Research On Key Node Identification Algorithm In Coupling Demand Of Electric Power Information Network

In recent years,various infrastructure network systems have continuously expanded in scale and increased in intelligence.The large-scale interconnection between the electric power network and the electric power communication network has continuously enhanced the dependence of the production control and automation of the electric power network on the electric power communication network.This mutual dependence between networks improves the efficiency of network cooperation,but also increases the risk of cascading failures in dependent networks as faults propagate between interdependent networks.This article uses complex network theory as a basis to identify key nodes in the electric power communication network,construct a load-capacity cascading failure model for dependent networks,and study the robustness of the dependent network model.The main work of this article is as follows:(1)A multi-attribute key node identification method for the electric power communication network based on game theory-VIKOR is proposed.First,the importance of nodes is quantified from three attributes: topology information,business information carried by nodes,and the influence of power grid sites.Then,the comprehensive weights of each attribute are obtained using the analytic hierarchy process,entropy weight method,and game theory,and the importance of nodes is evaluated using the VIKOR method.The results of the simulation demonstrate that the algorithm is capable of effectively integrating various characteristics of the power communication network,accurately identifying link nodes within the network,and providing a dependable foundation for ensuring the secure operation of the power communication network.(2)Based on the load-capacity cascading failure model of interdependent networks,a oneto-one coupled network model considering edge weights is constructed.The edge load related to the subnetwork and node degree,edge capacity,and load redistribution strategy are defined,and the proportion of nodes in the largest connected subgraph of the network is used as the evaluation index for the robustness of the coupled network.The impact of the network model,coupling method,and network model parameters on the robustness of the coupled network is studied under random and deliberate attack modes,and example analyses are conducted on the IEEE57 node system forming a coupled network.