A Decentralized Multi-Agent System for Management of En Route Electric Vehicles

Being energy producers and consumers (prosumers), electric vehicles introduce new challenges to the operation of the electricity grid. With the anticipated increase of the market for electric vehicles in the near future, systems that set rules to manage and organize their large-scale penetration into the power grid are needed. As the Internet of Things (IoT) vision evolves, the Internet of Electric Vehicles (IoEV) shall be a stepping-stone for establishing a link between the well-known hierarchical smart grid control architecture (primary, secondary, and tertiary control) and the emerging layer of the EVs as dynamic and mobile energy prosumers. Against this background, this paper proposes a decentralized multi-agent system designed for the management of en route connected electric vehicles. In this framework, each electric vehicle and electric vehicle park is equipped with an intelligent agent. These agents interact with a higher-level agent responsible for managing charging requests within their geographical area. The importance of this multiagent framework is demonstrated in a case study, where electric vehicles are routed to the most suitable electric vehicle charging parks in their vicinity in a manner that satisfies the charging preferences of customers, with the lowest cost possible, while ensuring that the generation-load balance at the EV parks is not violated.