Journal: IEEE Transactions on Transportation Electrification
Authors: Kihan Kwon, Dong-Min Kim, Kyoung-Soo Cha, Junhyeong Jo, Myung-Seop Lim, Seungjae Min
DOI: 10.1109/TTE.2024.3416173
Multi-motor and multi-speed transmission systems for electric vehicles (EVs) can outperform conventional systems in energy efficiency and dynamic performance. Since motor characteristics directly affect EV efficiency, they were analyzed and verified by the simulation and experiment, respectively. The efficiency and performance of the two-motor and two-speed EV were evaluated using the motor characteristic results, and the importance of the motor design parameters was confirmed to improve them. To maximize both, a multi-objective optimization problem, including the objectives representing electricity consumption per 100 km (EC100) and acceleration time, was formulated. As a solution to the excessive computational burden arising from the optimization process, an artificial neural network (ANN) model was proposed. The ANN-model-based optimization was performed to find the optimal solutions, and a Pareto front, indicating a trade-off between efficiency and performance, was obtained. Furthermore, the results of the optimal motor design values demonstrated the necessity of accurate motor characteristic analysis according to changes in motor design parameters. Finally, the optimization results between various EV powertrain systems were compared to confirm the superiority of two-motor and two-speed systems. Especially, the EC100 and acceleration time were enhanced by up to 11.7% and 14.1%, respectively, compared to a powertrain system employing single-motor and single-speed.