Torque Ripple Minimization of Predictive Current Control for IPMSMs with Improved Cost Function Design

Abstract

This paper applies model predictive current control (MPCC) to a typical 2-level voltage source inverter interior permanent magnet synchronous motor (VSI-IPMSM) rail vehicle drive system. However, simply controlling the current cannot minimize the torque ripple. An improved cost function design is proposed for VSI-IPMSM drives. The torque ripple information is integrated into the current error penalty function, so that the d-axis and q-axis current errors are reasonably distributed. The proposed cost function improves torque control accuracy at low sampling frequency and low switching frequency without adding torque and flux observer. The effectiveness of the proposed method has been verified by hardware-in-the-loop experiments.