Stochastic Load-Redistribution Model for Cascading Failure Propagation

TL;DR

This paper introduces a probabilistic model for cascading failures in interconnected systems, accounting for load redistribution and stochastic effects, with analytical solutions and detailed failure propagation analysis.

Contribution

It presents a novel stochastic load-redistribution model that incorporates real system characteristics, solved analytically using generalized branching processes.

Findings

Model captures load-dependent failure propagation

Analytical solutions provide insights into failure dynamics

Prototype analysis demonstrates model effectiveness

Abstract

A new class of probabilistic models for cascading failure propagation in interconnected systems is proposed. The models take into account important characteristics of real systems that are not considered in existing generic approaches. Specifically, it is assumed that the load increments after a failure are proportional to the failed load and that the load redistribution among the remaining elements is stochastic. The models are solved analytically in terms of generalized branching processes, and the failure propagation properties of a prototype example are analyzed in detail.