加权无标度网络上级联动力学建模研究  

Research on Model for Cascading Dynamics on Weighted Scale-free Networks

作　　者：丁琳[1,2] 张嗣瀛[2] 

机构地区：[1]南华大学计算机科学与技术学院,衡阳4210011 [2]青岛大学复杂性科学研究所,青岛266071

出　　处：《系统仿真学报》2014年第3期621-625,共5页Journal of System Simulation

基　　金：国家自然科学基金(61104075)

摘　　要：考虑以自然定律所决定的流量模型来建模分布式流,在此基础上,使用现实网络节点的负载-容量关系,提出一个带有可调权重参数的级联模型。依据新的度量网络鲁棒性指标,探讨了加权无标度模型网络和真实网络上的级联动力学行为。仿真结果表明,存在最优权重策略使得分布式流下的网络达到最强抵制级联失效的鲁棒性。而且,相比于以往基于最短路径流的最优权重策略,基于分布式流的最优权重策略使得网络更具鲁棒性,能更显著地降低级联动力学发生的可能性。Considering the flow governed by the natural laws to model distributed flow, and using a realistic load-capacity relation for network nodes, a cascading model with a tunable weight parameter was proposed. According to new measures to the robustness of a network, Cascading dynamics on weighted scale-free model networks and real-world network was further investigated. The simulation results show that the existence of an optimal weighting strategy for which the network under distributed flows exhibits a maximum degree of robustness against cascading failures. Furthermore, the optimal weighting strategy based on distributed flow makes the networks more robust as compared with the previous that is based on shortest-path flow, showing that the possibility of the occurrence of the cascading dynamics can be more significantly reduced.

关 键 词：级联失效 分布式流 鲁棒性 无标度网络 权重策略 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]