Adaptive Fault-Tolerant Operation of Three-Level ANPC Inverter in PMSM Drives Using Model Predictive Control

The three-level active neutral point clamped (3L-ANPC) inverter provides enhanced control flexibility and improved fault-tolerant capability. However, existing fault-tolerant approaches face challenges in achieving a balanced trade-off among various targets. Finite control-set model predictive control (FCS-MPC) is widely applied in multilevel inverter drives, primarily owing to its remarkable capability to handle multi-objective optimization using a dedicated cost function. Therefore, this article introduces FCS-MPC into the fault-tolerant control of 3L-ANPC inverters, achieving multi-objective optimization. Additionally, an artificial neural network method is employed to dynamically optimize the weight factors of FCS-MPC, addressing the issue of multi-objective optimization in fault-tolerant mode. Finally, the feasibility of proposed method is verified by the simulation and experiment results.