Indirect Power-System Contingency Screening for Real-Time Applications Based on PCA

This letter introduces the application of a data mining method with the purpose of contingency screening, by rapid recognition of hazardous, reoccurring power-system operating conditions. The method, suitable for real-time applications, is demonstrated on the north-western part of the Slovenian power-system, for first-swing stability issues. The presented demonstration consists of two steps: First, a database containing a set of prefault operating states and the corresponding critical clearing times of several contingencies is constructed. Second, the prefault measurements matrix is decomposed using the principal component analysis method and represented in a coordinate system, defined by the principal components. Since operating states form dense clusters of points in this coordinate system, the similarity between current and past conditions is established by identifying the shortest Euclidean distance metric. In this manner, the indication of each contingency impact is provided rapidly as long as a similar operating state exists within the database. Otherwise, the case is thoroughly investigated and included in the database. This approach is applicable to a wide spectrum of dynamic problems, providing that problem-relevant sets of input data are considered.