Research on optimization of pricing strategies for inter and intra provincial electricity spot markets based on deep reinforcement learning

As the construction of inter provincial markets becomes increasingly mature, strategic generation companies (GenCo) need to consider the impact of inter provincial spot markets, collaborate to optimize their pricing strategies in the two-tier market, and improve profits. This article uses deep reinforcement learning to construct an intelligent agent, simulating power generators participating in inter provincial spot market as price receiver and attend intra provincial spot market transactions as price setters. By using the deep deterministic policy gradient algorithm, the load forecasting data during the bidding period and the node electricity prices during the competition process are taken as state variables, and their declared electricity prices and retention amounts in the provincial spot market are used as bidding strategies. This enables the agent to gradually learn the environment during the bidding process and optimize its own bidding strategy based on profits. The case study validates the effectiveness of the model proposed in this paper and can improve the bidding revenue of GenCos.