Model Predictive Adaptive Fault-tolerant Control of DTP-PMSM Based on Multiple Optimization Objectives

The traditional fault-tolerant procedure is usually divided into fault detection and fault tolerance. Fault detection is a prepositive step of fault tolerance, which is applied to locate the specific fault. However, the complexity and misdiagnosis risk of fault-tolerant systems will increase. In response to the open-phase faults of dual three-phase permanent magnet synchronous machine(DTP-PMSM) drives, an adaptive fault-tolerant control without fault detection is introduced. When the open-phase fault of DTP-PMSM has occurred, firstly, the x-y subspace current is open-loop controlled, whereafter, according to the x-y open-loop current, the angle of reference current is acquired optimally, finally, the unified fault-tolerant reference current is built up by using the above angle. Appling this method of omitting fault detection, the fault-tolerant reference current can be given to the value function of model predictive control adaptively. Meanwhile, in order to realizing the fault-tolerant control of multiple optimization objectives, the control objectives of the maximum torque(MT), minimum copper loss(ML), and single 3-phase(S3) can be combined by importing the weight coefficient in the fault-tolerant reference current. The experimental results show that the proposed method can realize adaptively and rapidly fault-tolerant control and has a very smooth switching capability of different optimization objectives by adjusting the weight coefficient.