复杂网络上的相继故障模型及鲁棒性研究

发布时间：2018-07-11 15:19

本文选题：复杂网络 + 鲁棒性　；参考：《沈阳理工大学》2014年硕士论文

【摘要】：复杂网络的安全性已经成为人们日益关心的一个重要问题。人们做出了很多的努力，但是大规模的相继故障仍然时有发生。复杂网络的相继故障，就是指一个或少数几个节点或边发生的故障通过节点之间的耦合关系引起其他节点发生故障，这样就会产生连锁效应，最终导致相当一部分节点甚至整个网络的崩溃，也形象的称为“雪崩”。为满足人们对各种关乎国计民生网络安全性和可靠性的要求，有必要对复杂网络相继故障的发生机理、相继故障的预防与控制进行研究。本文首先介绍了四种经典的基本网络模型，分别是规则网络、随机网络、WS小世界网络和BA无标度网络，研究了网络的基本特性，分析了BA无标度网络、WS小世界网络和互联网在遭遇攻击时的鲁棒性。然后分析了三种网络在随机性攻击和蓄意性攻击上的鲁棒性。用参数攻击标度来控制随机攻击和蓄意攻击，，用最大连通子图比来表示网络的鲁棒性能，通过分析，在随机性攻击方面，三种网络都表现出很强的鲁棒性；在蓄意性攻击方面，BA无标度网络和互联网都表现出了很强的脆弱性，但互联网稍好点，而WS小世界网络几乎没表现出脆弱性。最后分析了网络负载对三种网络的鲁棒性影响，实验表明，当网络空负载时，WS表现出很强的鲁棒性，而BA无标度网络和互联网次之；在网络满负载的情况下，分为两种情况，在容许参数低时，三种网络都表现出了很强的脆弱性，在容许系数高时，在随机性攻击时BA无标度网络表现强鲁棒性，而互联网弱些，在蓄意攻击时BA无标度表现出高度的脆弱性，互联网次之。 WS小世界网络不管是随机攻击还是蓄意攻击都有很强的鲁棒性，但强度不及BA无标度网络。
[Abstract]:The security of complex networks has become an important issue that people pay more and more attention to. A lot of efforts have been made, but large-scale successive failures still occur from time to time. The sequential failure of complex network refers to the failure of one or a few nodes or edges causing the faults of other nodes through the coupling relationship between nodes, which will produce cascading effects. Eventually leading to the collapse of a considerable number of nodes or even the entire network, also known as the "avalanche." In order to meet the requirements of the safety and reliability of the network, it is necessary to study the occurrence mechanism, prevention and control of the successive faults of the complex network. This paper first introduces four kinds of classical basic network models, namely, regular network, random network, small world network and BA scale-free network, and studies the basic characteristics of the network. In this paper, the robustness of the BA scale-free network / WS small-world network and the Internet under attack is analyzed. Then, the robustness of three kinds of networks on random attack and deliberate attack is analyzed. The parameter attack scale is used to control random attack and deliberate attack, and the maximum connected subgraph ratio is used to represent the robustness of the network. Both scale-free networks and the Internet have shown a strong vulnerability to intentional attacks, but the Internet is slightly better, while the WS small-world network shows little vulnerability. Finally, the effects of network load on the robustness of the three networks are analyzed. Experiments show that when the network is empty, WS exhibits strong robustness, while BA scale-free network and Internet take the second place, and when the network is full, it can be divided into two cases. When the allowable parameters are low, the three kinds of networks exhibit strong vulnerability. When the allowable coefficient is high, the BA scale-free network exhibits strong robustness while the Internet is weaker. BA has a high degree of vulnerability in a deliberate attack, followed by the Internet. WS small world network has strong robustness whether random attack or deliberate attack, but its strength is not as strong as BA scale-free network.
【学位授予单位】：沈阳理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：O157.5;TP393.08

【参考文献】