Natural Fault-Tolerant Model Predictive Voltage Control for Symmetrical Six-Phase Vernier Permanent Magnet Motor With Open-Phase Fault

Most fault-tolerant control schemes work based on fault diagnosis and control reconstruction, which suffers the misdiagnosis risk and process complexity. To address this issue, the natural fault-tolerant schemes are being developed for multi-phase motors, however, the use of virtual vector synthesis still makes it a large computational burden. This paper proposes a new natural fault-tolerant model predictive voltage control (NFT-MPVC) for the open-phase fault of a symmetrical six-phase vernier permanent magnet motor (SSP-VPMM), which directly uses appropriate actual voltage vectors specifically existed in symmetrical six-phase motors to avoid the complex calculation and the high switching frequency caused by virtual vector synthesis. Meanwhile, by combing the deadbeat control principle with sector judgment, the number of candidate voltage vectors is reduced to only two in one control cycle, thus further simplifying the calculation. The simulation and experimental results show that the proposed NFT-MPVC can provide satisfactory fault-tolerant performance with lower switching frequency compared with traditional natural fault-tolerant scheme using synthesized virtual vectors, and the computational burden of the proposed method can be reduced by at least 22%.