With the increasing proportion of electric vehicles and distributed power supplies connected to the power grid

virtual power plant (VPP) provides a new way to effectively solve the problem of electric vehicles and distributed power supplies connected to the power grid. Aiming at the challenges such as high operating costs

large uncertainties in electricity prices and source-load when VPP operates independently

this paper proposes an optimization method for the collaborative operation of multi-VPP based on Nash three-stage robust optimization. In order to coordinate the conflict of economic interests between the VPP internal operator and the EV subject

the master-slave game theory is used to depict the interaction between the VPP system operator and the upper and lower levels of the EV. Meanwhile

the upper VPP operator fully considers the influence of uncertainty caused by the fluctuation of purchase and sale price and source load output in the power market. A three-stage robust optimization model is used to construct the upper body. The three-stage robust optimization model is different from the traditional model. The min-maxmin-maxmin construction is used to describe the internal relationship of the model. A multi-VPP group optimization operation model based on Nash negotiation theory is constructed. In order to solve the complex nonconvex nonlinear optimization problem

the model is transformed into two sub-problems: multi-VPP benefit maximization and power payment value between multiple VPPS. Considering the information privacy security between VPPS

alternating direction method of multipliers (ADMM) is adopted to solve the above two sub-problems. Three multi-functional VPPS are selected for example analysis

the results show that the proposed optimization method in this paper can not only establish the optimal dispatching scheme for VPP under the influence of multiple uncertainties but also improve the profits of each VPP and the overall revenue while taking fairness into account.