Internal and external coordinated distributionally robust bidding strategy of virtual power plant operator participating in day-ahead electricity spot and peaking ancillary services markets

Virtual power plant operators (VPPO) must consider external markets and internal members' coordination issues when bidding decisions and minimize the loss of benefits from wind and PV uncertainty. This study first clarifies the internal and external coordinated distributionally robust (DR) bidding decision process for VPPO participation in the day-ahead electricity spot and peaking ancillary services markets. Secondly, a fuzzy set based on the Wasserstein distance for determining the forecast error of wind and photovoltaic output was used to establish a two-layer optimization model for the VPPO internal and external coordinated DR bidding decision. The upper level is the VPPO external market DR bidding model, and the lower level is the master-slave game bidding model with the VPPO as the leader and controlled distributed power, flexible load, and energy storage (ES) as the followers. Finally, the genetic algorithm with elite strategy and Gurobi solver combining method was used to optimize the bidding strategy of VPPO. The analysis of the algorithm shows that the proposed method gives an optimized solution for VPPO's bidding in the external market, and the interests of both VPPO and internal members are enhanced at the same time. The comparative analysis of multiple scenarios found that wind power forecast error has a greater impact on VPPO's profit than PV. When the unit cost of ES drops to a certain level (200-300 yuan/MW & sdot;h), the cost of ES has less impact on the VPPO. The price of the day-ahead electricity spot market had a tremendous impact on VPPO's profits, and when the price of electricity fell by 15%, VPPO's profits fell by 38.63 %, and VPPO's use of ES declined dramatically.