Indentification and Ranking of Critical Contingencies in Dependent Variable Space

Abstract

Working in dependent-variable space, reliable yet simple computational procedures are presented for the purpose of identifying and ranking line outages which result in the violation of various limits related to power system security. Security limits defined in terms of real-power line flows, generator reactive powers and demand-bus voltage magnitudes are considered in normalised subspaces wherein critical contingencies are identified by a filtering algorithm using the infinite norm as a performance index. Critical contingencies are then ranked using performance indices which are defined in accordance with the subspace of interest. In the case of real-power line flows, ranking is based on a measure related to distance from the center of the corresponding security region whereas, in the case of generator reactive powers and demand-bus voltage magnitudes, ranking is based on measures related to distances from the boundaries of the corresponding regions. Furthermore, an operating point's violation of limits associated with generator reactive powers and demand-bus voltage magnitudes is measured in combined form form with due consideration being given to the relationships between these reactive powers and voltage magnitudes.