Multi - stage Stochastic Optimization Scheduling of Service Area Virtual Power Plant Considering Source and Load Uncertainty

This paper addresses the need for a low-carbon economic transformation and green and the green development of expressway service areas by constructing a virtual power plant (VPP) model. This model incorporates electric vehicle charging piles, photovoltaic power plants, traditional gas turbine power plants, and energy storage systems. And then, a multi-stage stochastic optimization dispatch model is proposed for day-ahead, intra-day, and real-time operations.The research begins by analyzing the development models and the assumptions underlying virtual power plant modethe assumptions underlying the virtual power plant model. To tackle the VPP’s optimal dispatch problem, a multi-stage stochastic optimization model is developed aimed at minimizing operation costs. The uncertainties associated with electric vehicle charging loads and photovoltaic output were modeled through the Markov Chain-Monte Carlo algorithm, which generates typical scenarios for simulation analysis. The proposed model’s validity is confirmed through a case study of a highway service area in Xinjiang. The results indicate that the model effectively lowers the operating costs of the service area, enhances the market competitiveness of the VPP, and improves the flexible operation capabilities of the distribution network.