A Study On Model Predective Control Of Permanent Magnet Synchronous Motor Multirate Sampling System

Due to the characteristics of simple structure, high power density, high starting torque, and simple controlling, Permanent magnet synchronous motors(PMSM) are widely used in the industrial field. It is important to improve the control algorithm and the operating performance.Model predictive control (MPC) is based on the controlled object models, Because MPC consider the effect of the control amount in the future state of the system, compared with PI control algorithm, can improve the control effect more effectively. Compared with traditional PID control, MPC has faster dynamic response and smaller overshoot. In recent years, the model predictive control has been applied to permanent magnet synchronous motor control. Further study on model predictive control algorithm for permanent magnet synchronous motors is of practical significance.First, the paper introduces the development of permanent magnet synchronous motor and current research status of AC motor control theory. And PMSM mathematical model is established respectively in the three-phase stationary coordinate system, the two-phase stationary coordinate system and synchronously rotating coordinate system. Studied the principle of permanent magnet synchronous motor vector control and SVPWM. then permanent magnet synchronous motor simulation model based on double closed loop PI control method is established in Matlab/Simlink.Second, model predictive control theory is well-studied. For continuous linear time-invariant systems, discrete-time model predictive control theories have been built. Combined with multirate digital control system, Established the model predictive control model based onmultirate digital control system.Third, Single-rate and multi-rate sampling single-loop model predictive control algorithm for permanent magnet synchronous motor are designed. Simulation comparison of single-loop model predictive control based on Single-rate simpling, multi-rate simpling and PI control is provided. Simulation results show that single-loop model predictive control algorithm is effective and feasible. Compared with PI control, single loop model predictive control has better control effect. Multirate single-loop model predictive control can improve the single-rate single-loop model predictive control effect in steady-state operation.Finally, PMSM experimental platform hardware circuit is designed, PMSM experiment platform is built. Performed experiments of motor start up, load variation and speed regulating. The effectiveness and feasibility of Single-rate and multi-rate single loop model predictive control algorithm for PMSM are verified by experiments. Experimental results show that the single-loop model predictive control has better performance than the PI speed control, and multi-rate single-loop model predictive control can improve the steady performance of the single-rate model predictive control.