Fuzzy Neural Network In The Ac Servo System

In this paper the research object is new model alternating current servo system. According to the request that was the static state- dynamic state precision and the celerity of response brought forward by task book, aimed at the bad circs of large inertia variety load and the sway of launch flat, We made deep research. At the same time attempting and groping for the combination of fuzzy logic control and neural network algorithm, putting forward the fuzzy neural network structure and control algorithm that is in series and validating it in practice.As soon as the servo system is concerned, the celerity of control algorithm is a very important capability guideline. First in the paper network parameter is processing, learning off-line and revising accidence, then learning on-line at real control. When the error is big, it should be controlled by time-best choiceness. System is permitted to move at the quickest velocity. The measure above all can resolve the contradiction of control precision and the response real time very good.The result of experimentation is implied: designed control can settle position revising question above all about complex and bad condition very well, it satisfied the request of design.