Design of sophisticated fuzzy logic controllers using genetic algorithms

The design of fuzzy logic controllers encounters difficulties in the selection of optimized membership functions and a fuzzy rule base, which is traditionally achieved by a tedious trial-and error process. This paper develops genetic algorithms for the automatic design of high-performance fuzzy logic controllers using sophisticated membership functions that intrinsically reflect the nonlinearities encountered in many engineering control applications. The controller design space is coded in base-7 strings (chromosomes), where each bit (gene) matches the 7 discrete fuzzy values. The developed approach is subsequently applied to the design of a proportional-plus-integral type fuzzy controller for a nonlinear water level control system. The performance of this control system is demonstrated to be higher than that of a conventional PID controller. For further comparison, a fuzzy proportional-plus-derivative controller is also developed using this approach, the response of which is shown to present no steady-state error