A Demand Side Response Strategy Considering Long-term Revenue of Electricity Retailer in Electricity Spot Market

In order to maximize the long-term revenue of electricity retailer in electricity spot market, a dynamic optimization scheme of demand response based on reinforcement learning is proposed. Firstly, a demand response model of retailer and users is established. By establishing the pre- and post-comfort cost function for the users, a dynamic optimized revenue function for the electricity retailer is constructed. The response loads are determined by the user's demand response income function. And then the revenue function of the retailer's current demand response is transformed into an immediate reward function. The Q function of reinforcement learning is constructed with BP neural network, and the BP neural network is trained with iterative method until the Q function is converged. The simulation results of one electricity retailer and five industrial users show effectiveness of the proposed method. © 2019, Power System Technology Press. All right reserved.