 KC < 8
8 + 8K C > 0 K C > −1

The important constraint is Kc<8. Any Kc ≥ 8 will cause instability.

Flowchart for performing a stability analysis.