Adaptive Model Predictive Current Control for Permanent Magnet Synchronous Motors under Poor Power Quality Conditions

Abstract

This study investigates an adaptive model predictive current control strategy for permanent magnet synchronous motors exposed to poor power quality. The work focuses on operating conditions marked by voltage sags and harmonic distortion, which are common in weak-grid environments. The proposed framework brings three features into a single predictive control structure: real-time sag compensation, multivector harmonic mitigation, and thermal derating. A complete PMSM model that includes electrical, mechanical, and thermal behaviour was developed in MATLAB and Simulink and used to test the controller under controlled disturbance scenarios. These scenarios covered voltage sags at 0.75 and 0.74 per unit, injected harmonic orders of the fifth, seventh, eleventh, and thirteenth, and combined disturbances. The adaptive controller was compared with a baseline predictive controller and a simple open-loop drive. Under nominal conditions, it maintained a rotor speed of 1500 revolutions per minute, a torque of 30.24 newton metres, and an efficiency of 85.8 percent. During sag events, it limited speed droop to roughly 0.4 to 0.5 percent, which was better than the baseline controller and far superior to the large drops recorded in open-loop operation. Under harmonic injection, it kept speed variation within 1.6 to 1.9 revolutions per minute and reduced torque ripple to about 2.4 percent. In extended operation it restricted the winding temperature rise to 20.9 degrees Celsius, which demonstrates improved thermal management. The findings show that the adaptive predictive controller offers stronger resilience, better efficiency, and safer thermal behaviour than the other methods. It represents a practical option for PMSM drives used in electric transport, industrial equipment, and renewable energy systems where grid quality is unreliable.