NEURAL NETWORK–DRIVEN EFFICIENCY OPTIMIZATION OF PMSM FOR ELECTRIC VEHICLE PROPULSION SYSTEMS

Abstract

Permanent Magnet Synchronous Motors (PMSMs) are widely adopted in electric vehicle (EV) propulsion systems due to their high-power density, fast dynamic response, and superior efficiency. However, maintaining optimal efficiency over wide speed–torque operating ranges and varying driving conditions remains a significant challenge. Conventional control techniques such as proportional– integral (PI) controllers and fixedparameter field-oriented control (FOC) schemes fail to ensure optimal efficiency under dynamic load and speed variations. This paper proposes a Neural Network (NN)–driven efficiency optimization strategy for PMSM-based EV propulsion systems. The proposed NN controller adaptively optimizes control parameters to minimize motor losses and improve overall efficiency. The effectiveness of the proposed method is evaluated under various speed and torque conditions. Simulation results demonstrate improved efficiency, reduced losses, and enhanced dynamic performance compared to conventional control approaches, validating the suitability of the NN-based strategy for EV applications.