FUZZY Logic and Genetic Algorithms Applied to Intelligent Control System Design

Abstract

The design methods of fuzzy logic controllers (FLCs) using the genetic

' algorithms

(GAs) are appearing as systematic methods. These methods easily provide an optimized design and form the framework for further progress. This thesis describes the optimal design of P-, PI-, PD-, PID-like FLCs based on theoretical fuzzy concepts and genetic­ based optimizations. In the case of PI-like FLC, the most important feature of the proposed controller is its simple structure consisting of three gain factors and a FLC with each of the two inputs and the output has only three membership functions. Through GAs, the optimal values for the gain factors associated with each of the inputs and the output of the FLC is achieved without changing the membership functions or the rule-base of the FLC itself.