0% found this document useful (0 votes)
202 views5 pages

Tugas 4 (Inverse Response Chapter 17)

The document discusses control strategies for an inverse-response system model. It provides the following key details: 1) Under pure proportional control, the stable range of the controller gain Kc is between -1/2 and 7/8. 2) With Kc = 0.5, the steady-state offset in response to a unit step change is zero as Kc is within the stable range. 3) Using an inverse-response compensator, the stable range of Kc for the overall system is between -1/2 and -7/2. 4) In the limit as Kc approaches infinity, the apparent controller is a PID controller with Kc as the proportional gain and
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
202 views5 pages

Tugas 4 (Inverse Response Chapter 17)

The document discusses control strategies for an inverse-response system model. It provides the following key details: 1) Under pure proportional control, the stable range of the controller gain Kc is between -1/2 and 7/8. 2) With Kc = 0.5, the steady-state offset in response to a unit step change is zero as Kc is within the stable range. 3) Using an inverse-response compensator, the stable range of Kc for the overall system is between -1/2 and -7/2. 4) In the limit as Kc approaches infinity, the apparent controller is a PID controller with Kc as the proportional gain and
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

Nama : Muhammad Jabber

NRP : 07111745000051
Mata Kuliah : Sistem Pengaturan Proses

17.3 It is required to control a certain inverse-response system whose transfer-function


model is given as :
2(1 - 4s )
g ( s) =
(2 s + 1)(5s + 1)
And several control strategies are to be investigated
(a) Under pure proportional control, what is the range of values the kc can take for the
closed-loop system to be stable ? (Assume there are no other elements in the
feedback loop, just the process and the feedback controller).
(b) When the system operates under pure proportional control with kc = 0.5, in
response to a unit step change in set-point, what will the steady-state offset in the
process output be ?
(c) Using an inverse-response compensator in aminor loop around a pure proportional
controller, if g '(s) , the transfer function in this minor loop, is chosen to be :
10s
g '( s ) =
(2 s + 1)(5s + 1)
Now find the range of values which Kc can take for the overall system to be stable.
(d) If the pure proportional controller and the inverse-response compensator in the
minor loop around it are combined into “apparent controller” block referred to as
gc *( s) , show that in the limit as Kc � �, this apparent controller, gc *( s) , in
t ,t
fact tends to a PID controller. What are th value of Kc, i D for this controller ?
(e) Implement the limiting PID controller derived in part (d) on the process model
given in Eq. (P17.5) and obtain the closed-loop system response to a unit step
change set point.

17.3 The dynamic behavior of abiological process which consist of two subprocesses in
opposition is represented by following transfer function :
3 3
g (s) = +
(7 s + 1) ( s + 1)( s + 3)
(a) Assume that the process is to be under proportional feedback control and obtain a
root locus diagram for the closed-loop system, and from it determine the range of
Kc values required for closed-loop stability. What is the frequency of oscillation
when the closed-loop system is on the verge of instability ?
(b) Use this information to design a PI controller for the process. Implement this
controller and obtain the closed-loop response to a unit step change in the set-point.
Briefly comment on the control system performance vis-a-vis the nature of the
process transfer function.
(c) For Extra Credit Only. Derive an inverse-response compensator for the class of
systems with two right-half plane zeros, whose transfer functions are of the form :
K (1 - h1s )(1 - h 2 s )
g (s ) =
(t 1s + 1)(t 2 s + 1)(t 3 s + 1)

17.8 Prove l = 2h , the inverse-response compensator is exactly the same as the Smith
predictor with a first-order Pade approximation.
Answer :
(a) Under pure proportional control, what is the range of values the kc can take for the
closed-loop system to be stable ? (Assume there are no other elements in the
feedback loop, just the process and the feedback controller).
Open-loop :
2(1 - 4 s)
g ( s) =
(2 s + 1)(5s + 1)
(2 - 8s )
g ( s) =
(2 s + 1)(5s + 1)
Closed-loop :
Kc(2 - 8s )
CLTF ( s) = 1 +
(2s + 1)(5s + 1)
(2 s + 1)(5s + 1) + Kc(2 - 8s)
=
(2 s + 1)(5s + 1)
(2 s + 1)(5s + 1) + (2 Kc - 8Kcs )
=
(2s + 1)(5s + 1)
(10s 2 + 2s + 5s + 1) + (2 Kc - 8 Kcs )
=
10 s 2 + 7 s + 1
10 s 2 + (7 - 8 Kc) s + (1 + 2 Kc)
=
10 s 2 + 7 s + 1

The characteristic equation for the closed-loop system is given by :


10 s 2 + (7 - 8 Kc) s + (1 + 2 Kc)
7 - 8 Kc = 0 1 + 2 Kc = 0
8 Kc = 7 2 Kc = -1
7 1
Kc = Kc = -
8 and 2
1 7
- < Kc <
The stable range of controller gain is 2 8 . Thus the system exhibits
conditional stability.

(b) When the system operates under pure proportional control with kc = 0.5, in
response to a unit step change in set-point, what will the steady-state offset in the
process output be ?

Simulink :

Gambar 1. Blok Diagram Simulink


Response :

Gambar 2. Respon System Using Kc=0,5

Dengan Kc = 0,5, respon keluaran dari sistem menjadi stabil karena nilai Kc masih
1 7
- < Kc <
memenuhi range 2 8 untuk batas kestabilan.

(c) Find the range of values which Kc can take for the overall system to be stable.
10 s
g '( s) =
(2 s + 1)(5s + 1)
Open-Loop :
g * ( s ) = g ( s ) + g '(s )
2(1 - 4 s ) 10 s
= +
(2s + 1)(5s + 1) (2 s + 1)(5s + 1)
(2 - 8s) + 10 s
=
(2s + 1)(5s + 1)
(2 + 2 s)
=
(2s + 1)(5s + 1)
Close-Loop :
CLTF = 1 + g * (s )
Kc(2 + 2 s)
= 1+
(2 s + 1)(5s + 1)
(2 s + 1)(5s + 1) Kc(2 + 2 s )
= +
(2 s + 1)(5s + 1) (2 s + 1)(5s + 1)
(2 s + 1)(5s + 1) + Kc(2 + 2 s)
=
(2 s + 1)(5s + 1)
10 s 2 + (7 + 2 Kc) s + (1 + 2 Kc)
=
(2 s + 1)(5s + 1)
The characteristic equation for the closed-loop system is given by :
10s 2 + (7 + 2 Kc) s + (1 + 2 Kc)
7 + 2 Kc = 0 1 + 2 Kc = 0
2 Kc = -7 2 Kc = -1
7 1
Kc = - Kc = -
2 and 2
7 1
- < Kc < -
The stable range of controller gain is 2 2 . Thus the system exhibits
conditional stability.

17.8 Prove l = 2h , the inverse-response compensator is exactly the same as the Smith
predictor with a first-order Pade approximation.

Pade approximation
a
1- s
-a s
e = 2
a
1+ s
2
a � a �-a s
1- s = � 1+ s � e
2 � 2 �

a
=h
2
1 - h s = ( 1 + h s ) e -2h s

Design Inverse-Response Compensator :


1
g '( s ) = ( 2h s )
( t1s + 1) ( t 2 s + 1)
2h s
g '( s ) =
( t1s + 1) ( t 2 s + 1)
Design Smith Predictor
g ( s) =
( 1 + h s ) e -2h s
( t 1s + 1) ( t 2 s + 1)
(
g m *( s ) = g ( s ) 1 - e
-2h s

)
( 1+hs)
( t 1s + 1) ( t 2 s + 1) ( )
-2h s
g m *( s ) = 1- e
-2h s -2h s
(1 + h s ) + (-e - h e s)
=
( t 1s + 1) ( t 2 s + 1)
-2h s -2h s
(1 - e ) + (h - h e ) s
=
( t 1s + 1) ( t 2 s + 1)
1 -h s 1 -h s
(1 - ) + (h - h )s
1 +h s 1 +h s
=
( t 1s + 1) ( t 2 s + 1)
�(1 + h s) - (1 - h s ) � � h + h 2 s h -h 2 s �
� 1 +h s � �1 + h s - 1 + h s �s
+
= � �� �
( t 1s + 1) ( t 2 s + 1)
�2h s � �2h 2 s �
� +� �s
�1 +h s � � � 1 +h s �
=
( t 1s + 1) ( t 2 s + 1)
�2h 2 s 2 + 2h s �
� �
1+hs �
= �
( t 1s + 1) ( t 2 s + 1)
� 1 +h s �
2h s �
� 1 +h s �

=
( 1 ) ( 2 1)
t s + 1 t s +
2h s
=
( t 1s + 1) ( t 2 s + 1)

Dengan pendekatan Pade, desain smith predictor sama dengan desain inverse-response
compensator.

You might also like