Day-Ahead Capacity Estimation and Power Management of a Charging Station Based on Queuing Theory

Electric vehicles (EVs) have many advantages over vehicles with an internal-combustion engine (ICE), and they have the capability to dominate the vehicles' market in the next years. In order to be charged, EVs require public charging stations that can be two-sided when vehicle-to-grid (V2G) services are indispensable to participate in an ancillary service market and in this case the capacity of the charging station should be estimated. Previous researches in the field of EVs used the queuing theory to model the arrival and departure times of EVs for estimating the V2G capacity. However, they did not consider the constraints of the queuing system. In this paper, a new queuing model is proposed to estimate the day-ahead capacity of a charging station, including balking, reneging, and retrial inside a finite-source queue system and phase-type services. A day-ahead service scheduling method is introduced for charging stations, which includes all constraints in the queuing model. Furthermore, control and management of the electric power at the charging station are carried out by using a nonpreemptive priority queue considering ac Level II and dc Level III chargers. Simulations are performed in three scenarios by using random input data of six types of EVs and the obtained results demonstrate that the proposed approach is able to obtain a realistic and precise estimation of the power capacity of a charging station.