Quotation Model of Energy Storage participating in Electric Power Day-Ahead Market based on Deep Learning Surrogate Model

摘要

Energy storage has been paid more and more attention in China and will become an important part of future power system and power market. How to incorporate the energy storages in the day-ahead market so as to maximize the economic benefits of both energy storages and the whole market has become an urgent problem to be studied. In this paper, the bidding model of energy storages under the current electric power day-ahead market rules (taking Guangdong's rules as an example) is established firstly. Then a new trading mode of energy storage participating in day-ahead market based on surrogate model is suggested, and a both model-driven and data-driven surrogate modeling method for energy storage participating in day-ahead market quotation is proposed. And further, a multi-objective optimization based market clearing model with the proposed surrogate-model is built and solved by NSGAII method. Case studies are accomplished to verify the advantages of the proposed trading mode and the feasibility and effectiveness of both surrogate modeling method and the market clearing model.