DocumentCode :
323423
Title :
The innovation of a variable structure control system-a striking fact about how A.I. can help system
Author :
Lu, Bingchao ; Xu, Hongbing
Author_Institution :
Dept. of Autom., Univ. of Electron. Sci. & Technol., Sichuan, China
Volume :
1
fYear :
1997
fDate :
28-31 Oct 1997
Firstpage :
736
Abstract :
The paper attempts to show how through artificial intelligence (AI) techniques such as knowledge based production systems etc., together with some novel ideas about making use of variable structure controllers (VSC), an innovative variable structure control system with better performance could be built. The system identifies online, the control characteristics related tightly to switching surfaces. With the aid of AI techniques and a novel approach of using each control mode according to the needs decided by judgement from the obtained control characteristic information, the system can be operated in a more favorable manner. Such an innovative variable structure controller (IVSC) would act more intelligently in dealing with plant/system with significant uncertainty, nonlinearity and other undesired properties for control. It can alleviate the chattering problem, and at the same time retain fast response, good robustness and disturbance rejection. The benefits achieved in IVSC are proved by computer simulation results
Keywords :
artificial intelligence; intelligent control; variable structure systems; AI techniques; IVSC; artificial intelligence; chattering problem; control characteristics; disturbance rejection; knowledge based production systems; switching surfaces; variable structure control system innovation; variable structure controllers; Artificial intelligence; Computer simulation; Control systems; Intelligent structures; Nonlinear control systems; Production systems; Robustness; Technological innovation; Time factors; Uncertainty;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Processing Systems, 1997. ICIPS '97. 1997 IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-4253-4
Type :
conf
DOI :
10.1109/ICIPS.1997.672885
Filename :
672885
Link To Document :
بازگشت