Dynamic pricing at electric vehicle charging stations for waiting time reduction

Abstract

The load balancing of electric vehicle (EV) charging at public charging stations has gained increasing attention in recent years. In reality, a majority of EV drivers choose to charge at the nearest public charging stations. This incurs uneven utilization ratio and longer waiting time at popular stations. This substantially reduces the QoS performance of this public service. Dynamically balancing the charging load thus is an important and challenging problem and yet remains unsolved. In this paper, we propose a novel dynamic pricing approach that allows charging stations to adjust their service fees in real time based on their charging load, which in turn encourages drivers to switch to farther while less crowded and cheaper stations for charging. We develop a notification system for drivers to receive valuable information about all nearby charging stations including electricity price, service fee, estimated waiting time and time to arrive there. Our solution is evaluated through a real-world dataset, called e-charge, which includes 1,851 charging stations in Beijing, China. Experimental results demonstrate that our approach reduces the average waiting time by 21.1% compared to not using the approach. Our work will further shed lights on general dynamic load balancing techniques in cost-sensitive human-in-the-loop IoT applications.