A Novel Model Predictive Current Control With Reduced Computational Burden Based on Discrete Space Vector Modulation for PMSM Drives

Abstract

Discrete space vector modulation (DSVM) technique is commonly adopted in model predictive control (MPC) to mitigate current harmonics and torque ripples. Nevertheless, the employment of DSVM typically leads to heavy computational burden and high switching frequency (SF). To solve these problems, a novel model predictive current control (MPCC) scheme based on DSVM is proposed in this paper for permanent magnet synchronous motor (PMSM) drives. Firstly, a simple voltage vectors (VVs) pre-selection strategy based on the stator flux increment is introduced to eliminate the redundant virtual VVs generated by DSVM for the purpose of lower computational burden. Then, a hierarchical search strategy is designed to generate the candidate VVs online, which can further simplify the DSVM technique. In addition, an efficient optimal switching sequence (OSS) method is also employed to reduce the switching losses without weakening the control performance. Compared to the existing strategies, the proposed scheme possesses lower complexity and SF as well as superior performance. The effectiveness of the proposed scheme is supported by comparative experimental results on a PMSM platform.