A Modified Self-Tuning Fuzzy-Neural Controller

Hsiao-Kang Hwang, Yu-Ju Chen, Chuo-Yean Chang, Rey-Chue Hwang

This paper presents a modified self-tuning fuzzy-neural controller in the applications nonlinear model reference control system. In order to make the controller have the adaptive control capability, the immediate system error (e(k)) and error change ( e(k)) are used to be the inputs for fuzzy-neural tuning mechanism. For simplifying the construction of fuzzy system, nine rules are used in the rule table. To demonstrate the superiority of the controller we developed, several nonlinear model reference control systems are studied and simulated. The simulation results clearly show that the self-tuning fuzzy-neural controller has quite promising potential in the real control applications.