A Dynamic Social Network Matching Model for Virtual Power Plants and Distributed Energy Resources with Probabilistic Linguistic Information

Abstract

Virtual power plants (VPPs) offer an effective means to address the imbalance issue between electricity supply and demand to advance the world’s low-carbon development. To fully utilize the limited resources in the virtual power plant planning stage, a two-sided match between VPPs and distributed energy companies is needed to better implement resource aggregation management. Because of the vagueness in this matching environment, the probabilistic linguistic term set (PLTS) is necessary to apply to express the decision makers’ preference. Considering the complex social relationships and intense competition among companies, a dynamic social network two-sided matching model is proposed for solving the multi-attribute two-sided matching decision-making problem. Firstly, we present a matching satisfaction degree described by PLTS. A dynamic social trust degree based on the sliding time concept is proposed. Secondly, the social trust network relationships are built based on the direct and indirect dynamic trust degree among companies. This relationship is then combined with an improved trust rank algorithm to identify the most authoritative and the most trusted company to provide the target company with a recommendation for the next moment. Besides, given that companies compete for limited resources, we further define the competitive satisfaction degree and apply the two-sided matching model. Additionally, then a two-sided matching model is developed. Finally, our model is tested numerically to ensure its accuracy and reliability.