Mining Vulnerable Entities In Power Communication Network Based On Complex Network

The mining of vulnerable entities of power communication network mainly measures the impact on the vulnerability of power communication network when entities such as communication stations and optical cables suffer deliberate attack or failure in normal operation.Based on the complex network theory,this paper constructs a weighted power communication network model for the physical topology of prefecture-level power communication network,comprehensively considers the network topology and business characteristics,designs the mining algorithm of Vulnerable entities of power communication network,and analyzes the impact of Vulnerable entities on the Vulnerability of power communication network.Firstly,the characteristics of various connection relationships between communication stations and optical cables in the actual prefecture level power communication network are analyzed.According to the gap between communication stations and communication stations,optical cables and optical cables and different connection relationships between communication stations and optical cables,the weight distribution is carried out,and the weighted power communication network model is constructed.Then,using the complex network characteristic indexes: node degree,betweenness,clustering coefficient,average path length,small world and scale-free characteristics,this paper makes an empirical analysis and Research on the complex network characteristics of the constructed power communication network model,and finds that the prefecture-level power communication network has the characteristics of small world and scale-free.Secondly,aiming at the problem of inaccurate mining of vulnerable nodes without considering the relationship between nodes in the mining of vulnerable nodes in power communication networks,an algorithm for mining vulnerable nodes in power communication networks based on improved structural holes is proposed.The algorithm considers the efforts made by nodes to maintain neighbor nodes in the idea of improving structural holes,and designs constraint coefficients to measure the degree of local influence of nodes.Using the influence coefficient matrix to combine the betweenness value and the constraint coefficient value of the nodes,an improved structure holes is designed as the node vulnerability value measure to mine the vulnerable nodes in the power communication network.Finally,the mined vulnerable nodes set is regarded as a deliberate attack on the power communication network,and the vulnerability evaluation index is used to verify the rationality of the proposed algorithm.The vulnerability evaluation index shows that compared with the other four algorithms,this algorithm improves the maximum number of split blocks under susceptibility is increased by 4.2% on average.When the top-15 nodes are removed,the sensitivity index increased by 3.6% on average,the network efficiency index increased by 23.5% on average,and the largest connected subgraph index increased by 52.87% on average,which can effectively mine vulnerable nodes in the power communication network.Finally,in view of the problem of ignoring the interaction between vulnerable nodes and vulnerable links in the mining of vulnerable links in power communication networks,which leads to deviations in the mining of vulnerable links,an algorithm for mining vulnerable links in power communication networks based on improved degree product and edge betweenness is proposed.The algorithm incorporates the node’s vulnerability value into the link’s degree product formula to measure the link’s local vulnerability.Then,considering the position of the links and the role of information transmission in the transmission process of the power communication network,the degree product and edge betweenness is designed as the link’s vulnerability value measure to mine the vulnerable links in the power communication network.Finally,the mined vulnerable links set is regarded as a deliberate attack on the power communication network,and the vulnerability evaluation index is used to verify the rationality of the proposed algorithm.The vulnerability evaluation index shows that when this algorithm and the other two algorithms remove the top-20 links,the number of network split blocks of the sensitivity index increased by an average of 25%.The network efficiency index increased by 16.75% on average,and the largest connected subgraph index increased by 18.3% on average,which can effectively mine vulnerable links in the power communication network.