Torque Ripple Suppression of BLDCM with Optimal Duty Cycle and Switch State by FCS-MPC

This paper proposes a scheme that combines Finite control set and model predictive control for BLDC motors to suppress torque ripples. Traditional six-step commutation control method in brushless DC motors results in torque ripples, which limits its applications. To address this issue, the study employs the FCS-MPC technique. The research begins with a mathematical analysis of BLDC motors to identify the causes of commutation torque ripples. By maintaining a consistent slope of the non-commutation current, a discrete model is utilized to predict the current at the next time step. This prediction is then used to compute the optimal duty cycle. Additionally, to overcome the limitations of Pulse Width Modulation (PWM) control performance at high speeds, the paper proposes a switch insertion compensation strategy. This enhancement aims to improve the overall regulation and control of the motor at higher operational speeds.

Keywords: BLDCM, model predictive control, torque ripple, cost function.