Identification and design of multivariable fuzzy neural network system
Hongwei Yao, Mei Xiaorong, Xianyi Zhuang · 2002
A new method of fuzzy neural network identification is proposed. A function for measuring cluster validity is defined with which the number of fuzzy rules can be determined. A sufficient criterion that guarantees the global stability of the fuzzy system is presented. Based on this, a design method to optimize parameters of the fuzzy neural network controller by genetic algorithms is presented. This method is proved to have better effect through a double inverted pendulum by experiments.