Study On Improved Direct Thrust Control Technology For Permanent Magent Linear Synchronous Motor

Permanent Magnet Linear Synchronous Motor(PMLSM) is used in various of linear running domain, such as in industrial operation, transportation, automation and defense departments, and has a broad application prospects, with characteristics of a large thrust, low-loss, small electrical time constant and fast response. But it is a multi-variable, nonlinear and strong coupling system, so direct thrust control(DTC) technology is applied, which has some advantages of fast dynamic response, strong robustness for the system parameter perturbation and external interference.Mathematical models of PMSLM in different coordinates are created based on fundamental theory for PMLSM as well as structure diagram of the direct thrust control system is analyzed in detail. Then, in order to reduce the flux and thrust ripples predicted, predicted voltage vector control technology is adopted based on the analysis for hysteresis control of stator flux and electromagnetic force in DTC, which has strong dependence on motor parameters and determined by the motor mathematical model, and predicted voltage vector is obtained. A direct thrust control system based on sliding mode controller is developed, which has a insensitive trait and strong robustness for parameter. Aiming at obvious "chattering" phenomenon innately in sliding mode variable structure, and the defect of slow convergence rate, adaptive linear neural network will be applied to the sliding mode controller.A novel expected voltage direct thrust control system base on sliding mode variable structure and nerual network is designed.Results that the system not only enhances the dynamic quality of system and improves the convergence speed, raise the system overall performance, but also chattering is greatly reduced are showed based on the simulation.