加权无标度网络的级联失效模型

Cascading Failure Model of Weighted Scale-Free Networks

在加权的无标度网络中,为了抵抗网络的级联失效,增强网络的鲁棒性,提出了一种参数可调的级联失效模型.该模型从全局和局域的角度,将节点介数、节点度、节点权重和邻居节点权重相结合构建节点的初始负载,并建立节点容量与初始负载的比例关系,当节点失效后,通过结合失效节点邻居的容量来制定负载重分配规则,进而通过对网络级联失效的分析,推导负载参数的演化过程,得出模型中的参数对网络鲁棒性的影响.最后,通过实验验证了所提方法的有效性.

The robustness of complex network against cascading failures is of great significance. Based on the weighted scale-free network, a new cascading failure model with adjustable parameters is proposed. In this model, the nodes＇ initial loads are constructed with node betweermess, node degree, node weight and neighbor node＇s weight from both global and local perspective. Meanwhile, the initial load is made proportional to the node capacity. Adopting a new redistribution rule, the failed nodes＇ loads are assigned to their neighbors. Then the load parameters can be obtained through analyzing the cascade failure process. Finally, the effectiveness of the proposed method are verified by experiment.