The transition to electric vehicles (EVs), coupled with the rise of renewable energy sources, will significantly impact the electric grid. Unlike conventional fuel sources, electricity for EVs is constrained by grid capacity, price fluctuations, and long EV charging times, requiring new pricing solutions to manage demand and supply. This paper proposes a model for online dynamic pricing of reserved EV charging services, including reservation, parking, and charging as a bundled service priced as a whole. Our approach focuses on the individual charging station operator, employing a stochastic demand model and online dynamic pricing based on expected demand. The proposed model uses a Markov Decision Process (MDP) formulation to optimize sequential pricing decisions for charging session requests. A key contribution is the novel definition and quantification of discretization error introduced by the discretization of the Poisson process for use in the MDP. The model's viability is demonstrated with a heuristic solution method based on Monte-Carlo tree search, offering a viable path for real-world application.

翻译：向电动汽车的转型，结合可再生能源的兴起，将对电网产生重大影响。与传统燃料不同，电动汽车的电力供应受到电网容量、价格波动以及较长的充电时间的限制，需要新的定价方案来管理供需。本文提出了一种针对预约式电动汽车充电服务的在线动态定价模型，该模型将预约、停车和充电作为整体定价的捆绑服务。我们的方法聚焦于单个充电站运营商，采用随机需求模型和基于预期需求的在线动态定价。所提出的模型使用马尔可夫决策过程（MDP）框架来优化针对充电会话请求的序列定价决策。一个关键贡献是对为用于MDP而对泊松过程进行离散化所引入的离散化误差进行了新颖的定义和量化。通过基于蒙特卡洛树搜索的启发式求解方法，证明了该模型的可行性，为实际应用提供了一条可行的路径。