Improved Fault-Tolerant Model Predictive Torque Control of Five-Phase PMSM by Using Deadbeat Solution

This paper proposes an improved fault-tolerant model predictive torque control (MPTC) for a five-phase permanent magnet synchronous motor based on deadbeat solution, which offers the reduced computation burden, good steady-state performance and simple control structure. Under the open-circuit fault condition, the virtual voltage vectors (VVV) which can restrain the harmonics work as the candidate vectors. The deadbeat solution is adopted to predict the reference voltage vector. The optimal VVV can be selected directly and rapidly according to the sector of the reference voltage vector. Hence the computation burden can be reduced largely because the process of testing all candidate VVVs is avoided. In order to reduce the error between the reference voltage vector and the selected VVV, a null vector is introduced by a simple geometric principle. Then, the duration of the selected VVV can be obtained. Therefore, the steady-state performance is enhanced because of the deadbeat solution and the multi-vectors in each control period. The experiments are presented to verify the proposed fault-tolerant MPTC.