A review on robust predictive control of permanent magnet synchronous motor drives: Strategies and future trends

Abstract

Permanent magnet synchronous motors (PMSMs) are widely used in electric vehicles and industrial drives due to their high power density and efficiency. However, the performance of traditional predictive control strategies—such as sliding mode control and deadbeat predictive control—is highly dependent on the accuracy of motor parameters, which are prone to vary under real-world operating conditions. These parameter mismatches, along with external disturbances, can lead to current tracking errors, torque ripple, and degraded system robustness. This paper provides a comprehensive review of robust predictive current and speed control methods for PMSMs, focusing on three main approaches: disturbance observation, online parameter identification, and predictive model improvement. Furthermore, emerging trends such as multi-step prediction, enhanced ultra-local models, hybrid observers, and artificial intelligence-based techniques are discussed. This paper concludes with future research directions aimed at achieving higher robustness and intelligence in PMSM drive systems.