Optimal Aggregation Method of a Virtual Power Plant Based on a Nash-Stackelberg Game

The aggregation of massive heterogeneous distributed energy resources (DERs) is a crucial challenge in the optimal operation of a virtual power plant (VPP). To effectively exploit the flexibility and economy of DERs, this study proposed a novel method to aggregate DERs belonging to different stakeholders into a VPP based on a Nash-Stackelberg game. The DER-based generators act as leaders in the game and decide their output to maximize profits, while the VPP acts as a follower on the lower level to maximize the profit in the market. Moreover, the network constraints and uncertainties are considered in the lower model. Subsequently, a method to find the Pareto-optimal equilibrium points is proposed. The bi-level model can be transformed into a single-level model via the Karush-Kuhn-Tucker conditions. Thus, the proposed Nash-Stackelberg game model is transformed into a non-cooperative multi-objective optimization problem. Furthermore, the Nash equilibrium point can be obtained by the modified column-and-constraint generation algorithm. Finally, case studies demonstrated that the method can efficiently find the Nash equilibrium solution of the VPP aggregation model and provide more comprehensive Nash equilibrium points than other methods. Moreover, the distribution of Nash equilibrium points can be used to guide the dispatching scheme of the VPP.