Optimal Design of Type_1 TSK Fuzzy Controller Using GRLA for AVR System

Author

Naderi, F. ; Gharaveisi, A.A. ; Rashidinejad, M.

Author_Institution

Shahid Bahonar Univ. of Kerman, Kerman

fYear

2007

fDate

10-12 Oct. 2007

Firstpage

106

Lastpage

111

Abstract

A new methodology for designing optimal systematic GA-based fuzzy controller is presented in this paper. Our design is based on genetic reinforcement learning algorithm (GRLA), unlike normal GA, GRLA is only based on the competition and cooperation among chromosomes for survival. The proposed method tries to find best chromosomes with good combination to form an optimal fuzzy controller. GRLA as design method has been applied to the automatic voltage regulator (AVR) system. The controller is able to follow the input reference. A Mamdani controller designated via an ordinary GA is applied using five membership functions for inputs and output variables. A PID controller designed by Ziegler-Nichols method is also implemented to the same problem. A comparison between the obtained results shows the effectiveness of the proposed GRLA method.

Keywords

control system synthesis; functions; fuzzy control; genetic algorithms; learning (artificial intelligence); three-term control; voltage regulators; GA-based fuzzy controller; Mamdani controller; PID controller design; Type1 TSK fuzzy controller; Ziegler-Nichols method; automatic voltage regulator system; genetic reinforcement learning algorithm; membership functions; Algorithm design and analysis; Automatic control; Biological cells; Control systems; Design methodology; Fuzzy control; Fuzzy systems; Genetics; Learning; Optimal control;

fLanguage

English

Publisher

ieee

Conference_Titel

Power Engineering, 2007 Large Engineering Systems Conference on

Conference_Location

Montreal, Que.

Print_ISBN

978-1-4244-1583-0

Electronic_ISBN

978-1-4244-1583-0

Type

conf

DOI

10.1109/LESCPE.2007.4437362

Filename

4437362