Attack robustness of cascading model with node weight

作者：Wang Jianwei^*; Zhang Chuan; Huang Yi; Xin Chong

来源：Nonlinear Dynamics, 2014, 78(1): 37-48.

DOI：10.1007/s11071-014-1420-3

• Cascading failure
• Node weight
• Robustness
• Attack

Considering the node weight and the effect of the neighboring nodes, we introduce a method to define the initial load of a node in a network and propose a cascading model. Our aim is to explore how to allocate the initial load and select some nodes to be protected so as to maximize the network robustness against cascading failures. According to the distribution of the node weight and the normalized effect of the neighboring nodes, we focus on investigating different roles of high-load and low-load nodes and the correlation between some parameters in our model and the strongest robust level of a network against cascading failures. We obtain by the numerical simulations the optimal values of the parameters in our model at which the network can reach the strongest robust level against cascading failures. In addition, we find that the effect of two attacks strongly depends on the parameters of the node weight and the normalized effect of the neighboring nodes, i.e., the network robustness of attacking the low-load nodes has a positive correlation with the parameter of the node weight, while has a negative correlation with the parameter of the normalized reaction of the neighboring nodes. While the result of attacking the high-load nodes is almost on the contrary. Finally, we verify the numerical results by the theoretical analysis.

• 出版日期2014-10
• 单位东北大学