On Fuzzy Ac Servo System

The combination of fuzzy logic and neural network becomes an attractive research field in recent years. Neural network specializes in system identification and self adaptation to changeable environment, while fuzzy inference system is good at reasoning and decision making based on expert knowledge. Combining both of them can lead an access to solve practical problem more efficiently.CMAC, as a neural network, could approximate local optimum effectively. FMAC results from the integration of CMAC and fuzzy logic. This controller represents fuzzy and continuous cerebella cognizance. Its strong self-learning capacity could adapt to various complex control requirement, avoiding the drawbacks of either CMAC or fuzzy control.This paper studies deeply on the CMAC structure identification and fuzzy control, and then applies FCMAC to solve the control problem of servo system on certain long distance rocket launcher. The content includes as follows:①Apply CMAC to identify the structure of AC servo system;②Build an intelligent controller based on fuzzy CMAC, and introduce normal PD feedback controller to refrain disturbance;③Simulate and analyze the AC servo system through MATLAB. The simulation results show that Fuzzy CMAC performs well due to reduced overshoot and enhanced reaction speed.