Robust Predictive Current Control for Fault-Tolerant Operation of Five-Phase PM Motors Based on Online Stator Inductance Identification

Abstract

The parameter mismatch in model predictive control (MPC), especially in fault-tolerant control, will lead to degradation of output performance. This article presents a new robust predictive current control based on online d-q inductance identification for a five-phase permanent-magnet (PM) motor under open-circuit fault. First, the MPC of the five-phase motor under the fault-tolerant operation of A-phase open-circuit is built, and then, the parameter mismatch caused by the d-q-axis inductance and the stator resistance is analyzed and discussed. Afterward, the real-time d-q-axis inductance identified by a fault-tolerance model reference adaptive system is used to replace the original parameters in MPC. Moreover, the flux linkage parameter mismatch can be rejected by the incremental predictive model in the proposed method. Therefore, the proposed method can eliminate almost all parameter mismatches, thereby improving output performance and enhancing parameter robustness under open-circuit fault. Finally, the validity and effectiveness of the proposed method are verified by experiments.