Classification of short-circuit faults in high-voltage energy transmission line using energy of instantaneous active power components-based common vector approach

Yumurtaci M., GÖKMEN G., KOCAMAN Ç., ERGİN S., KILIÇ O.

TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES, cilt.24, sa.3, ss.1901-1916, 2016 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 24 Sayı: 3
  • Basım Tarihi: 2016
  • Doi Numarası: 10.3906/elk-1312-131
  • Dergi Adı: TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, TR DİZİN (ULAKBİM)
  • Sayfa Sayıları: ss.1901-1916
  • Anahtar Kelimeler: Common vector approach, support vector machine, artificial neural network, wavelet packet transform, fault classification, short circuit, transmission line, WAVELET TRANSFORM, LOCATION, IDENTIFICATION, ALGORITHM, MACHINES
  • Marmara Üniversitesi Adresli: Evet

Özet

The majority of power system faults occur in transmission lines. The classification of these faults in power systems is an important issue. In this paper, the real parameters of a 28 km, 154 kV transmission line between Simav and Demirci in Turkey's electricity transmission network is simulated in MATLAB/Simulink. Wavelet packet transform (WPT) is applied to instantaneous voltage signals. Instantaneous active power components are obtained by multiplying instantaneous currents obtained from a voltage source side with these WPT-based voltage signal components. A new feature vector extraction scheme is employed by calculating the energies of instantaneous active power components. Constructed feature vectors are treated with a classifier for short-circuit faults that occurred in high-voltage energy transmission lines; this is known as the common vector approach (CVA). This is the first implementation of CVA in the classification of short-circuit faults that occurred in high-voltage energy transmission lines. Furthermore, the same feature vector is applied to a support vector machine and artificial neural network for a comparison with the CVA method regarding classification performance and testing duration issues. Additionally, a graphical user interface is designed in MATLAB/GUI. Various noise levels, source frequencies, fault distances, fault inception angles, and fault exposure durations can be investigated with this interface. Classification of short-circuit faults in high-voltage transmission line is achieved by using an offline monitoring methodology. It is concluded that a combination of the proposed feature extraction scheme with the CVA classifier gives substantially high performance for the classification of short circuit faults in transmission line.