A Three-Stage Method for Designing Genetic Fuzzy Systems by Learning from Examples

摘要

In this paper, we present a three step method for designing Genetic Fuzzy Systems combining an iterative and increasing rule derivation stage and two genetic-based simplification and tuning processes. The performance of the method proposed is shown by measuring the accuracy of the Fuzzy Logic Controllers designed in the fuzzy modeling of two three-dimensional control surfaces and comparing them with others generated by using Wang and Mendel's method, one of the most widely known iterative rule derivation processes.