Understanding power grid network vulnerability through the stochastic lens of network motif evolution

Modern cyber-physical systems must exhibit high reliability since their failures can lead to catastrophic cascading events. Enhancing our understanding of the mechanisms behind the functionality of such networks is a key to ensuring the resilience of many critical infrastructures. In this paper, we develop a novel stochastic model, based on topological measures of complex networks, as a framework within which to examine such functionality. The key idea is to evaluate the dynamics of network motifs as descriptors of the underlying network topology and its response to adverse events. Our experiments on multiple power grid networks show that the proposed approach offers a new competitive pathway for resilience quantification of complex systems.