You are on page 1of 4

2008 International Conference on Advanced Computer Theory and Engineering

Design and Research Based on Fuzzy PID-parameters Self-tuning


Controller with MATLAB

Wang Xiao-kan1 Sun Zhong-liang1 Wanglei1 Feng Dong-qing2


(1 Henan Mechanical and Electrical School, Zheng zhou, Henan, 450002 2 Institute
of Information and Control, Zheng zhou University 450001)

Control takes error "e" and Change-in-error


"ec" as the input of Fuzzy PID controller,
meets the request of the different moments of
"e" and "ec" to PID parameters self-tuning.
Using fuzzy control rules on-line, PID
parameters "kp", "ki", "kd" are amended,
which constitute a self-tuning fuzzy PID
controller, the principle of which control
programme as shown in Figure 1[1,3].

Abstract
Designed
a
parameter
self-tuning
PID-controller based on fuzzy control, which
can adjust PID-parameters according to error
and change-in-error. A simulation study of the
system shows that the controller possesses
both the precise character of PID controller
and the flexible advantage of fuzzy controller.
It can eliminate the steady error of the system
promptly.
Key words: fuzzy control; parameter
self-tuning; fuzzy PID ;MATLAB
1. The self-tuning control principle of fuzzy
PID parameter
PID control requirements model structure
very precise, and in practical applications, to
different extent, most of industrial processes
exist to the nonlinear, the variability of
parameters and the uncertainty of model, thus
using conventional PID control can not
achieve the precise control of the process. But
the dependence on the mathematical model of
the fuzzy control is weak, so it isn't necessary
to establish the precise mathematical model of
the process, and the fuzzy control has a good
robustness and adaptability. Accoding to their
own characteristics, we combinate fuzzy
control with PID control, and provide a based
on fuzzy PID parameters self-tuning controller
with MATLAB.
Fuzzy PID parameters Self-tuning
978-0-7695-3489-3/08 $25.00 2008 IEEE
DOI 10.1109/ICACTE.2008.9

Figure 1 The structure of self-tuning fuzzy


PID controller
PID parameters fuzzy self-tuning is to
find the fuzzy relationship between the three
parameters of PID and "e" and "ec", and
according to the principle of fuzzy control, to
modify the three parameters in order to meet
different requirements for control parameters
when "e" and "ec" are different, and to make
the control object a good dynamic and static
performance. selecting the language variable
values of error "e" and the error rate of change
"ec" is (NB, NM, NS, ZO, PS, PM, PB) seven
fuzzy values, so is selecting the language
996

variable values of kp, ki, kd. And then setting


up the suitable fuzzy control table for"kp",
According to the fuzzy rules table, to
select the appropriate vague and ambiguous
methods, and to make dynamic tuning for
"kp" , "ki" , "kd". Given "kp", "ki", "kd" is the
pre-tuning value of conventional tuning , the
formula as following[2,3]:
Table 1 the fuzzy rules table of kp

"ki",
"kd"
three
parameters
tuning
separately[3].
self-tuning through processing of results,
look-up table, calculation of the fuzzy logic
rules.
2. The analysis of system simulation
Assume that the controlled object[2] is

G p ( s) =

500000
s + 80 s 2 + 15000 s
3

Sampling time is 1 ms, using


conventional PID control and fuzzy PID
control to simulate respectively, and
comparing their results.
2.1 The MATLAB / Simulink PID
simulation without disturbance
Using conventional PID controller to
control, through the SIMULINK MATLAB
simulation, debugging and setting the three
parameters "kp" , "ki" , "kd" of the PID
controller , their values are as follows:

Table 2 the fuzzy rules table of ki

k p = 0.4, ki = 1, k d = 0.005
the output curve of step-response is shown in
Figure 2.
Using MATLAB onboard FUZZY
toolbox, we create two input and three output
fuzzy controller, and determine membership
functions and fuzzy rules, and then we can
create fuzzy self-tuning PID simulation map
in SIMULINK. The self-adaptive-adjusting
step response output curve by using MATLAB
programming is shown in Figure 3.

Table 3 the fuzzy rules table of kd

k p = k p + k p
'

ki = ki + ki
'

k d = k d + k d
'

Figure 2 The Conventional PID response


curve (ithout disturbance)

In the running process online, the control


system completes PID parameters' online

997

output response curve, as shown in Figure 6


and Figure 7.

Figure 3 The Fuzzy PID response


curve (without disturbance)
2.2 The simulation result when t=0.35s and
adding the pulse amplitude of 1.5 for
interference
When t=0.35s and adding the pulse
amplitude of 1.5 for interference[2,3], we can
use matlab / simulink to analyze the changes
of the two step output curve, as shown in
Figure 4 and Figure 5.

Figure 5

the fuzzy PID response


curve (with disturbance)
2.3 The simulation result when the
controlled object is changed
Assume that the controlled object:

G p ( s) =

70000
s 3 + 30 s 2 + 100000 s

debugging and setting the three


parameters "kp" , "ki" , "kd" of the PID
controller , their values are as follows:

k p = 0.6, ki = 0.5, k d = 0.001


We have the conventional PID step output
response curve and the fuzzy PID control step
output response curve, as shown in Figure 6
and Figure 7.

Figure 4 the Conventional PID response


curve (with disturbance)
2.3 The simulation result when the
controlled object is changed
Assume that the controlled object:

G p ( s) =

70000
s + 30 s 2 + 100000 s
3

debugging and setting the three


parameters "kp" , "ki" , "kd" of the PID
controller , their values are as follows:

Figure 6 The conventional PID response


curve when the controlled object is changed

k p = 0.6, ki = 0.5, k d = 0.001


We have the conventional PID step output
response curve and the fuzzy PID control step

998

control has greater volatility, the high


overshoot and long adjust time, and the fuzzy
PID control has the small overshoot, shorter
adjust time, strong anti-interference capability
and strong robust.
3. Conclusion
This paper introduces a design method of
fuzzy self-tuning PID controller and make use
of MATLAB fuzzy toolbox to design fuzzy
controller, organically combine fuzzy PID
controller with SIMULINK, and achieve the
design and simulation of PID parameters
self-adjusting fuzzy control system. The
simulation results show that: compared with
the traditional PID controller, fuzzy
self-tuning PID controller has a better
dynamic response curve, short response time,
small overshoot, high steady precision, good
static and dynamic performance.
References:
[1] Li Shiyong, Fuzzy Control. Neural
Networks and Intelligent Control Theory[M].
Harbin: Harbin Institute of Technology Press,
2003.
[2] Liu Jinkun, Advanced PID Control
and Its MATLAB Simulation[M]. Electronics
Industry Publishing House, 2003.
[3] Liu Zengliang, Fuzzy Control
Technology and Application Selected[M].
Beijing: Beijing University of Aeronautics and
Astronautics Press, 2002.
Project Fund: Henan Natural Science
Fund(0511010800).
Address: on the 2ndAgricultural Road in
Zhengzhou City of Henan Province, Henan
Mechanical and Electrical Vocational School
Mechanical and Electrical Department Wang
Xiaokan(450002).
E-mail: wxkbbg@163.com
Tel: 13633806151

Figure 7 The fuzzy PID response curve


when the controlled object is changed
2.3 Comparison of three simulation results
We have the comparative results of their
respective performance indicators through
several over methods to simulate, as shown in
table 4.

Analyzing the various performance


indicators of table 4, we have the following
conclusions:
(1) Fuzzy self-tuning PID parameter
control has smaller overshoot and shorter
adjust time than conventional PID control,and
has better dynamic response properties and
steady-state properties;
(2) The three parameters "kp", "ki", "kd"
of conventional PID control need to be
constantly online to adjust in order to achieve
better control performance; Fuzzy self-tuning
PID parameters controller can automatically
adjust PID parameters in accordance with the
error and the rate of error-in-change in the
status of the change process ,so it has better
self-adaptive capacity;
(3) When the object controlled is changed
or joining disturbance, conventional PID

999

You might also like