Cascading failures caused by node overloading in complex networks

Abstract

It is well known that complex networks are vulnerable to the failure of hubs in terms of structural robustness. An as yet less researched topic is dynamical robustness, which refers to the ability of a network to maintain its dynamical activity against local disturbances. This paper analyzes the impact of overload attacks in complex networks and gives a precise definition of this type of attack using the load redistribution model. The main contribution is a greedy algorithm to select a small number of candidates for an overload attack maximizing the impact with respect to the number of failed nodes and load increase. The quality of the algorithm is analyzed for a real power grid network.