Identification And Vulnerability Analyis Of Key Nodes In Power Optical Network

Power optical network is a part of power system and communication system.With the increasing demand of power grid development,power optical network not only carries a lot of power business,but also contains the corresponding communication business,and its position in the whole power grid system is becoming more and more important.However,the network performance of power optical network is more diverse and the structure is more complex,so the network is more prone to collapse after failure.In this thesis,the structure,state,correlation and other complex characteristics of the power optical network are studied deeply,and the static parameters of the network are analyzed,so that the internal structure of the power optical network can be understood more clearly.By analyzing node importance with multidimensional indicators,key nodes can be accurately identified,and vulnerable sources can be protected in advance to improve the ability of network to resist failures.By evaluating network vulnerability through multidimensional indicators,the effectiveness of the key node identification algorithm can be verified,and the impact of node failure on network vulnerability can be clarified.The key node identification and vulnerability analysis of the power optical network in this thesis can effectively analyze the network performance,accurately identify vulnerable sources and protect them in advance,improve the ability of the network to resist faults and avoid network crashes,and provide useful guidance for the optimization planning of the power optical network.Current research on the identification and vulnerability of key nodes in power optical networks has shortcomings such as a single evaluation index and failure to consider the strong power characteristics of the network.In this thesis,a key node identification model based on long-term and short-term comprehensive importance is proposed.This model is evaluated from two aspects:power optical network topology structure and network power characteristics.The main research contents include the following three aspects:(1)Take the network topology as the research objective,calculate the static parameters of the network according to the complex network theory,and construct the long-term importance of nodes.It combines local and global information of nodes and uses subjective and objective multidimensional fusion algorithm for fusion to obtain the long-term importance of nodes.Through experimental simulation,after comparing and deleting high-importance nodes obtained by different algorithms,the changes of network in three dimensions are analyzed.Compared with other algorithms and a single index,node long-term importance can accurately identify key nodes in network topology.(2)Taking the characteristics of the power industry as the research object,combined with the power background,and considering the site,traffic and business related information in the real power optical network,the short-term importance index of nodes is constructed.This index describes the importance of nodes in the power optical network,a power system related network.Then,combined with the long-term importance of nodes,the entropy weight+approximate ideal solution algorithm was used to obtain the long-term and short-term comprehensive importance of nodes.By synthesizing the importance of nodes in the context of topology and power,a more comprehensive and accurate key node identification model was formed.(3)In order to carry out a comprehensive vulnerability analysis on the power optical network,by deleting nodes of high importance successively,observe and compare the changes of network loss function,loss entropy and node failure influence under different algorithms,and verify that the proposed algorithm can more accurately identify key vulnerable sources of the network and analyze network vulnerability in multiple dimensions.