In recent years, energy management systems have become an emerging research topic. This concept allows the distribution of energy-intensive loads among various energy sources. An appropriate resource allocation scheme is necessary for the controller to efficiently allocate its energy resources in different operating conditions. Recent advances in artificial intelligence are instrumental to solve complex energy management problems by learning large repertoires of behavioral skills. This consists of hand-engineered policy and human-like expertise representations. In this paper, a deep reinforcement learning based resource allocation scheme is proposed for electric vehicles avoiding to work at the level of complex vehicle dynamics. Using multiple energy storage devices, like batteries, in parallel increases their maintenance due to their different behavior in various operating conditions. Thus, the proposed strategy aims to learn optimal policies to equilibrate the state of charge (SOC) of all batteries extending their lifespan and reducing their frequent maintenance.

Additional Metadata
Persistent URL dx.doi.org/10.1109/VPPC.2018.08605023
Conference 15th IEEE Vehicle Power and Propulsion Conference, VPPC 2018
Citation
Chaoui, H, Gualous, H. (Hamid), Boulon, L. (Loic), & Kelouwani, S. (Sousso). (2019). Deep reinforcement learning energy management system for multiple battery based electric vehicles. In 2018 IEEE Vehicle Power and Propulsion Conference, VPPC 2018 - Proceedings. doi:10.1109/VPPC.2018.08605023