Uncovering the Dependence of Cascading Failures on Network Topology by Constructing Null Models

Cascading failures are the significant cause of network breakdowns in a variety of complex infrastructure systems. Given such a system, uncovering the dependence of cascading failures on its underlying topology is essential but still not well explored in the field of complex networks. This study offers an original approach to systematically investigate the association between cascading failures and topological variation occurring in realistic complex networks by constructing different types of null models. As an example of its application, we study several standard Internet networks in detail. The null models first transform the original network into a series of randomized networks representing alternate realistic topologies, while taking its basic topological characteristics into account. Then considering the routing rule of shortest-path flow, it is sought to determine the implications of different topological circumstances, and the findings reveal the effects of micro-scale (such as degree distribution, assortativity, and transitivity) and meso-scale (such as rich-club and community structure) features on the cascade damage caused by deliberate node attacks. Our results demonstrate that the proposed method is suitable and promising to comprehensively analyze realistic influence of various topological properties, providing insight into designing the networks to make them more robust against cascading failures.

Computational modeling of radiation-induced segregation in concentrated binary alloys

ABSTRACTA sharp-interface model to study radiation-induced segregation in binary alloy has been developed. This model is based on a set of reaction-diffusion equations for the point defect and atomic species concentrations, with a stochastic, spatially-resolved, discrete defect generation terms representing the cascade damage. An important feature of this model, which is significantly different from the way radiation-induced segregation has been studied in the past, is that the role of the boundaries as defect sinks has been ensured by defining defect-boundary interactions via a set of reaction boundary conditions. Defining defect-boundary interactions in this way makes it possible to capture the process of segregation as a consequence of boundary motion. The model is tested in 2D for Cu-Au solid solution with the material surface being free to move. The Gear method has been used to solve the reaction-diffusion equations. Enrichment of Cu and depletion of Au have been observed near to the boundaries.