As a global problem in the power quality area, voltage sags are matter of high interest for both utilities and customers. With a view to resolving the problem of sag source location in the power network, this paper introduces a new method based on dimension reduction capability of Multiway Principal Component Analysis (MPCA). MPCA models are developed using three dimensional databases of voltage and current Root Mean Square (RMS) values. Computed scores are then used for training commonly used classifiers for putting sags in two classes. A feature selection algorithm is successfully applied for determining the optimal subsets of scores for training classifiers and also the number of principal components in the MPCA models. The proposed method is tested with success using some real voltage sags recorded in some substations. Also, through some experiments we demonstrate that satisfactorily high classification rates must be attributed to the applied feature selection algorithm.
IOS Press, Inc.
6751 Tepper Drive
Clifton, VA 20124
Tel.: +1 703 830 6300
Fax: +1 703 830 2300 firstname.lastname@example.org
(Corporate matters and books only) IOS Press c/o Accucoms US, Inc.
For North America Sales and Customer Service
West Point Commons
Lansdale PA 19446
Tel.: +1 866 855 8967
Fax: +1 215 660 5042 email@example.com