1.College of Energy and Electrical Engineering, Hohai University, Nanjing 211100, China;2.Economic and Technology Research Institute of State Grid Jiangsu Electric Power Co., Ltd., Nanjing 210008, China

With the large-scale access of massive, decentralized and diverse distributed resources, the virtual power plant (VPP) technology has become an effective method of unified management and efficient regulation for demand-side resources. In order to explore a new mode of VPP multi-variety trading, a multi-VPP electricity-carbon-reserve peer-to-peer (P2P) trading model is proposed, and the conditional value-at-risk theory is used to quantify the potential risks brought by the randomness of renewable energy. Firstly, a VPP aggregation model including photovoltaic units, fuel cells, energy storage, central air conditioning and flexible load is established, and a multi-VPP electricity-carbon-reserve P2P trading method is proposed to realize the interactive sharing of energy, carbon emission and reserve resources. Secondly, a VPP trading model based on the conditional value-at-risk is established to realize the risk measurement of photovoltaic output uncertainty in the VPP. Thirdly, the consensus-based adaptive step size alternating direction method of multipliers is used to solve the constructed model efficiently. Finally, through an example, it is verified that the electricity-carbon-reserve P2P trading mechanism for multiple VPPs can encourage VPPs to participate in the sharing of various resources, while increase their own benefits and social welfare. Conditional value-at-risk also provides an effective decision reference for the trade-off of profit and risk in VPP.