MRL-Based Model for Diverse Bidding Decision-Makings of Power Retail Company in the Wholesale Electricity Market of China

Abstract

Power retail companies in the electricity market make profits through buying and selling power energy in the wholesale and retail markets, respectively. Traditionally, they are assumed to bid in the wholesale market with the same objective, i.e., maximize the profit. This paper proposes a multiagent reinforcement learning (MRL)-based model to simulate the diverse bidding decision-making concerning various operation objectives and the profit-sharing modes of power retail companies in China’s wholesale electricity market, which contributes to a more realistic modeling and simulation of the retail companies. Specifically, three types of operation objectives and five types of profit-sharing modes are mathematically formulated. After that, a complete electricity market optimization model is established, and a case study with 30 retail companies is carried out. The simulation results show that the proposed method can effectively model the diverse bidding decision-making of the power retail companies, which can further assist their decision-making and further contribute to the analysis and simulations of the electricity market.

Keywords

• multiagent reinforcement learning
• bidding strategy
• decision making
• electricity market
• power retail company
• electricity market simulation