Research On Predictive Control Algorithm Of Permanent Magnet Synchronous Motor

With the continuous development of permanent magnet synchronous motor control algorithms and DSP technology, digital PMSM servo control system of high-performance control strategy will be the trend of servo control system development. Since the high-performance PMSM servo system requires a rapid response of the current loop to ensure high dynamic performance of the system, and the traditional PI controllers are often difficult to meet this requirement because of its poor dynamic characteristics, and it is prone to oscillation. Although the current predictive control depends on accurate mathematical model, but it can achieve high dynamic response while avoiding overshoot, so further research in predictive control has important practical significance.This paper introduces the basic vector control framework of the permanent magnet synchronous motor, including the mathematical model of the motor space vector model, the stationary coordinate system, synchronous rotating coordinate system, the electromagnetic torque equation, and the equation of motion vector control technology and test the entire system in Simulink. By analyzing the PMSM mathematical model and designing PI controller parameters, and the inherent contradictions between overshoot and rapidity in PI controller can be proved. There is current error problem in term of the use of traditional predictive control, this article will introduce extended state observer, design and verify the effectiveness of the current extended state observer, and this provides the theoretical basis for subsequent compensation algorithm.Then, derived prediction mathematical model of permanent magnet synchronous motor and voltage source inverter in synchronous rotating coordinate system. Based on the prediction model of motor and inverter, further analyze the impact of the deviation of the controller parameters on current error and steady performance, simulation results under certain operating conditions conducted to illustrate the problem at the same time. According to the depth research of PMSM digital control technology, a compensation algorithm is proposed to eliminate the current steady-state error. Predictive control simulation model in Simulink environment are build up to simulate the proposed algorithm and the result shows good compensation effect. Finally, experiment of the proposed algorithm in physical experimental setup based on TMS320F2812 is build up. Simulation and experimental data verify the effectiveness of the proposed algorithm.