Research On Optimizing Fuzzy Logic Controllers With Genetic Algorithm

In designing fuzzy logic controller, obtaining optimal fuzzy rules and membership functions is a bottle-neck problem, hi traditional methods, fuzzy rules and membership function was obtained by expert. Furthermore, when they have once designed, they did not change in control process, which did not make fuzzy control performance the best. As a new full round and random optimize methods, genetic algorithm has been applied successfully in many fields. Many specialist had made research on combination of fuzzy control and genetic algorithm and have succeed in optimized fuzzy rules and membership functions.This paper presents the researches on the design process of fuzzy controller by analyzing the importance of fuzzy rules and membership functions and noticing some problem in the ordinary design procedure. We analyzed the successful point and the insufficiency for optimized fuzzy logic controller with genetic algorithm. Hence, two methods is purposed "A GA-based Optimization Method for Completeness and Semantics of Membership Functions" is presented for the completeness in optimized membership functions with genetic algorithm, "A GA-based Generating Method Based on Completeness of Fuzzy rule" is presented for the unreasonable in optimized fuzzy rules with genetic algorithm. And we give some simulation results.