DocumentCode
2565324
Title
The optimization design of fuzzy controller based on an improved artificial immune algorithm
Author
Miao Yu ; Jie Chen ; Li-hua Dou ; Ming-gang Gan
Author_Institution
Dept. of Autom. Control, Beijing Inst. of Technol., Beijing
fYear
2008
fDate
2-4 July 2008
Firstpage
3383
Lastpage
3387
Abstract
An improved artificial immune algorithm is proposed. A large and a small neighborhood are constructed to perform global and local search respectively, so that two-level neighborhood searching mechanism is realized to ensure the search capabilities of the algorithm, and its combination with the control system is presented. Aiming at the fuzzy controller difficulties in the field of intelligent control, it is suggested a design method that employs the improved artificial immune algorithm to adjust the scale factors and membership functions of the fuzzy controller respectively. Then, it is designed a fuzzy controller that is composed of two planar fuzzy controllers by using Sugeno fuzzy reasoning method, which successfully controls an inverted pendulum system. The simulation results show that the improved algorithm for single-stage inverted pendulum has an implementation of effective control, and it makes the designed performance of fuzzy controller greatly improved.
Keywords
artificial immune systems; control system synthesis; fuzzy control; fuzzy reasoning; intelligent control; Sugeno fuzzy reasoning; artificial immune algorithm; fuzzy controller; intelligent control; optimization; Algorithm design and analysis; Automatic control; Control systems; Design optimization; Fuzzy control; Fuzzy reasoning; Fuzzy systems; Gallium nitride; Intelligent control; Laboratories; Artificial immune system; Fuzzy control; Optimization design;
fLanguage
English
Publisher
ieee
Conference_Titel
Control and Decision Conference, 2008. CCDC 2008. Chinese
Conference_Location
Yantai, Shandong
Print_ISBN
978-1-4244-1733-9
Electronic_ISBN
978-1-4244-1734-6
Type
conf
DOI
10.1109/CCDC.2008.4597957
Filename
4597957
Link To Document