Real-Time Torque-Distribution for Dual-Motor Off-Road Vehicle Using Machine Learning Approach
Marouane Adnane,Chi Thi Phuong Nguyen,Ahmed Khoumsi,Joo Pedro F. Trovo,Chi T.P. Nguyen,João Pedro F. Trovão
DOI: https://doi.org/10.1109/tvt.2024.3355186
IF: 6.8
2024-01-01
IEEE Transactions on Vehicular Technology
Abstract:Recently, demand for electric vehicles (EVs) has increased significantly as people are becoming more conscious of the environment and the need to reduce carbon emissions. The introduction of multi-motor systems in EVs has brought new challenges in terms of energy efficiency and performance. This paper presents a Multi-Ensemble Learning (MEL)-based approach to design an Energy Management Strategy (EMS) for a Dual Motor Electric Vehicle (DMEV) where MEL is a new powerful Machine Learning approach implemented using Python programming language. To make our study concrete, we studied a real DMEV that is modeled using Energetic Macroscopic Representation and whose control is simulated using MatlabSimulink. The designed EMS aims to distribute the instant torque between the two electric motors in an efficient manner, with the objective of minimizing energy consumption as much as possible. Contrary to existing EMSs, an important advantage of our designed EMS is that it determines the instant torque distribution in real-time (while the vehicle is running), without knowing in advance how physical parameters (such as the speed and traction force) will evolve during the current trip. The real-time simulation is carried out under unknown driving cycles based on a validated numerical EV model with a significantly lower computational cost while achieving a high degree of accuracy in predicting and allocating torque, and a high degree of performance in terms of energy consumption.
telecommunications,engineering, electrical & electronic,transportation science & technology