Model Parameter Self-Correcting Deadbeat Predictive Current Control for SPMSM Drives

Abstract

To achieve ideal deadbeat current control of the surface-mounted permanent magnet synchronous motor (SPMSM) drives under parameter mismatch, a model parameter self-correcting deadbeat predictive current control method (MPSC-DPCC) is proposed in this article. First, the impact of parameter mismatch on conventional deadbeat predictive current control (DPCC) using an extended state observer (ESO) is analyzed, indicating that neglecting resistance leads to a transient drop in the current step response, and the under/overshoot caused by inductance mismatch cannot be improved by ESO. Afterward, a model parameter self-correction scheme (MPSC) is proposed to form a closed-loop disturbance regulation system by integrating with ESO. In this system, the observed disturbances are regulated to zero by adjusting the model parameters, enabling indirect correction of all model parameters without signal injection. The proposed MPSC-DPCC, comprising DPCC, ESO, and MPSC, exhibits strong robustness against initial model parameter mismatch and motor parameter variation. The ideal deadbeat current control without any transient drop and under/overshoot can be easily achieved, and it is not limited by the observer bandwidth. Experimental results validate the superiority of the MPSC-DPCC compared with conventional methods.