Market game model construction method based on reinforcement learning algorithm and considering V2G participation

The invention discloses a market game model construction method based on a reinforcement learning algorithm and considering V2G participation. The method comprises the following steps: constructing an upper-layer V2G income model with a plurality of decision variables and constraint conditions; and constructing a lower-layer market liquidation game model based on reinforcement learning, obtaining a bidding strategy obtained by a power generator in an optimal decision of a decision-making layer through the upper-layer V2G income model, and taking the bidding strategy as input data of the lower-layer market liquidation game model to obtain a power market lower-layer game model considering V2G participation. Market income maximization is realized through the power market lower-layer game model considering V2G participation; and based on an improved WoLF-PHC multi-agent reinforcement learning algorithm, optimizing a power market lower-layer game model considering V2G participation. The V2G pricing and power resou