Linear Permanent Magnet Synchronous Motor Servo System Intelligent Neural Network Control

Since the beginning of the 21st century, because of the rapid development of science and technology. Make linear permanent magnet synchronous motor gained widespread application and development. Compared with the traditional rotating motor, linear permanent magnet synchronous motor has advantages such as simple structure,high efficiency,high precision,good reliability and other advantages. So it have been widely used in aviation,spaceflight, Numerical control machine tools,machining center and robot occasion.This paper mainly made a research in the linear permanent magnet synchronous motor servo,and select the linear permanent magnet synchronous motor as the controlled object,For linear permanent magnet synchronous motor servo system itself existed defects,the main task is that the control scheme for servo system optimization design.With classic'rotating electrical machine add ball screw'compared to the way to enter,although linear permanent magnet synchronous motor servo system greatly eliminate the bad influence caused by mechanical transmission, but it gives linear permanent magnet synchronous motor control brought some difficulties,in order to linear permanent magnet synchronous motor servo system obtains good control effect,the search for an effective control strategy is imminent.So far,the traditional PID control is the oldest and most widely used one of the basic control modes.But because linear permanent magnet synchronous motor has strong nonlinearity, time-varying uncertain and is hard to establish practical mathematical model and so on characteristics; So the traditional PID control system has been difficult to achieve good control performance.With the continuous development of intelligent control technology,especially the fuzzy control and neural network control to appear,To solve the complicated nonlinear time-varying control problem of uncertain systems opened up new horizons.Will the traditional PID control and fuzzy logic formed by the combination of fuzzy PID control and the RBF neural network with the traditional PID control formed by the combination of RBF neural network setting PID control,in a certain extent,improved linear permanent magnet synchronous motor servo system control performance.Although fuzzy PID control,RBF neural network setting for linear PID control in linear permanent magnet synchronous motor servo system control opened up new ideas,but when we use the fuzzy control and neural network to try to overcome their nature of existence defects,such as fuzzy control get bogged down in the control rules heuristic adjustment,and neural network cannot make full use of using engineering and technical personnel of experience and knowledge,especially parameters iterative algorithm is easy to local minima and even hard to meet the real-time control. In order to improve the fuzzy control and neural network above the disadvantages and limitations,this paper proposes a kind of intelligent neural network control strategies:with RBF neural network as servo system identification device,comprehensive fuzzy control,neural network and the traditional PID control in one,s own advantages by the formation of the intelligent neural network as controller,parameter optimization algorithm is using classical BP algorithm. Using MATLAB separately on the traditional PID control,fuzzy PID control,RBF neural network setting PID control and intelligent neural network simulation.The simulation results indicates that the proposed intelligent neural network control has good robustness and strong tracking.Therefore,it shows that using intelligent neural network control strategy into linear permanent magnet synchronous motor servo system control is a kind of effective control method.