Research On Application Of AC Servo System Based On Multi - Modal Control

Rocket gun has a series of advantages compared with other weapon,including wider firing coverage range, longer firing distance and stronger rapid deployment capability, and so on.So,it has been favored by the armies.Each army also increased their demands for rocket gun’s shooting accuracy to cope with the modern war environment. One feasible and efficient approach to meet this demand is to adjust the accuracy of rocket as rocket gun will be disturbed by some unpredictable nonlinear factors when it is being shot.This paper would study the control strategies of rocket gun’AC servo position system. By looking into the system, it is hoped that the accuracy of rocket gun’s adjustment could be improved.The main work of this paper contains the following four aspects:(1)Analysing AC servo position system’s structure and principle, establishing and deducing mathematic model for the system.(2) Since the mathematic model of the system is based under assumption,so this paper uses BP neural network and wavelet neural network which includes particle swarm algorithm to conduct system identification. And the comparing results shows that the particle swarm wavelet neural network program works better.(3) The paper designs a multi-model controller which is combined with fuzzy-PID controller and RBF networks self-tuning controller. It has adopted fuzzy switching strategy to achieve soft switches between different modes.The simulation test shows that the multi-model controller meets the system’s requirements well.(4) The paper uses simulation test bench to verify the performance of control.The test results showed that the control could good work for rocket system.