Borderline Edge Mining And Edge Addition Strategy Of Power Optical Fiber Network Based On Community Structure

The power fiber network is a network that supports the power network and has the same scale as the power network.When some lines in the power optical fiber network are damaged,it will cause cascading failures,and even lead to the paralysis of the entire network.It is of great significance to analyze the line vulnerability of the power optical fiber network,improve the connectivity and efficiency of the entire network to ensure the stable operation of the power optical fiber network.Therefore,this paper constructs a power fiber network model based on the complex network theory based on the physical topology of the actual provincial power fiber network.Firstly,the community division of the power optical fiber network model is carried out.Then,identify the important lines that play a key role in the safe and stable operation of the power grid,and implement differentiated management and operation and maintenance.Finally,a reasonable edge is added to ensure the operation reliability of the vulnerable lines of the power grid.The main work is as follows:（1）In order to make up for the shortcomings of relying only on administrative divisions to manage the power optical fiber network,under the community structure,the links have a major impact on the network.Therefore,a community partition algorithm for power optical fiber network is designed.Firstly,extract the backbone network from the constructed power fiber network model to obtain the initial community,define and calculate the local edge centrality（）of all edges in the network.Secondly,delete the edges whose value is greater than the threshold to obtain the first community division.Finally,the small communities are merged according to the edge density to determine the network partition.Through the experimental analysis of Jilin Province’s power optical fiber network,it is found that the divided regions are all connected internally,the overall toughness is increased by 60% compared with the division method according to the administrative region,and the community structure is more obvious.Compared with the other four community partitioning algorithms,the algorithm ILEC also achieves good results in network efficiency and modularity.（2）In order to reasonably protect the network,identify the weak edge structure of the network,and design a borderline edge mining algorithm for the power optical fiber network.First of all,for the communities divided by the power fiber network model,the topological reliability metrics of each community are calculated,and the edge set between communities is obtained.Then,based on network heterogeneity,an H-index based fragile edge metric is proposed.Finally,considering the and the topological reliability measure of the community,the set of borderline edges inside the community is obtained.Integrate and sort with the borderline edges between communities to get the borderline edge se tof the entire network.Through the experimental analysis of the power optical fiber network in Jilin Province,it is found that,compared with the other four algorithms of edge mining,the algorithm reduces the network connectivity by 37.6% and the network efficiency by 46.95% on average after removing borderline edges.It can more effectively dig out the vulnerable edges of the network.（3）Aiming at achieving the optimal connectivity and network efficiency at the same cost or low cost,an edge-adding strategy for power optical fiber network is designed.Firstly,based on the borderline edge set of the power fiber network model,the edge node set is obtained,and the node vulnerability measure is calculated according to the measure.Secondly,to avoid double edges,adding the maximum value node of of the edge node set to the node with the maximum degree of the full set except the borderline edge set.Through the experimental analysis of Jilin Province’s power optical fiber network,it is found that the average shortest distance of network efficiency increased by 17.49%.At the same time,the efficiency increased by 13.71%.Thus,the effectiveness of the edge addition strategy is confirmed.