A Dynamic Algorithm for Facilitated Charging of Plug-In Electric Vehicles
Plug-in Electric Vehicles (PEVs) are a rapidly developing technology that can reduce greenhouse gas emissions and change the way vehicles obtain power. PEV charging stations will most likely be available at home and at work, and occasionally be publicly available, offering flexible charging options. Ideally, each vehicle will charge during periods when electricity prices are relatively low, to minimize the cost to the consumer and maximize societal benefits. A Demand Response (DR) service for a fleet of PEVs could yield such charging schedules by regulating consumer electricity use during certain time periods, in order to meet an obligation to the market. We construct an automated DR mechanism for a fleet of PEVs that facilitates vehicle charging to ensure the demands of the vehicles and the market are met. Our dynamic algorithm depends only on the knowledge of a few hundred driving behaviors from a previous similar day, and uses a simple adjusted pricing scheme to instantly assign feasible and satisfactory charging schedules to thousands of vehicles in a fleet as they plug-in. The charging schedules generated using our adjusted pricing scheme can ensure that a new demand peak is not created and can reduce the consumer cost by over 30% when compared to standard charging, which may also increase peak demand by 3.5%. In this paper, we present our formulation, algorithm and results.