Analysis of virtual power plant transactions based on distributionally robust optimal scheduling

Abstract Virtual power plants aggregate resources such as distributed power sources, energy storage systems, and controllable loads through digital means, promote the matching between the demand side and the supply side, and promote the large-scale consumption of renewable electricity. This paper introduces the operation mode of virtual power plants, including direct power purchase and sales transactions with external large power grids and its role as an auxiliary service provider for peak shaving, and expounds the key supporting technologies of virtual power plants, such as state awareness and flexible aggregation, information prediction and capacity estimation. The characteristics and mathematical models of the robust optimization method are analyzed in detail, including single-stage classical robust optimization, two-stage classical robust optimization, and two-stage distributed robust optimization model. The results of the combined robust optimal scheduling of heat and power in the virtual power plant and the distributed robust optimal scheduling under diversified flexible loads are demonstrated through the case. The results show that by adjusting the load invocation cost and optimizing the scheduling strategy, the utilization efficiency of flexible load can be significantly improved, the curtailment of wind and solar power can be reduced, the operating cost can be reduced, and the total revenue can be increased.