Improving performance of electric vehicle drive system based a five-phase PMSM under fault using ANN and MPC

This study presents an advanced speed tracking control strategy designed to handle open-phase motor faults in EV drive systems. The proposed control strategy is evaluated using two distinct drive cycles. A five-phase interior permanent magnet synchronous motor is employed due to its notable advantages, including high efficiency, reliability, power density, and inherent fault tolerance. The control strategy leverages a multi-layer perceptron artificial neural network for online tuning of proportional-integral controller gains, enabling adaptive performance. This approach is benchmarked against a recent metaheuristic optimization algorithm known as Honey Badger. To further enhance performance, model predictive control is applied using a tailored cost function to minimize current harmonics and torque ripple. Simulation results, conducted in MATLAB Simulink, validate the effectiveness of the proposed method. Compared to conventional technique, the new approach achieves lower values for motor torque ripples and speed percentage overshoot, mean square error and integral absolute error across both test drive cycles. Additionally, the proposed method gives lower THD and attains energy saving.