Scribd
Upload
1 views

[13].Simulation of Fuzzy Auto-tuning PID Control Based on MATLAB

This paper presents a fuzzy auto-tuning PID controller designed for CNC machine tools' servo feed systems, addressing the challenges of uncertain parameters and nonlinearities. The simulation results in MATLAB demonstrate that this controller enhances the dynamic performance and robustness of the servo feed system compared to conventional PID controllers. The fuzzy auto-tuning approach allows for real-time adjustment of PID parameters based on error and error rate, leading to improved stability and precision in control.

Uploaded by

thangtdhvlvc
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
1 views

[13].Simulation of Fuzzy Auto-tuning PID Control Based on MATLAB

This paper presents a fuzzy auto-tuning PID controller designed for CNC machine tools' servo feed systems, addressing the challenges of uncertain parameters and nonlinearities. The simulation results in MATLAB demonstrate that this controller enhances the dynamic performance and robustness of the servo feed system compared to conventional PID controllers. The fuzzy auto-tuning approach allows for real-time adjustment of PID parameters based on error and error rate, leading to improved stability and precision in control.

Uploaded by

thangtdhvlvc
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

Advanced Materials Research Vols 308-310 (2011) pp 579-583 Online: 2011-08-16

© (2011) Trans Tech Publications, Switzerland


doi:10.4028/www.scientific.net/AMR.308-310.579

Simulation of Fuzzy auto-tuning PID Control Based on MATLAB

JianwenChen1,a leili2 ,b JinchunSong3,c ZhanJunHe4,d


1,2,3,4
School of Mechanical engineering and Automation Northeastern University , Shenyang,
Liaoning, 110004, China
a b c d
[email protected] , [email protected] , [email protected] , [email protected]

Keywords: Fuzzy auto-tuning, PID control, Servo feed system, Simulink, Simulation.

Abstract. In CNC machine tools servo feed system, there are many uncertain parameters, it is
difficult to establish accurate mathematical model of the Servo feed system. The parameters of
conventional PID controller can’t adjust immediately to the controlled object. Fuzzy PID controller
[1]
, which has the advantages of conventional PID control and fuzzy control, can solve these problems.
We apply it in numerical control machine servo feed system. The simulation in MATLAB shows that
the fuzzy auto-tuning PID controller improve the dynamic performance of the servo feed system
and has better robustness.

Introduction
The position and velocity are the control objects of the servo system in Precision CNC machine tools,
CNC machine tools servo feed system include electrical transmission systems and mechanical
transmission system, electrical transmission systems and mechanical transmission system have many
parameters, such as: Drive parameters , Velocity feedback parameters, Friction coefficient,
Coefficient of friction, Stiffness of transmission and anti-hunting, these parameters are uncertain and
nonlinear. It is difficult to establish an accurate mathematical model of feed mechanism.
Although the conventional PID controller has some advantages such as: simple principle, easy to
implement and stable performance, It is mainly used for the controlled object who has the exact
model, The fuzzy control [2] is a promising branch of the field of modern control and has better
robustness, it is not sensitive to the parameters variation of the controlled object., This paper
combines the advantages of conventional PID control with fuzzy control concept, formed fuzzy
auto-tuning PID control method. Through the simulation analysis results indicate, Fuzzy auto-tuning
PID controller can adjust automatically the control parameters in process of system feed control, this
method solve the problem of parameter uncertainties, nonlinear problems in servo feed system of
precision CNC machine tools.

CNC machine tools servo control system modeling


The servo control system of CNC machine tools usually include servo motor, current and speed and
position control links and mechanical feed mechanism etc. According to control commands, the servo
system of CNC machine tools can perform automatically the speed, the direction and location of the
components, which is different from the general common machine feeding system, this paper
discusses the problem of auto-tuning PID parameter of servo system, the object is position servo
system of the permanent magnet synchronous motor, the component [2] is shown in Figure 1.

All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans
Tech Publications, www.ttp.net. (ID: 130.113.151.253, McMaster University, Hamilton, Canada-02/04/15,20:36:32)
580 Advanced Design Technology, ADME 2011

Fig.1 Component of servo system


Taking appropriately these links into consider while establishing servo system model, and
conducting shortcutting according to the concrete circumstance, In order to facilitate calculation, The
output shaft of motor connect directly with the ball screw, And current loop WACR(s) is simplified a
typical first-order system, Speed loop WASR(s) simplified one-order inertial link, the controlled object
simplified mathematical model diagram is shown in Figure 2.

Fig. 2 The dynamic structure diagram of servo system

The design of fuzzy auto tuning PID controller


The fuzzy control don’t require a precise mathematical model of controlled object, and just need to
take technical personnel of experience knowledge into control strategy, This paper shows that the
complex systems is controlled effectively and Fuzzy auto-tuning PID controller can stabilize control
system rapidly by the thought of fuzzy reasoning. Fuzzy auto tuning PID controller take error E and
error rate Ec for input to adjust parameters Kp , Ki and Kd of PID controller online, the PID controller
parameters meet the different requirements of the E and Ec in different time, and the controlled object
have a static and dynamic performance. The structure [2] is shown in Figure 3.

Fig .3 Structure of fuzzy auto tuning PID controller


Advanced Materials Research Vols. 308-310 581

Fuzzy auto-tuning PID controller discrete expressions [3] as follows:


k
u(k ) = K P (k ) ∗ e(k ) + K I (k ) ∗ ∑ e(i ) + K D (k ) * ec (k )
i =0

The final output of the control parameters is as follows：


K P = K P' + ∆K P , K I = K I' + ∆K I , K d = K d' + ∆K d . Among them: K P' , K I' , K d' is the initial value of
PID controller, ∆K P , ∆K I , ∆K d is the output value of the Fuzzy control reasoning device, K P , K I ,
K d is the ultimate control parameter.

Membership functions of each variable


The core algorithm of Fuzzy auto-tuning PID control is: Using Fuzzy language variable e, ec and
KP , KI, KD fuzzy subsets of membership functions ，and according to the fuzzy adjustment rules of
each parameters，the fuzzy controller adjust PID controller parameters online based on different e and
ec. Each fuzzy language variables have multiple linguistic values , Each fuzzy subsets can be
described as: Negative Big(NB),Negative Medium(NM),Negative Small(NS),Zero(Z),Positive S
mall(PS),Positive Medium(PM),Positive Big(PB). each language value can be described by one
membership function .This paper build membership function for each fuzzy linguistic variable,
These membership function is shown in Figure 4.

Fig.4 Membership function

The Fuzzy control rule table


According to the technical engineering knowledge and experience of staff, considering the system
stability, response time, overshoot, Steady precision and other aspects ， this paper build the
appropriate fuzzy reasoning rules to follow when setting parameters Kp, Ki, Kd. These rules [2],[4],[7]
are shown as fellows:
(1) When the e is larger , to speed up the response time and avoid differential saturation which
will leave the control action beyond the approved scope. Kp should be larger but Kd should be smaller,
at the same time, Ki should be zero in order to prevent the system response showed a larger overshoot
which caused by integral saturation.
(2)When the e and ec is Medium size ，in order to ensure the system response speed and to reduce
the overshoot of system response. Kp should be smaller .Ki should be Medium. Kd .important to
system .should be bigger.
(3)When the error is smaller, Kp and Ki should be larger to make the system have good stability,
meanwhile, In order to avoid output response oscillations in nearby the set point and to enhance the
anti-jamming capability, Kd should be a proper value, usually when ec is smaller, Kd can be
bigger.when ec is bigger, Kd should be smaller, According to the above rules, the fuzzy rule table of
Kp, Ki, Kd is shown in table:
582 Advanced Design Technology, ADME 2011

Table.1 Fuzzy rules of Kp, Ki, Kd

The Fuzzy self-tuning PID Controller simulation of the System


The object of study is Z feeding system of HTC3250μn，workbench quality m=1000kg，screw
palpitation L = 0.02 m，screw span l = 0.96m ，electromotor is made by Siemens , model number
1FT6105-8AF7，the rotation inertia of electromotor J m = 0.1487 kgm 2 ，the rotation inertia of load
J L = 0.042420031kgm2 ， the armature resistance of electromotor R = 1.4Ω ， the armature
inductance of electromotor Ls = 0.0135H ，the coefficient of Induction EMF K u = 0.04v ⋅ m / A ，
the coefficient of torque K L = 1.1( N ⋅m A) ，the axial rigidity of screw KB = 1.12 × 108 N m ，the
contact rigidity of screw nut KN = 2.02 × 108 N m ， According to the above parameters ， the
transfer function of system can be confirmed，considering the actual situation and referencing GA
genetic algorithm [5] , the order of system transfer function become lower, the transfer function can be
simplified as follows:
308.5
G (s) = 2
s ( s + 14.4 s + 40.5)
According to the Structure of fuzzy auto-tuning PID controller (Fig. 3), the system simulation
model[6] is designed in Simulink environment , which is shown in Fig. 5.

Fig. 5 The Simulink model of the control system


According to the actual debugging, factors of deviation quantization are K e =80, K ec=5.9, factors
of actual quantification are k1=0.7 ， k2=0.05 ， k3=0.78, The initial value of PID controller
parameters are Kp=2.1 ,Ki=0.5,Kd=0.3, after running Simulink, you can get both fuzzy auto-tuning
Advanced Materials Research Vols. 308-310 583

PID control `s and conventional PID control’s step response as the Fig. 6. The upside is the step
response of conventional PID control; the underside is the step response of fuzzy auto tuning PID
control, according to the comparison: the dynamic response curves of fuzzy auto-tuning PID control
is better, steadier and higher precision, smaller overshoot. It has better robustness.

Fig. 6 Step response curve of system simulation

Conclusion
The paper design a fuzzy auto-tuning PID controller having both advantages which combine
effectively fuzzy control concepts with conventional PID control. And this controller is used in CNC
servo system. According to the error and the rate of error change, the controller can regulate the three
parameters of PID controller online .The simulation in the Simulink environment shows the steady
precision and anti-jamming performance of servo system have been improved. Servo feed system
obtains better dynamic performance and better robustness.

References
[1] Rajani K.Mudia, Chanchal Deyb , Tsu-Tian leec, in: An improved self-tuning scheme for PI
controllers[J],ISA Transactions,2008(47) : p.45-52.
[2] Wenhua Gao, Hongzhi Su, Fuzzy self-tuning PID control of NC machine tool feed servo system.
Manufacturing Automation.2009.
[3] Yunhua Yin, Shuikang Fan, Mine Chen: The Design and Simulation of Adaptive Fuzzy PID
Controller.Fire Control and Command Control, 33(7) (2008), p.96-99.
[4] Haoxiang Yu, Hongxia Chu, Xifeng Wang: MATLAB Practical Course. (Chemical Industry
Press,Beijing, 2009), p. 133-157.
[5] Xiaodong Ye, MeiYing Ye. A Method for Model reduction of High-order System. Journal of
basic science and engineering,2007.
[6] Zhibing Shu, Junquan Liu: Mathematical Model of Closed-loop Servo System.System
Simulation (2002), p.1611-1613.
[7] Shihong Xie: MATLABR2008 Control System Dynamic Simulation Tutorial examples.
(Chemical Industry Press Beijing,2008), p.234.
Advanced Design Technology, ADME 2011
10.4028/www.scientific.net/AMR.308-310

Simulation of Fuzzy Auto-Tuning PID Control Based on MATLAB


10.4028/www.scientific.net/AMR.308-310.579

You might also like