The Research Of Permanent Magnet Synchronous Motor Based On Model Predictive Control

In recent years, belong to the development of high-performance rare earth permanent magnet materials and a new generation of power electronic devices, permanent magnet synchronous motor(PMSM) get the attention because of its small size, light weight, high efficiency and high power density. It has been promoted more and more widely in the industrial, agricultural production, defense and other aspects of daily life. The performance of PMSM mainly depends on its control technology, so the control technology deserves to be studied deep.First this paper introduces the development of PMSM and its control technology, then analyses the mathematical model of PMSM in different coordinate systems. Based on the study of the PMSM classic speed control method, i.e. vector control(VC) and direct torque control(DTC), this paper take the focus on the model predictive control(MPC). VC and DTC are the two mature control method in the industry. VC has good steady performance, wide speed range, but its dynamic performance is not very good due to the double closed-loop, at the same time tuning the PI parameters is relatively complex. DTC uses hysteresis comparator and switching vector table. The merit of DTC is fast dynamic response, good robustness, but the ripple of DTC is larger, and steady performance has been criticized by scholars. MPC has simple control idea, through the PMSM model and control variables it makes the cost function, then calculates the cost function and chooses the minimum as the output vector. According to different control variables, MPC can be divided into model predictive torque control(MPTC) and model predictive current control(MPCC). This paper researches both methods and points their advantages and disadvantages.Under the premise of the simulation by MATLAB/Simulink and building a set of experiment platform, this article finishes the simulation and experiment of VC, DTC and MPC and makes a contrast through the experiment data, finally gets a conclusion. Simulation and experimental results demonstrate the feasibility study and achieve the expected topic result.