Research On Bidding Strategy Of Grid-side Energy Storage Participating In Spot Joint Market Based On Game Theory

The grid-oriented battery energy storage system is of great significance to promote the consumption of renewable energy and solve problems such as blockage of regional power transmission.As an emerging subject of competitive power market,how does energy storage compete with other decision-making entities and make reasonable quotations to improve its operation economy and promote its participation in the power market has become an important research topic.Under the double opportunities of rapid development of energy storage industry and steady progress of power market reform,starting from the game relationship of grid-side energy storage participating in the spot market,combined with the construction process of China's power spot market,this paper studies the bidding strategy of energy storage participating in the spot market,and constructs the bidding strategy model of energy storage in the primary stage of the spot market and the bidding strategy model of energy storage in the perfect stage.Firstly,the basic theories of cooperative game and non-cooperative game are introduced;combined with the construction process and organizational points of China's spot market,the trading mechanism of grid-side energy storage participating in the spot joint market is designed;starting from the game relationship of grid-side energy storage participating in the spot market,the game relationship and bidding mechanism between energy storage and market entities are systematically analyzed,including the game bidding relationship between energy storage and dispatching/trading center,between market bidding entities,and between energy storage power stations.The game process framework of grid-side energy storage participating in market bidding is given.Then,according to the characteristics of the initial stage of the spot market construction,the grid-side energy storage is regarded as the maker of the market price.Based on the most basic single energy storage,adopt a simple price-quantity bid structure;comprehensively considering the participation in the day-ahead energy market,the day-ahead frequency modulation market and the intra-day balance market,a bilevel game model of energy storage participating in the spot joint market based on the stackelberg game is constructed;the upper-level of the model takes energy storage as the leader to make bidding decisions,and the lower-level takes dispatching/trading center as the follower to complete market clearing.In addition,the introduction of multi-scene description transaction decision game process is the risk of uncertainty to the energy storage income.By using KKT optimality condition and strong duality theorem,the bilevel model is reconstructed into a mixed integer linear programming model and solved.The analysis of calculation examples verifies the feasibility and effectiveness of the proposed model.Finally,the paper further explores the bidding strategy of energy storage participating in the joint market during the perfect stage of the spot market,the energy storage alliance is composed of the energy storage power stations with unlimited location and diversified investors on the grid side to realize the unified scheduling of resources;use the frequency modulation performance mechanism to encourage energy storage to participate in the market,and introduce frequency modulation mileage to ensure the fairness of bidding between different performance resources;by adopting complex bidding structure,a bilevel bidding model of energy storage participating in the spot joint market considering performance is constructed.The market operators perform the technical constraints of energy storage when clearing the market to avoid market speculation.In addition,aiming at the inapplicability of the previously used algorithm caused by the non-convex lower-level model under the complex bidding structure,a reinforcement learning algorithm is introduced to combine with the proposed model to solve the inapplicability problem of the algorithm and simulate the energy storage transaction decision-making process.The results of the calculation example show that taking into account performance can drive energy storage to participate in frequency modulation,especially in the downward direction.At the same time,grid-side energy storage can alleviate or even eliminate the congestion,and the closer to the blocked line,the better.