GraphVPP: Enabling Optimal Bidding Strategy of Virtual Power Plants in Graph Computing Frameworks

This paper proposes the GraphVPP, a graph computing framework for developing the optimal bidding strategy of virtual power plants (VPPs) participating in the wholesale electricity market. Compared with existing literature that only position the VPP as a commercial unit, GraphVPP considers both the economic operation of distributed energy resources (DERs) within the VPP and their technical impacts towards the network security of transmission and distribution systems. First, GraphVPP stores the power grid data and market information in a graph structure, and then formulates a two-stage optimization problem for deriving the market clearing results including local marginal prices (LMPs) and generation dispatches. To ensure the power grid’s stability under the scheduled generation dispatches, a graph based power flow algorithm is introduced for the network security analysis. Once the network security check is approved, a profit-maximization bidding decision model is introduced for each VPP to derive the optimal bidding strategy. Finally, case study compares different types of VPPs within the wholesale electricity market to demonstrate the feasibility and validity of the proposed GraphVPP.