A Performance Model for Single Charger Multi-Socket Charging Stations in Shared Mobility Systems

Abstract

It is commonly recognised that the availability of more pervasive networks of public fast-charging stations is a key incentive for the market growth of electric vehicles (EVs). In order to reduce the huge capital investments that are needed for the deployment of such charging infrastructure, new types of charging facilities have been proposed, which allow to charge several EVs simultaneously by sharing the resources of a single charger over multiple co-located charging sockets. To characterise the behaviour of a single-charger multiple-socket (SCMS) system under stochastic EV charging demands, in this work we propose a continuous time and discrete state space Markov chain model. Our analytical model applies to scenarios in which the duration of charging periods is uncertain. This typically occurs in shared mobility systems (e.g., car sharing services) in which customers randomly arrive at the charging station to pick up available shared EVs. We examine two scenarios of increasing complexity. In the first one, customers can pick up only fully-charged EVs. In the second scenario, EVs can leave the station before the charging process is complete. Our numerical results assess the impact of station capacity (both physical space and grid connection) on the system performance from the perspective of both the customers and the infrastructure owner.