A rule self-regulating fuzzy controller

Abstract

The most effective way to improve the performance of a fuzzy controller may be to optimize the fuzzy control rules. Usually fuzzy control rules are generated by: (1) translating the operators' experience into fuzzy linguistic form directly; (2) monitoring and summarizing the control behaviour of the operators; (3) modelling the process to be controlled using fuzzy set theory; (4) self organizing in running of the control systems. Different from the above means, the authors in [10] have proposed a method to regulate fuzzy control rules where the control rules are represented by an analytic expression with a regulating factor α. The authors in this paper employ such a method and use a simplified fuzzy control algorithm which allows fuzzy control rules to be regulated on line, thus constructing a Rule Self-regulating Fuzzy Controller (RSFC). Simulation results demonstrate that the performance of RSFC is better than that of a non-rule-regulating fuzzy controller. © 1992.