An improved double-vector model predictive control strategy for four-switch three-phase inverter-fed permanent magnet synchronous motor based on visualisation analysis considering DC voltage pulsation

Abstract

To address the problem of considerable current distortions in traditional single-vector model predictive control (MPC) method for four-switch three-phase inverter (FSTPI)-fed permanent magnet synchronous motor, a double-vector MPC scheme has been studied. However, it still lacks a sufficient theoretical foundation, causing doubts about its effectiveness. To perfect this deficiency, a novel visualisation analysis scheme considering DC voltage pulsation is presented to demonstrate the validity of the traditional double-vector MPC method for FSTPI. The analysis results reveal for the first time that the traditional double-vector MPC method is not always superior to the single-vector method as there is no zero voltage vector for FSTPI. So, a new virtual voltage vector is defined to compensate the deteriorated control performance caused by the absence of zero voltage vectors, based on which an improved double-vector MPC method is further put forward to reduce the current ripples. Then, the presented visualisation analysis is adopted to demonstrate the validity of the improved MPC strategy under pulsating DC voltage conditions. Finally, a new voltage sector division technology is further put forward to simplify the implementation steps of the presented method. Contrastive experimental studies demonstrate the availability of the improved MPC strategy.