Optimal Dynamic Allocation and Space Reservation for Electric Vehicles at Charging Stations

Abstract

We propose an optimal allocation and reservation system for Electric Vehicles (EVs) at charging stations distributed in an urban environment. The system assigns and reserves an optimal space at a charging station based on the user's cost function that combines proximity to current location (or destination) and charging cost. Our approach is motivated by a similar system we have developed for “smart parking”, where resources are parking spaces rather than EV charging station spaces. We solve a Mixed Integer Linear Program (MILP) problem at each assignment decision point over time. The solution of each MILP is an optimal allocation based on current state information, and is updated at the next decision point. Formal guarantees are included that there is no resource reservation conflict and that no user is ever assigned a resource with a higher than this user's current cost function value. Simulation results are included to illustrate how our system, compared to uncontrolled processes or guidance-based approaches, reduces the average time to find a charging space and the associated user cost, while the overall charging space capacity is more efficiently utilized.