Assessment of electric vehicle charging hub based on stochastic models of user profiles

Abstract

A significant challenge in the electric mobility transition is the planning of proper charging infrastructures to incentivize the use of electric vehicles (EV) and guarantee a reliable charging service to EV users. This paper proposes to model generic EV user profiles (e.g. worktime, commuters, etc.) together with a simulation framework to appropriately assess charging hubs that become undersized due to growing EV demand. First, Gaussian Mixture Models (GMM) of different EV user profiles are developed in order to simulate multiple scenarios of EV sessions per day (N). Second, an algorithm is presented to simulate the occupancy of a charging hub based on two parameters: (1) the number of charging points (P) and (2) the connection time limit (H). Finally, the charging hub assessment is performed according to a metric designed to consider the interests of both the EV user and the charging hub operator, recommending the optimal P for expandable hubs, or the optimal H for limited hubs. Both cases are analysed in the validation section of this work employing a real-world use case. Results validate that the presented methodology can be used by EV charging hub operators to achieve a balance between the exploitation of the charging installation and the satisfaction of EV users