Two-Stage Admission and Scheduling Mechanism for Electric Vehicle Charging

Abstract

In order to provide an efficient charging service to future electric vehicles (EVs), it is important for the charging station to estimate its available capacity. As the charging station usually has a fixed number of chargers and a limited charging capacity, it cannot accommodate all the arriving cars. Thus, a decision has to be made by the charging station whether a new arriving car can be admitted and then be scheduled for charging. This not only depends on the charging station’s physical constraints but also depends on how certain the charging station is toward its energy supply. We study the problem of EV charging using various energy sources (such as energy from the grid, renewable energy such as solar photovoltaic panels, and local energy storage) and their impacts on the charging station’s performance. Unlike the energy from the grid, solar generation varies with time and cannot be predicted precisely. To this end, we introduce a multiplier ${k}$ to measure the effect of solar prediction error and a composite performance index (the figure of merit) to capture the charging station’s utility, also taking account of EVs’ charging requirements. We further propose a two-stage admission and scheduling mechanism to find the optimal trade-off between accepting EVs and missing charging deadlines by determining the best value of ${k}$ under various energy supply scenarios.