Abstract:

This work reviews the design process and real-time implementation of energy management systems (EMS) for dual-motor battery electric vehicles (DM-BEVs), starting from the fundamental concepts of two-motor coupling, to achieve higher energy savings and dynamic performances compared to single-speed, single-motor BEVs. The advantages and disadvantages of the most popular dual-motor architectures and their influence on the EMS design complexity are presented, followed by a revision of the reported energy management controllers. Besides the most prominent methods, classified as rule-based or optimization-based techniques, reinforcement learning-based EMSs are discussed in detail, given their near-optimal, real-time implementation and adaptability to newer, unforeseen drive cycles. Finally, the standard procedures and equipment required to assess the EMS’ performance with hardware-in-the-loop tests are presented. Conclusions and open challenges for the energy management controllers of DM-BEVs are discussed at the end of this work.