A Dynamic Business Model for Battery Swapping Stations with Pricing and Charging Scheduling

To mitigate carbon emissions and achieve carbon neutrality, developing electric vehicles (EVs) has proven to be an effective strategy. Battery swapping stations (BSSs) are emerging as efficient facilities for replenishing EV batteries. While existing research has predominantly concentrated on scheduling problems related to battery charging within BSSs, there remains a notable gap in developing effective models for pricing as well as scheduling battery swaps. In this study, we investigate the pricing mechanisms of a single BSS operating within a system where the power grid engages in electricity trading, subject to load constraints and dynamic, price-sensitive swapping demand. By employing a deep reinforcement learning approach within a Markov Decision Process framework, we aim to optimize real-time battery charging, discharging, and pricing decisions, which can enhance the overall profit of the BSS compared to fixed-price models while balancing the charging demand.