Power System Fault Diagnosis Based on Gradual Optimization in Consideration of Alarm Information Aberrance

Nian Mei,Dongyuan Shi,Yinhong Li,xianzhong Duan
DOI: https://doi.org/10.3321/j.issn:1000-6753.2009.09.027
2009-01-01
Abstract:Recent optimization-based fault section estimation methods for power systems can be categorized as consistency diagnostic methods or parsimonious covering diagnostic methods. The two kinds of methods may reach wrong diagnosis results in complex scenerios in which maloperation of protections and alarm aberrance exist. To overcome their drawbacks, a multi-objective gradual optimization method was built by combining the merits of the two methods. The proposed method took turns to make the consistency diagnosis based on all the alarm and the protection alarm repectively, and then the parsimonious covering diagnosis. By simple calculation of the diagnosis results, several kinds of fault section candidates are listed according to their fault probability. Case studies show that the method presented can accurately locate the faulty element in the aforementioned complex scenerios but also give a detailed evaluation of the operation of protections. To reduce the heavy calculation load caused by multiplicity of characteristic alarms, a method to construct adaptive characteristic alarms is proposed. Specefic approaches to confine and list various state vectors are provided based on the outage area, which make the diagnosis process much faster.
What problem does this paper attempt to address?