Recharge intelligente décentralisée de véhicules électriques à l’aide d’un système multi-agent adaptatif utilisant des bandits manchots

The drastic growth of electric vehicles and pho- tovoltaics can introduce new challenges, such as electrical current congestion and voltage li- mit violations. These issues can be mitigated by ENS Rennes, SATIE/IETR, Rennes, France controlling the operation of electric vehicles i.e., smart charging. Centralized smart charging so- lutions have already been proposed in the lite- rature, but may lack scalability, resilience, and data privacy. Decentralization can help tackle these challenges. In this paper, a fully decentralized smart char- ging system is proposed using the philosophy of adaptive multi-agent systems. The proposed system utilizes multi-armed bandit learning to handle uncertainties in the system. The presen- ted system is decentralized, scalable, real-time, model-free, and takes fairness among different players into account. A detailed case study is also presented for performance evaluation.