Torque Distribution Prediction for Dual-Motor Electric Vehicle Using Ensemble Learning Algorithms

The increasing demand for Electric Vehicles (EVs) has made energy efficiency and performance crucial. This paper proposes a Multi-Ensemble Learning-based Energy Management Strategy (EMS) approach for a Dual-Motor Electric Vehicle (DM-EV) to address these challenges. Energetic Macroscopic Representation is used to model the DM-EV, and Matlab/SimulinkT is used to simulate the control. The proposed model is designed with Python programming language and aims to distribute the instant torque between the two electric motors efficiently, minimizing energy consumption in real-time, without prior knowledge of physical parameters. A real-time simulation under an unknown driving cycle was validated using a numerical EV model and achieved promising results while having a significantly lower computational cost compared to existing EMSs. The proposed model shows a high degree of efficiency in predicting and allocating torque, making it a promising solution for efficient energy management in multi-motor EVs.