Price-based demand response in the coupled power and transportation network via EV charging station

The increasing adoption of electric vehicles (EVs) presents an opportunity for demand response (DR) programs in the power network to improve the overall operation efficiency. However, in the transportation network (TN), the potential of EV charging station (EVCS) participating in DR needs to be better understood. In this paper, we propose a price-based DR strategy for the coupled power and transportation network through EVCS. The model is built as a three-layer framework that encompasses the power grid, EVCS, and EV users. By ensuring information security and independent operation within each layer, this framework aims to expand overall interests and achieve a mutually beneficial outcome. To address the challenge of inherent computational complexity in the proposed model, we employ deep reinforcement learning (DRL) and utilize multi-agents iterative training process to obtain optimal decision-making policy. The effectiveness of our proposed method is validated through a real-world test system.