Multi-Virtual-Vector-Based Predictive Current Control for Fault-Tolerant Inverter-Fed Dual Three-Phase Permanent Magnet Synchronous Machines

Dual three-phase permanent magnet synchronous machine (DTP-PMSM) has received extensive interests due to its inherent fault-tolerant capability. However, conventional single-virtual-vector-based predictive current control (SVV-PCC) for DTP-PMSM with open-phase fault (OPF) presents poor steady-state performance and high computational burden. To address these issues, a multi-virtual-vector-based PCC (MVV-PCC) for DTP-PMSM with OPF is proposed in this paper. Therein, two virtual vectors and one zero vector with optimal duty cycles are determined in each sampling period to improve the steady-state performance. A simplification strategy for selecting the optimal virtual vectors and calculating the optimal duty cycles is established to reduce the computational burden in digital implementation. Extensive comparisons are conducted to verify the effectiveness of the proposed MVV-PCC scheme.