An Extended M/M/K/K Queueing Model to Analyze the Profit of a Multiservice Electric Vehicle Charging Station

Abstract

Recent trends in plug-in electric vehicles (PEVs) include the investment in public charging stations (PCS) construction, wherein PEVs can charge their batteries, and PCS owners can make profits. Scheduling algorithms used in the PCS can optimally assign PCS resources to the vehicles in order to maximize their profit. Although the main purpose of the PCS is to provide charging capability, it needs to address other customer requirements such as different deadline constraints, battery swapping, and vehicle to grid (V2G). In this paper, we introduce a comprehensive model for multiservice PCS profit maximization in which battery charging, discharging, and swapping services are included. The model has been used in formulating a mixed-integer linear programming (MILP) problem wherein charging and discharging of vehicles are scheduled to maximize the PCS profit. We utilize queuing theory to analyze the performance of the model at the presence of random variables such as solar power, electricity price, state of charge (SoC), amount of discharge (AoD), battery capacity, and vehicle arrival time to the PCS. Low complexity and near-optimal online scheduling algorithm is proposed to complete the offline scheduling solution. Our simulation results show that more profit for the PCS and less delay for vehicles are better achieved when the number of vehicles requesting battery swapping service is increased. Extensive simulations are conducted to evaluate the impact of different parameters on long-term profit and the performance of the online algorithm.