A hybrid methodology based battery power management and torque control of open-end winding induction motor drive for electric vehicles

This paper proposes a hybrid strategy for an isolated battery-powered induction motor drive with two stages fed by an inverter and open-end winding, designed for electric vehicle applications. The hybrid approach, termed EOO-RERNN, combines the Eurasian Oystercatcher Optimizer (EOO) and Recalling Enhanced Neural Network methods. RERNN predicts the control pulses generated by EOO, which are utilized to control the voltage source inverter (VSI) switches. The system comprises dual motors, two battery packs (BPs), and VSIs supplying power to both the rear motor (RM) and front motor (FM). The primary objective of the EOO-RERNN technique is to regulate each DC-link's power share to maintain equilibrium in the electric vehicle's charge. Direct Torque Control (DTC) is employed due to its quick and precise torque response and tolerance for parameter variations. Proper tuning is required to control the state of charge battery level. The EOO-RERNN method is implemented in the Matrix Laboratory (MATLAB) platform, with performance contrasted to other methods such as Wild Horse Optimizer (WHO), Grasshopper Optimization Algorithm (GOA), and Heap-Based Optimizer (HBO). Simulation results indicate a motor speed of 4850 rpm for the EOO-RERNN method.