Simplified Multi-Step Predictive Current Control for Surface Permanent Magnet Synchronous Motor

Yaohua Li,Xiaoyu Wang,Zikun Liu,Guixin Chen,Dongmei Liu,Chao Ren

DOI: https://doi.org/10.1109/PRECEDE51386.2021.9680881

2021-01-01

Abstract:In order to solve the problem of computational burden in the process of multi-step model predictive current control (MPCC) of surface-mounted permanent magnet synchronous motor (SPMSM), two simplified multi-step predictive control strategies are proposed. The conventional multi-step model predictive current control (C-MPCC) model of SPMSM is established. Based on the law of voltage vector selection, a simplified strategy (S-MPCC) is proposed. Taking n-step prediction (n >= 2) as an example, C-MPCC needs (7((n+1))-7)/6 current prediction calculations and (7(n)-1) data comparison operations, while S-MPCC only needs (2n-1)x7 current prediction calculations and (25n-26) data comparison operations. Based on S-MPCC, another strategy to use an adding judgment to reduce calculation burden (S-MPCC-II) is proposed. Simulation results show that the motor system works properly under the control of the S-MPCC and S-MPCC-II. The control performances are almost the same as the C-MPCC. The execution time of three strategies in a single control cycle are compared based on the STM32H743 MCU platform. Experimental results show that the prediction time and the data comparison time in cost function optimization of S-MPCC are only 0.32% and 0.59% of C-MPCC, and the whole execution time of single control cycle is reduced to 0.30%. And the S-MPCC-II control strategy can further reduce the computational burden and improve the real-time performance of the system while maintaining the same control performances as the S-MPCC.