Research On Model Predictive Control Method For Permanent Magnet Synchronous Motor Servo System

Permanent magnet synchronous motor(PMSM)is widely used in AC speed regulation system,owing to its specific characteristics such as high power factor,compact structure,light weight and easy maintenance.However,it is difficult for conventional linear control strategies,such as the proportional-integral(PI)control method,to achieve a sufficiently high control performance,due to the existence of nonlinear friction,parameter uncertainties and load disturbances in PMSM servo system.With the development of modern control theory and micro-computer technology,many advanced control algorithms have been applied to control the PMSM.Model predictive control(MPC)is prevalent in practical application based on its capabilities of taking full advantage of the model knowledge and handling constrains effectively.In order to improve the comprehensive performance,MPC is proposed as an effective control scheme of PMSM servo system.In this paper,the MPC method of MPSM servo system is studied.Firstly,the speed-loop model predictive control algorithm of PMSM is designed on the basis of vector control.Besides,a variable prediction horizon MPC is designed based on the traditional MPC to further improve the dynamic performance of the system.The simulation and experimental results show that MPC has better dynamic performances than traditional PI control.Considering that friction reduces the tracking accuracy of servo system in low speed and round trip motion,a composite MPC method based on friction compensation and extended state observer(ESO)is designed in this paper.Firstly,the friction model of the system is established by the piecewise linear identification method.Then an ESO is proposed to estimate the lumped disturbance including friction compensation deviation and load disturbance.Finally,a more delicate prediction model is established by embedding the friction and disturbance information,and a MPC algorithm is designed based on this model.The simulation and experimental results show that this method can improve the speed response performance and position tracking accuracy when the motor moves at low speed and round trip.Although the traditional cascade control method has a strong robustness,it also limits the system bandwidth.Aiming to solve this problem,a single-loop MPC method is designed by combining the q-axis current controller and the q-axis speed controller together.The simulation and experimental results show that the single-loop MPC method has a better dynamic performance and a higher steady-state accuracy than the traditional cascade control method.