Study Of Linear Motors Adaptive Disturbance Rejection Control Strategy

Permanent magnet linear synchronous motor (PMLSM) is a transmission devicedirectly converted electric energy into a linear motion mechanical energy without anyintermediate conversion mechanism, with characteristics of high-precision positioningand high acceleration and many other advantages. It is widely used in the field ofprecision machining semiconductor and robotics in recent years. Linear motor controlsystem is a nonlinear, strong coupling parameters change system, and there are additionaldisturbance factors in this system, all of these make it more difficult to control. Thetraditional direct thrust control and the general intelligent control algorithm parameteradjustment problem, system parameter variations and external disturbance robustnessissues are the problems need to be solved in the PMLSM control systems.As for the problem of traditional direct thrust system, hysteresis controller isimproved based on fuzzy theory, the pulsating of direct thrust control is reduced. Linearmotor parameters are uncertain and changing, traditional methods of control are difficultto achieve disturbance rejection, so the conventional adaptive control methods arereplaced by model-free adaptive control methods to get rid of the dependence on motorparameters and system stability so that real-time adjustment of the controller parameterscannot be achieved when the system is disturbed or motor parameter change. Model-freeadaptive control method is easy to be adaptive, it greatly improve the ability of motordisturbance rejection and the properties of static and dynamic in the system.The combination of radial basis function neural network estimation and themodel-free adaptive control method can solve the impact of additional disturb and timedelay caused by linear motor servo system speed measurement as well as other factors, toachieve permanent magnet linear synchronous motor disturbance rejection.According to model free adaptive control theory and radial basis function neuralnetwork theory, this paper established the simulation model of model-free adaptive directthrust control system based on neural network estimation through Matlab/Simulink, analyzed and verified its feasibility.