Application Of Genetic Algorithm On The Design For Optimal Fuzzy Controller

The Genetic Algorithm (GA) is a kind of global optimal searching algorithm based on Darwin's nature evolution theory and Mendel's genetics and mutation theory. Through simulating the evolution of creature (nature selection, crossover, mutation), Problem-space's corresponding code-space is searched efficiently and simultaneously, and optimal result is get. GA is apposite to compound and non-linear problems where traditional methods can hardly get good results.Fuzzy Control is an important branch of Intelligent Control, it mainly not depends on controlled-object but simulates human's experience, thus, Fuzzy Controller can fulfill some human's intelligence and is widely used in complex process and object-model control.Efficiency of Fuzzy Control depends on several key parameters, membership functions, fuzzy control-rule table and scale factors. Conventional methods determining the parameters are man-operated especially based on experts' experience and practical modulation, so subjectivity and randomness exist. In this thesis a new method based on decimal-code is provided, which unites the membership functions and fuzzy control rule-table and scale factors into integrated one for whole-scope searching, and the operators of GA is modulated fit for the change of the code. In view of the plenty of redundant information in the fuzzy control-rule table, GA is used to filtrate the control-rule table, at last a less optimal resultis obtained under the less control rules. The result of simulation shows that this method is effective and applicatory.