Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
ARROW@DIT
Conference papers
2004-01-01
This Conference Paper is brought to you for free and open access by the
School of Electrical and Electronic Engineering at ARROW@DIT. It has
been accepted for inclusion in Conference papers by an authorized
administrator of ARROW@DIT. For more information, please contact
yvonne.desmond@dit.ie, arrow.admin@dit.ie.
This work is licensed under a Creative Commons AttributionNoncommercial-Share Alike 3.0 License
II
1
G c ( s) = Kc 1 +
+ Td s
Ti s
1 1 + sTd
G c ( s) = K c 1 +
T
Ti s
1+ s d
N
This architecture is used, for example, on the
Honeywell TDC3000 Process Manager Type A,
interactive mode product [15].
3. The non-interacting controller based on the two
degree of freedom structure:
1
Tds
E(s )
U( s) = K c 1 +
+
Ti s 1 + Td s
Tds
R( s)
Kc +
Td
1+ s
N
This architecture is used, for example, on the Omron
E5CK digital controller with = 1 and N=3 [15].
The most dominant PI controller architecture is the
1
ideal PI controller: G c ( s) = Kc 1 +
Ts
i
The wide variety of controller architectures is
mirrored by the wide variety of ways in which
processes with time delay may be modeled.
Common models are:
plus
delay
(IPD)
model:
s m
K e
G m ( s) = m
s
3. First order lag plus integral plus delay (FOLIPD)
Km e s m
model: G m (s) =
s(1 + sTm )
4. Second order system plus time delay (SOSPD)
model:
K m e s m
G m (s) =
or
2
Tm1 s2 + 2 m Tm1s + 1
K m e s
(1 + Tm1s )(1 + Tm 2 s )
It has been shown that 89% of the PI controller
tuning rules have been defined for the ideal PI
controller structure, with 47% of tuning rules based
on a FOLPD process model. The range of PID
controller variations has lead to a less homogenous
situation than for the PI controller; 44% of tuning
rules have been defined for the ideal PID controller
structure, with 37% of PID tuning rules based on a
FOLPD process model [14].
Of course, the modeling strategy used will
influence the value of the model parameters, which
will in turn affect the controller values determined
from the tuning rules. Twenty-six modeling
strategies have been proposed to determine the
parameters of the FOLPD process model, for
example. Space does not permit a full discussion of
this issue; further details are provided by ODwyer
[14].
G m ( s) =
III
IV
CONCLUSIONS
REFERENCES
[1] K.J. strm and T. Hgglund. PID Controllers:
Theory, Design and Tuning. Instrument Society
of America, Research Triangle Park, North
Carolina, 2nd Edition, 1995.
[2] H.N. Koivo and J.T. Tanttu. Tuning of PID
Controllers: Survey of SISO and MIMO
techniques. Proc. IFAC Intelligent Tuning and
Adaptive
Control
Symposium,
75-80,
Singapore, 1991.
[3] W.L. Bialkowski in The Control Handbook , ed.
W.S. Levine. CRC/IEEE Press, Boca Raton,
Florida, 1996, 1219-1242.
[4] W.L. Luyben and M.L. Luyben. Essentials of
process control. McGraw-Hill International
Edition, 1997.
[5] M.A. Hersh and M.A. Johnson. A study of
advanced control systems in the workplace.
Control Eng. Practice, 5(6):771-778, 1997.
[6] H. Takatsu and T. Itoh. Future needs for control
theory in industry report of the control
technology survey of Japanese industry. IEEE
Trans. Control Syst. Tech., 7(3):298-305, 1999.
[7] A. ODwyer. PID compensation of time delayed
processes: a full survey, Technical Report
AOD-00-05, Version 3. School of Control
Systems, Dublin Institute of Technology,
Ireland (January 2004).
[8] D.B. Ender. Control Engineering. 180-190,
September 1993.
[9] Protuner U.K. Ltd. The state of art in control.
Sales literature, 1997.
[10] Universal Dynamic Technologies. Brainwave:
the new concept in process control. Sales
literature, 1998.
[11] Entech Control Engineering Ltd. Competency in
Process Control - Industry Guidelines. Version
1.0, 3/94, 1994
[12] C.-C. Yu. Autotuning of PID controllers.
Advances in Industrial Control Series,
Springer-Verlag London Ltd., 1999.
[13] P. Van Overschee and B. De Moor. RaPID: the
end of heuristic PID tuning. Preprints Proc.
PID 00: IFAC Workshop on digital control,
687-692, Terrassa, Spain, 2000.
[14] A. ODwyer. Handbook of PI and PID
controller tuning rules. Imperial College Press,
London, 2003.
[15] ISMC. RAPID: Robust Advanced PID Control
Manual. Intelligent System Modeling and
Control nv, Belgium, 1999.
[16] K.J. strm and T. Hgglund. Benchmark
systems for PID control. Preprints Proc. PID
00: IFAC Workshop on digital control, 181182, Terrassa, Spain, 2000.
K me s
1
; Ideal PI controller G c ( s) = Kc 1 +
1 + sTm
Ts
i
m
Kc
Ti
Time domain criteria
min (Tm ,8 m )
Direct synthesis
0. 5Tm
Km m
Skogestad [17]
1
Gorez [18]
Model parameters assumed
known
Tm
Tm
K m m
Kc
(1 v) m + Tm
(1) 2
Skogestad [17]
m
1
m + Tm
K me s m
1
; Ideal PI controller G c ( s) = Kc 1 +
s
Ts
Kc
Direct synthesis
Chidambaram and Sree [19]
Comment
Comment
Ti
Time domain criteria
1. 1111
Km m
4.5 m
0.5
Km m
8 m
Kc
Servo or regulator
tuning minimum
IAE Huang and
Jeng [20]
Model parameters
identified using a
relay feedback method
Ti
Td
Other tuning: performance index minimisation
0. 36 + 0. 76
Tm
m
0.47 m + Tm
0. 47 Tm m
0.47 m + Tm
Km
Kc
(2 ) 3
Ti
1.5M s 2 m
1
+
M s (M s 1) 0.32 M s 2 (M s 1) m + Tm
2
K m e s
1
; ideal PID controller G c ( s) = Kc 1 +
+ Td s
1 + sTm
T
s
(2 )
Td
5 m
3
( + T ) M
m
m
s
( 2)
Comment
m
< 0.33
Tm
m
0.33
Tm
2. 5 m
0.5
1
M s (M s 1) ( m + Tm ) M s
Kc
(1)
Kc
(2 )
(1 v) m + Tm
K m m
2
2
0.4 M s
0.4 M s
0.1M s
+ Tm
m + Tm
with v = 1 0.5 m
, = M 1 7. 5
1
0
.
4
M
1
0
.
4
M
s
s
s
0.8194 Tm + 0. 2773 m
Km m
0.9738
Tm
0.0262
, Ti
(2 )
= 1.0297Tm + 0.3484 m , Td
(2 )
0. 4575 Tm + 0.0302 m
=
1.0297 Tm + 0.3484 m
2 M s 2 1
Kc
Chidambaram and
Sree [19]
1.2346
K m m
2
K m ( + 0.5 m )
Km e s m
1
; ideal PID controller Gc ( s) = K c 1 +
+ Td s
s
Ti s
Td
Comment
0.45 m
Model parameters
assumed known
+ 0.25 m
m
2 + m
Model method
not specified
Ti
Direct synthesis
4.5 m
Robust
2 + m
Km e s m
; non-interacting controller based on the two degree of freedom
s
1
T
s
T
s
d
d
E( s) K c +
R (s)
structure U (s) = K c 1 +
+
Tis
Td
Td
1+
s
1+
s
N
N
Comment
Kc
Ti
Td
Rule
Direct synthesis
Chidambaram and
Sree [19]
Model parameters
assumed known
1.2346
K m m
Kc
Minimum ISE
Arvanitis et al. [22]
1.4394
K m m
Minimum
performance index
Arvanitis et al. [22]
5
Minimum
performance index
Arvanitis et al. [22]
4.5 m
0.45 m
N = 0; = 0 ;
= 0. 6
Km e s m
K
controller U(s ) = c E(s ) K c (1 + Td s)Y( s) .
Ti s
s
Comment
Ti
Td
Regulator tuning: Minimum performance index
Model parameters
assumed known
2.4569 m
0.3982 m
1.2986
K m m
3.2616 m
0.4234 m
Model parameters
assumed known
1.1259
K m m
6.7092 m
0.4627 m
Model parameters
assumed known
Performance index =
[e
( t) + K m u 2 ( t ) dt
2
Performance index =
2
2
2 du
e ( t) + K m dt
dt
[21] B.W. Bequette. Process Control: Modeling, Design and Simulation. Prentice-Hall International Series in
the Physical and Chemical Sciences, Upper Saddle River, New Jersey 07458, USA, 2003.
[22] K.G. Arvanitis, G. Syrkos, I.Z. Stellas and N.A. Sigrimis. Controller tuning for integrating processes with
time delay. Part I: IPDT processes and the pseudo-derivative feedback control configuration. Proc. 11th
Mediterranean Conf. On Control and Automation, Rodos, Greece, Paper T7-040, 2003.
e 2 ( t) dt