Research on Key Node Evaluation Algorithm of Multi-index Based on Betweenness Centrality-Structure Holes

Abstract

The power optical network has the characteristics of complex structure and diversified power business. The safe operation of the power grid puts forward higher requirements for the survivability of the power optical network. In order to more accurately identify the key nodes in the optical network. This paper comprehensively considers the impact of the global importance and local importance of nodes on network performance. Based on the betweenness centrality, paper considers the structural hole index that shows local characteristic attributes of nodes. Based on betweenness centrality -structure holes, the paper innovatively proposed a multi-index key node evaluation algorithm. This algorithm uses the entropy method to quantify the amount of data information of betweenness centrality and structural hole indicators, which assigns weights to them according to the degree of index differentiation. On the basis of this, algorithm taking into account the impact of business importance, obtains a comprehensive evaluation index for node importance identification. In the simulation process, paper compares with other algorithms to analyze the impact of the network topology layer to the power grid after the node fails. It analyzes the vulnerability of the network by using the global network efficiency and the maximum sub-graph connectivity rate, and integrate the business importance to analyze the impact of the network topology and the business multi-layer to the power grid after the node fails. Paper use CVE (Cross Layer Vulnerability Entropy) to describe the comprehensive loss of the network. Then paper uses the distinguishability of the node importance value as an index to quantify the degree of granularization of the node sorting results. Finally, the accuracy of the algorithm is verified according to the changes in the vulnerability of the network and the comprehensive loss.