Research Of Model Predictive Control In Permanent Magnet Synchronous Motors Drives

As the application of permanent magnet synchronous motor (PMSM) becomes extensive, the demand for high-performance and control accuracy of PMSM drive system is getting increasingly high. In many occasions, traditional control methods come into trouble to satisfy higher control needs than before. Model Predictive Control (MPC) does not depend on the control of error, so it is able to obtain high control precision, and has already achieved success in the applications of process control. But so far, the algorithm of MPC applied to the study of PMSM is still in its infancy at home and abroad. Depth study of MPC algorithm in PMSM can dramatically improve energy efficiency of motors, and also has great practical significance. For this, researches are made on MPC algorithm in PMSM, the main contents are as follows:1. Start from PMSM mechanism, analyze the mathematical model of PMSM, the current vector control method, the coordinate transformation principle, the principle and implementation of voltage space vector modulation. Simulate the PI double closed-loop control of PMSM using Matlab/Simulink, the simulation prepares for the research of predictive control.2. Combined with the basic principles and characteristics of Model Predictive Control technology, the design method of MPC controller is given. S-function module in Simulink has been used to create a simulation model of MPC in PMSM,3. An automatic differentiation (AD) based model predictive control (MPC) algorithm for Permanent Magnet Synchronous Motors (PMSM) is proposed. AD is used to expand the Taylor series. Then the obtained terms are adopted to calculate sensitivity to solve the problem. To reduce prediction error caused by parameter variation during operations, the extended state observer (ESO) is adopted to estimate a lumped disturbance term that is feed-forward then. Simulation results are presented to demonstrate the effectiveness of the proposed control scheme.