The Cascading Failure Model And Robustness Analysis Based Complex Network Theory

In real life, many real presence networks, such as transportation networks,mobile networks, the Internet networks, electronic communication networks, etc.,are connected to each other because they are connected to each other. And the failure of the network system is often due to the failure of one or some nodes in the network to be attacked, at the same time the failure of the node will trigger the network and other nodes between the network failure, and then lead to failure In the interdependent association to the network to spread and spread, and ultimately may lead to a large area of the node failure, and even lead to the collapse of the entire network system. In order to prevent and protect the cascading failure in the real network system to bring great disaster, reduce the loss of social and economic property, for this reason, this paper has a series of related problems on the network,from the network dynamic point of view the use of Matlab simulation software to establish the corresponding This paper discusses the dynamic process and robustness of the associated fault of the connected directional network, and puts forward the local and global preventive measures to resist the cascading failure of the associated network. From the perspective of management, the practical A Comparative Study of Robustness between American West Power Grid and Server Networks. The main contents of the text are as follows:(1) Based on the research of cascading failure on complex networks, a capacity-load cascading failure model based on the proportion of load capacity reallocation on a single-layer network based on node free capacity is proposed from the perspective of network dynamics. Based on the model of node degree distribution mechanism, the model is derived from the mathematical model, which is more optimized and close to the reality than the traditional model,and the successive fault models on the single layer network are extended to the associated directional network.(2) Based on three interconnected modes, two kinds of interconnected dependent directed network models with different topologies are established by using BA scale network and ER random network. In the Matlab simulation process, in order to study the successive fault phenomena on the network, we deliberately attack the network with the largest node, and use the network connectivity normalization under different load control intensity a and different average degree <k>.The index G, the failure node normalized index F and the key threshold Tc, and analyzes the robustness of the network to resist the successive faults from the perspective of the local and global nature of the network and gives the corresponding prevention and protection strategies. Through Matlab simulation, it can be seen that under different topologies,the correlation network has different local and global robustness under different load intensity or average degree. Therefore,in the actual associated network,depending on the situation to determine the value of different parameters, so as to get the best robustness.(3) The cascading failure model is applied to the USA western power grid and server network to analyze how to effectively deal with the failure of the actual connected network,and put forward the countermeasures to prevent and protect the network system, Power grid and server network and related real network system to provide a reasonable basis for the prevention and control of disaster in the network system on a large area of the spread and spread, reduce network construction costs,the rational allocation of networkresources and effective maintenance of the network and more management Meaning and practical application value.