Data-mining-based Approach for Predicting the Power System Post-contingency Dynamic Vulnerability Status

JAIME CEPEDA; JOSÉ LUIS RUEDA TORRES; D. GRACIELA COLOMÉ; I. ERLICH

International Transactions on Electrical Energy Systems

John Wiley & Sons, Ltd.

Año: 2014 vol. 8 p. 1 - 32

This paper proposes a data-mining-based approach for predicting the power system post-contingency vulnerability status in real time. To this aim, the system dynamic vulnerability regions (DVRs) are firstly determined by applying singular value decomposition, by means of empirical orthogonal functions (EOFs), to a post-contingency data base obtained from phasor measurement units (PMU) adequately located throughout the system. In this way, the obtained pattern vectors (EOF scores) allow mapping the DVRs within the coordinate system formed by the set of EOFs, which permits revealing the main patterns immersed in the collected PMU signals. The data base along with the DVRs enable the definition, training, and identification of a support vector classifier (SVC) which is employed to predict the post-contingency vulnerability status as regards three short-term stability phenomena, that is: transient stability, short-term voltage stability, and short-term frequency stability (TVFS). Enhanced procedures for feature extraction and selection as well as heuristic optimization-based parameter identification are proposed to ensure a robust performance of the SVC. Numerical results, obtained by implementation of the proposed approach on two different size test power systems, demonstrate the methodology viewpoint and effectiveness.