Design of TSK-type fuzzy controllers using differential evolution with adaptive mutation strategy for nonlinear system control

摘要

This study proposes a differential evolution with adaptive mutation strategy (DE_AMS) for TSK-type fuzzy controllers (TFC) in nonlinear system control applications. The DE_AMS can optimize both the parameters of the membership functions and the weights of the consequent part for the TFC model. The proposed DE_AMS uses three common mutation strategies: DE/rand/1, DE/best/1, and DE/current-to-best/1. These three mutation strategies were chosen using a roulette wheel selection method in order to avoid the expensive computational costs of searching by a trial-and-error procedure. The experimental results demonstrate that in some circumstances, the proposed TFC-DE_AMS can increase the global search ability and obtain a lower error compared to other existing methods.