Generalized additive-multiplicative fuzzy neural network optimal parameters identification based on genetic algorithm
Donghai Zhai, Li Li, Fan Jin · 2003
In additive-multiplicative fuzzy neural networks (AMFNN), its membership functions have no adaptability and the number of fuzzy rules is determined subjectively. In this paper, a generalized additive-multiplicative fuzzy neural network (generalized AMFNN) is presented, and the parameters of membership functions can be adjusted. Therefore, there are many parameters to be determined. The matrix coding in genetic algorithm (GA), which combines binary coding with real number coding, is adopted to search the optimal parameters of the generalized AMFNN and determine the number of fuzzy rules. The generalized AMFNN has lower complexity and can approximate to a nonlinear system at high accuracy degree. A numerical simulation has demonstrated the validity of this approach.