Virtual Power Plant Operation Based on Entropy-Weight-Improved Cloud Risk Assessment Model

Many virtual power plant demonstration projects are currently in operation in China, but research on the risk assessment of virtual power plants is insufficiently deep. Virtual power plants are complex systems operated by multiple participants and can provide a variety of products or services. Based on the operation practices of virtual power plants, in this study, we considered the characteristics of different components of virtual power plants and their impact on the operation process. We developed a risk assessment model for virtual power plants from many aspects involved in their operation processes. First, we analyzed the operation modes that can be selected during the operation of a virtual power plant, including the scenario in which electric vehicles participate and considering the comprehensive demand response. Second, we considered the risk faced by a virtual power plant from five dimensions: external policy, participation, coupling technology, bidding transaction, and credit management risk; we then designed an operation risk indicator system for virtual power plants with 29 indicators. Third, based on the entropy weight order relation method and cloud model for determining the index uncertainty in the process of risk assessment, we developed a cloud risk assessment model and specific algorithm flow based on the entropy weight order relation method are proposed. Finally, we compared the optimal economic operation strategies of a virtual power plant under different operating characteristics. The results showed that the comprehensive risk in the operation of gas virtual power plants can be effectively reduced when considering various uncertainties, electric vehicles, and comprehensive demand response risk. The results verify the effectiveness and scientificity of the improved cloud model risk assessment model proposed in this paper.