Prediction of cascade failure using probabilistic approach with AC load flow

Abstract

Cascading failures have been the cause of major blackouts in power systems which lead to economical loss and system instability. Following an initial triggering event, it is challenging to predict whether a system would remain stable or tend towards cascade. In order to predict cascading phenomena of power system which is a complex interconnected network, existing deterministic approach may fail to take into account network uncertainties which is taken care by probabilistic approach. Given an initial triggering event, this paper proposes a probabilistic approach to cascading failures in power system which predicts the next line to trip by considering the thermal limit of each line as a constraint. For every line outage, the deviations in line flows with respect to the mean are calculated and the line with the least deviation is the next probable line to trip. The results of this approach are verified using power world simulator.