Scribd
Upload
61 views3 pages

Benchmark Systems For PID Control

This document presents 11 benchmark systems that are useful for testing PID controllers. The systems range from simple models with poles and zeros to more complex models representing phenomena like heat conduction. Systems 1-5 are basic models well-suited for parametric studies, while systems 6-10 illustrate more challenging control problems. System 11 adds integrators to previous models. References provide more details on PID controller design and tuning for these benchmark systems.

Uploaded by

Marlon José Do Carmo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
61 views3 pages

Benchmark Systems For PID Control

This document presents 11 benchmark systems that are useful for testing PID controllers. The systems range from simple models with poles and zeros to more complex models representing phenomena like heat conduction. Systems 1-5 are basic models well-suited for parametric studies, while systems 6-10 illustrate more challenging control problems. System 11 adds integrators to previous models. References provide more details on PID controller design and tuning for these benchmark systems.

Uploaded by

Marlon José Do Carmo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Benchmark Systems for PID Control

K. J. Åström and T. Hägglund


Department of Automatic Control
Lund University, Lund, Sweden
Email: [email protected]
Fax: +46 46 13 81 18

Introduction
This is a collection of systems that are useful for testing PID controllers. The material
is based on many di erent sources. The systems 1-5 are standard systems that
are well suited to parametric studies. Their properties can easily be changed by
varying a parameter. The systems 6-10 are more specialized. They illustrate systems
with various diculties of control. PID control is not well suited for all of them.
Parameters for PID controllers for many of these systems are found in Åström and
Hägglund (1995) and Åström et al. (1998).

1. System with Multiple Equal Poles


Transfer function
G(s) =
1 n = 1; 2; 3; 4; 8 (1)
(s + 1)n
These systems are very common. For n = 1 and 2 anything can be achieved by PI
or PID control respectively. For large values of n the system behaves like systems
with long dead times. The systems have been used by controller manufacturers as
test cases for a long time. We have obtained this information from Eurotherm and
Foxboro.

2. Fourth Order System


Transfer function
G(s) =
1 = 0:1; 0:2; 0:5; 1:0 (2)
(s + 1)(1 + s)(1 + 2 s)(1 + 3 s)
This system has four poles whose spacing is determined by parameter . For small
values there are drastic improvements when going from PI to PID control. For = 1
the system is identical to the system (1) for n=4.

1
3. System with Right Half Plane Zero
Transfer function
G(s) =
1 , s
(s + 1)3 = 0:1; 0:2; 0:5; 1; 2; 5 (3)
This system has three equal poles and a right half plane zero. The achievable per-
formance is determined by parameter . The diculty of control increases with
increasing .

4. First Order System with Dead Time


Transfer function
G(s) =
1 e,s T = 0; 0:1; 0:2; 0:5; 2; 5; 10 (4)
1 + sT
This is the classic system which has been used in many investigations of PID control.
The system reduces to a pure time delay for T = 0 and represents lag dominated
systems for large T . Many of the early tuning rules were derived based on this model.
A drawback with the model is that it has slow roll-o at high frequencies.

5. Second Order System with Dead Time


Transfer function
G(s) =
1 e,s T = 0; 0:1; 0:2; 0:5; 2; 5; 10 (5)
(1 + sT )2
This system is similar to (4) but it has more high frequency roll o . The system
reduces to a pure time delay for T = 0.

6. Heat Conduction
Transfer function
ps
G(s) = e, (6)
This system represents the dynamics of one dimensional heat conduction. The ul-
timate gain is ku = e . Analog implementations of this system has been used by
Eurotherm to test temperature controllers.

7. Fast and Slow Modes


Transfer function
G(s) =
100  1 + 0:5  (7)
(s + 10)2 s + 1 s + 0:05
The essential dynamics of this system has a fast time constant T = 1 with a moderate
gain (1) and a slow time constant T = 20 with a large gain (10). Simple tuning rules
based on the step response do normally not give good tuning for systems of this
type because it is dicult to get a good estimate of the gain and the time constant.

2
8. Conditionally Stable System
Transfer function
G(s) =
(s + 6)2 (8)
s(s + 1)2(s + 36)
This system is conditionally stable. The stability region under PI control consists of
two disjoint sets.

9. Oscillatory System
Transfer function
!02
G(s) =
(s + 1)(s + 2!0 s + !02) ;  = 0:1; !0 = 1; 2; 5; 10:
2
(9)

Systems of this type with small damping  are not good candidates for PID control.
The system is easy to control if !0 is large. The performance can often be improved
drastically by more general controller structures.

10. An Unstable System


Transfer function
G(s) =
1 (10)
s2 ,1
This is a simple model of an inverted pendulum.

11. Systems with Integral Action


It is very useful to also have systems with integral action. A good collection is obtaind
by adding an integrator to the systems 1-5.

References
Åström, K. J. and T. Hägglund (1995): PID Controllers: Theory, Design, and Tuning.
Instrument Society of America, Research Triangle Park, North Carolina.
Åström, K. J., H. Panagopoulos, and T. Hägglund (1998): \Design of PI controllers
based on non-convex optimization." Automatica, 35:5.

You might also like