Improved Multistep Model Predictive Control for the Permanent-Magnet Synchronous Motor

Abstract

To further improve the control performance of the model predictive control (MPC) and balance the contradiction between the calculation complexity and prediction steps of the system, an improved multistep MPC is proposed in this article. First, the mathematical model of the permanent-magnet synchronous motor and the basic principle of the conventional MPC are introduced. Second, a multistep MPC method is proposed to calculate the rate of change of current in the case that the different basic voltage vectors are applied to the motor. Then, the current after several control periods can be predicted by this rate, which can help us select the appropriate voltage vector. In addition, the implementation process of the proposed method is described. Finally, the comparison experiments between the conventional MPC method and the proposed multistep MPC one are presented. The experimental results show that the proposed method is able to reduce the inverter switching frequency under the same control frequency, and can also improve the steady-state performance under the same switching frequency.