Synchronous Optimal Pulse Pattern Modulation and Predictive Control for Permanent-Magnet Synchronous Motor Drives in Traction Applications

Abstract

Permanent-magnet synchronous motors (PMSMs) are widely used in traction applications due to their high efficiency and power density. However, their performance is often limited by conventional modulation and control strategies, especially under low switching-to-fundamental frequency ratio (SFR) conditions. Synchronous optimal pulse width modulation (SOPWM) offers reduced switching losses and low current total harmonic distortion (THD), but it does not explicitly suppress torque harmonics and exhibits limited dynamic performance due to sensitivity to sudden variations in the reference voltages. This thesis proposes a comprehensive control framework that enhances PMSM drive performance in traction applications by addressing torque harmonics, dynamic response, and switching frequency regulation. First, a novel optimal pulse pattern for torque harmonics minimization (OPP-THM) is introduced. Torque harmonics are analytically derived as functions of back-electromotive force (back-EMF) and inverter voltage harmonics, enabling the design of switching angles that directly suppress low-order torque harmonics. Experimental validation confirms that the proposed OPP-THM technique outperforms SOPWM and space-vector pulse width modulation (SVPWM). Second, to improve dynamic response, a ranking-based finite control set model predictive control (FCS-MPC) method is developed and integrated with SOPWM. This approach eliminates the need for conventional SOPWM-based field-oriented control (FOC) and PI controllers, offering faster transient response while maintaining low current THD. The ranking-based formulation also simplifies the traditionally complex weighting factor tuning process in FCS-MPC. Finally, a novel predictive switching frequency and current control (PSFCC) method is proposed. Unlike standard FCS-MPC, which lacks inherent switching frequency regulation, PSFCC introduces direct control over switching actions, decoupled from the controller sampling frequency. Experimental validation on a PMSM test bench demonstrates excellent performance across a wide range of operating conditions, making the proposed strategy a robust solution for next-generation electric vehicle drives.