A Multilevel Edge Computing Architecture and Edge Generation Method of Distribution Networks

With the development of ubiquitous power Internet of things in the power system, in order to satisfy the intelligent and refined requirements of management in distribution networks and alleviate the increased communication and calculation burden of master stations, a multilevel edge computing architecture and edge generation method of distribution networks are proposed. First, a multilevel edge computing architecture of distribution networks based on edge computing and hierarchical distributed control is proposed. Then, the principles of edge generation are explained from three aspects: size, communication burden, and autonomous ability. Finally, based on community detection, a distribution network is abstracted into a weighted undirected graph model and divided into edges by the Louvain algorithm. The case analysis shows the validity and rationality of the proposed edge generation method.