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

Area of Science:

• Electrical Engineering
• Automotive Engineering
• Control Systems

Background:

• Electric vehicle (EV) drive systems require robust control strategies to ensure reliability during motor faults.
• Open-phase faults in EV motors can degrade performance and necessitate adaptive control solutions.
• Five-phase interior permanent magnet synchronous motors offer advantages in efficiency, power density, and fault tolerance.

Purpose of the Study:

• To develop and evaluate an advanced speed tracking control strategy for EV drive systems experiencing open-phase motor faults.
• To enhance the fault tolerance and performance of five-phase interior permanent magnet synchronous motors.
• To minimize torque ripple, current harmonics, and energy consumption under fault conditions.

Main Methods:

• A five-phase interior permanent magnet synchronous motor model was used for simulations.
• A multi-layer perceptron artificial neural network was employed for online tuning of proportional-integral controller gains.
• Model predictive control with a tailored cost function was implemented to minimize harmonics and torque ripple.
• The proposed strategy was benchmarked against the Honey Badger optimization algorithm.

Main Results:

• The proposed control strategy significantly reduced motor torque ripples and speed percentage overshoot compared to conventional methods.
• Lower values for mean square error and integral absolute error were achieved across two distinct drive cycles.
• The method resulted in lower total harmonic distortion (THD) and demonstrated energy savings.

Conclusions:

• The advanced speed tracking control strategy effectively handles open-phase motor faults in EV drive systems.
• The integration of artificial neural networks and model predictive control provides adaptive and optimized performance.
• The proposed method offers a promising solution for improving the efficiency, reliability, and fault tolerance of EV powertrains.