基于拓展矢量组 MPC 的 PMSM 传导干扰抑制研究

永磁同步电机(permanent magnet synchronous motor, PMSM)驱动系统是电动汽车中最大的传导干扰来源。在考虑死区的影响后，传统单、双矢量模型预测电流控制在不使用零矢量的情况下依然存在共模电压尖峰，而增加抑制共模电压尖峰的约束会导致输出电流谐波增加。为此，文章提出一种基于拓展非零矢量组的有限集模型预测控制(model predictive control, MPC)方法，使用6组相邻的3个非零矢量作为备选矢量来扩大矢量合成范围，并合理安排矢量作用顺序，结合无差拍和调制的方法优化矢量作用时间，最后通过约束矢量作用时间和矢量组作用顺序，在消除共模电压尖峰的同时优化输出电流质量。仿真和实验结果证明了文章所提方法的有效性。

The permanent magnet synchronous motor (PMSM) drive system is the largest source of conducted interference in electric vehicles. After considering the impact of deadband, the traditional single-vector and dual-vector model predictive current control methods still exhibit common-mode voltage spikes when zero vectors are not utilized. Moreover, imposing constraints to suppress these spikes results in increased output current harmonics. To address this challenge, a finite set model predictive control (MPC) method based on extended vector groups is proposed. This method employs six groups of adjacent three non-zero vectors as alternative options to widen the range of vector synthesis. Additionally, it involves the rational arrangement of vector action sequences, the optimization of vector action times using space vector modulation and pulse width modulation techniques, and ultimately, the constraint of vector action times and vector group action sequences. This approach aims to eliminate common-mode voltage spikes while optimizing output current quality. Both simulation and experimental results validate the effectiveness of the proposed method.