Detection of blackouts by using K-means clustering in a power system

Özgönenel O., Thomas D., Yalcin T., Bertizlioglu I.

11th IET International Conference on Developments in Power Systems Protection, DPSP 2012, Birmingham, İngiltere, 23 - 26 Nisan 2012, cilt.2012 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 2012
  • Doi Numarası: 10.1049/cp.2012.0079
  • Basıldığı Şehir: Birmingham
  • Basıldığı Ülke: İngiltere
  • Anahtar Kelimeler: Anomaly (Outlier) Detection, Blackout, K-Means Clustering, Principal Component Analysis (PCA)
  • Ondokuz Mayıs Üniversitesi Adresli: Evet

Özet

This paper presents a novel approach for the detection of abnormal power system states that force systems into blackout. K-means clustering techniques and two types of distances for identifying pattern clusters are used to detect abnormal conditions. PCA is used for the reduction of the data matrix for faster calculations. The proposed hybrid technique is then demonstrated on an IEEE 14-bus system.