A New Link Isolation Algorithm based on Complex Network Theory for Preventing Cascading Failures in Power Grid

Abstract

Weak links is one of the main causes of large-scale blackouts in power systems. How to accurately identify the weakest links so as to disconnect the links that could cause cascading failures when an accident happens to a power system, is a vital research topic for keeping power system safety by preventing large-scale blackouts caused by cascading failures. In this paper, based on complex network theory we first propose a method to identify the weakest links according to vertices second order centrality and links joint vulnerability metric by combining link betweenness and link load level, which is called joint vulnerability method (JVM). Based on JVM, we design a links isolation algorithm when a weakest link failure happens to a power system, which is called joint isolation algorithm (JIA). The simulation results show that JVM could identify the weakest links accurately, and the JIA could avoid large-scale blackouts effectively which plays a guiding role in the optimization of power system safety.