A fuzzy based memetic algorithm for tuning fuzzy wavelet neural network parameters

The present paper proposes a memetic algorithm for tuning Fuzzy Wavelet Neural Network (FWNN) parameters in an adaptive way; to achieve this goal, our proposed algorithm combines Particle Swarm Optimization (PSO) as an evolutionary algorithm and an innovative local search which is based on a Fuzzy Inference System (FIS). The PSO increases the exploration ability of the memetic algorithm while the local search enhances its exploitation ability. To evaluate the performance of the proposed method, we have assessed our method by three known nonlinear problems commonly applied in the literature for modeling. In comparison with other methods used in the literature, our proposed method showed certain advantages, namely: a fewer number of obtained rules for FWNN, much better results in terms of error criteria, and faster convergence speed.