Low-Complexity Model Predictive Control for Series-Winding PMSM with Extended Voltage Vectors

This paper proposes a low-complexity model predictive current control (MPCC) strategy based on extended voltage vectors to enhance the computational efficiency and steady-state performance of three-phase series-winding permanent magnet synchronous motors (TPSW-PMSMs). Compared to conventional MPCC methods, this approach increases the number of candidate voltage vectors in the alpha-beta plane from 8 to 38, thereby achieving better steady-state performance. Specifically, the proposed method reduces the total harmonic distortion (THD) by 59%. To improve computational efficiency, a two-stage filtering strategy is employed, significantly reducing the computational burden. The number of voltage vectors traversed in one control period is reduced from 38 to a maximum of 4, achieving an 89% reduction in traversals. Additionally, to mitigate the impact of zero-sequence currents, zero-sequence current suppression is implemented within the control system for effective compensation. By combining low computational complexity, reliable steady-state performance, and real-time control capabilities, this strategy provides an efficient solution for TPSW-PMSM systems. Simulation results validate the effectiveness of the proposed method.