Reinforcement self-adaptive evolutionary algorithm for fuzzy systems design
Yung-Chi Hsu, Sheng-Fuu Lin · 2008
This paper proposes a reinforcement self-adaptive evolutionary algorithm (R-SAEA) with fuzzy system for solving control problems. The proposed R-SAEA combines the modified compact genetic algorithm (MCGA) and the modified variable-length genetic algorithm (MVGA) to perform the structure/parameter learning for constructing the fuzzy system dynamically. That is, both the number of rules and the adjustment of parameters in the fuzzy system are designed concurrently by the R-SAEA. The illustrative example was conducted to show the performance and applicability of the proposed R-SAEA method.