Güç Kalitesi Bozulmalarının 2 Boyutlu Ayrık Dalgacık Dönüşümü ve Torbalama Karar Ağaçları Yöntemi ile Sınıflandırılması

Bu çalışmada, Güç Kalitesi (GK) bozulmalarının sınıflandırılması için 2 Boyutlu Ayrık Dalgacık Dönüşümü (2B-ADD) yöntemi ile öznitelikler çıkartılmakta ve Destek Vektör Makineleri (DVM), Yapay Sinir Ağları (YSA) ve Torbalama Karar Ağaçları (TKA) yöntemleri ile sınıflandırma işlemi yapılmaktadır. Gürültülü (40 dB, 30 dB ve 20 dB) ve gürültüsüz durumları içeren 11 farklı GK bozulması için toplamda 2200 adet sinyal sentetik olarak üretilmektedir. Sinyaller 2 boyutlu görüntü matrislerine çevrilmekte ve her birine 2B-ADD uygulanmaktadır. Farklı ayrıştırma seviyesi ve istatistiksel özellikler uygulanarak öznitelikler oluşturulmaktadır. Özniteliklerden en uygun olanları Sıralı İleri Seçim (SİS) ve ReliefF yöntemleri ile seçilmektedir. Benzetim çalışmasına göre 3 farklı sınıflandırıcının başarımı birbirleri ile kıyaslanmaktadır. Sıralı ileri seçim ile seçilen öznitelikleri kullanan TKA yönteminin %99.12±0.12 oranı ile en iyi başarımı veren yöntem olduğu görülmektedir.

Anahtar Kelimeler:

Güç kalitesi, 2 boyutlu ayrık dalgacık dönüşümü, torbalama karar ağaçları sınıflandırıcısı, sıralı ileri seçim, ReliefF

Classification of Power Quality Disturbances with 2D Discrete Wavelet Transform and Bagged Decision Trees Method

In this study, to classify Power Quality (PQ) disturbances, attributes are extracted by 2D Discrete Wavelet Transform (2D-DWT) method and Support Vector Machines, Artificial Neural Networks and Bagged Decision Trees (BDT) methods are used for classification stage. 2200 signals are synthetically produced for 11 different PQ disturbances, including noisy (40 dB, 30 dB and 20 dB) and noiseless states. Signals are transformed into 2D image matrices and 2D DWT is applied to each. Attributes are created by applying different level of decomposition and statistical properties. The most appropriate ones are selected with Sequential Forward Selection (SFS) and ReliefF methods. BDT method, which uses selected attributes with SFS, is the method that gives the best performance with a rate of 99.12±0.12%.

Keywords:

Power quality, 2D discrete wavelet transform, bagged trees classifier, sequential forward selection, ReliefF,

PDF

___

[1] Thirumala K., Jain T. and Umarikar A.C., "Visualizing time-varying power quality indices using generalized empirical wavelet transform", Electric Power Systems Research, 143: 99-109, (2017).
[2] Nashad N.R., Islam M.J., Alam S., Rahat R.M., Begum M.T. and Alam M.R, "A simplistic mathematical approach for detection and classification of power quality events", Electrical, Computer and Communication Engineering (ECCE), International Conference on IEEE, 698-703, (2017).
[3] Granados-Lieberman D., Romero-Troncoso R.J., Osornio-Rios R.A., Garcia-Perez A. and Cabal-Yepez E., "Techniques and methodologies for power quality analysis and disturbances classification in power systems: a review", IET Generation, Transmission & Distribution, 5(4): 519-529, (2011).
[4] Saini M.K. and Kapoor R., "Classification of power quality events–a review", International Journal of Electrical Power & Energy Systems, 43(1): 11-19, (2012).
[5] Khokhar S., Zin A.M., Mokhtar A.S., Ismail N.M. and Zareen N., "Automatic classification of power quality disturbances: A review", Research and Development (SCOReD), IEEE Student Conference on IEEE, 427-432, (2013).
[6] Stańczyk U. and Jain L.C., "Feature selection for data and pattern recognition", New York: Springer, (2015).
[7] Tan P.N., Kumar V. and Steinbach M., “Introduction to Data Mining”, Pearson, (2005).
[8] Montoya F.G., García-Cruz A., Montoya M.G. and Manzano-Agugliaro F., “Power quality techniques research worldwide: A review” Renewable and Sustainable Energy Reviews, 54: 846-856, (2016).
[9] Karasu S. and Başkan S. “Classification of power quality disturbances by using ensemble technique”, 24th Signal Processing and Communication Application Conference (SIU), IEEE, 529-532, (2016)
[10] Ece D.G. and Gerek O.N., “Power quality event detection using joint 2-D-wavelet subspaces”, IEEE Transactions on Instrumentation and Measurement, 53(4): 1040-1046. (2004).
[11] Shareef H., Mohamed A. And Ibrahim A.A., “An image processing based method for power quality event identification”, International Journal of Electrical Power & Energy Systems, 46: 184-197, (2013).
[12] Krishna B.V. and Kaliaperumal B., “Image pattern recognition technique for the classification of multiple power quality disturbances”, Turkish Journal of Electrical Engineering and Computer Science, 21(3): 656-678, (2013).
[13] Uyar M., Kaya Y. and Ataş M., “Classification of power quality disturbances based on S-transform and image processing techniques”, 21st Signal Processing and Communications Applications Conference (SIU), IEEE, 1-4, (2013).
[14] Gonzalez R.C. and Woods R.E., “Digital Image Processing”, Prentice Hall, 2002.
[15] Robnik-Šikonja M. and Kononenko I., “An adaptation of Relief for attribute estimation in regression”, Machine Learning: Proceedings of the Fourteenth International Conference (ICML), 296-304, (1997).
[16] Kira K. and Rendell L.A., “A practical approach to feature selection”, Machine Learning Proceedings, 249-256, (1992).
[17] Kononenko I., “Estimating attributes: analysis and extensions of RELIEF”, European conference on machine learning, Springer, 171-182, (1994).
[18] Karasu S. and Saraç Z., “Classification of power quality disturbances with S-transform and artificial neural networks method”, 25th Signal Processing and Communications Applications Conference (SIU), IEEE, 1-4, (2017).
[19] Karasu S., Altan A., Saraç Z. and Hacıoğlu R., “Prediction of wind speed with non-linear autoregressive (NAR) neural networks”, 25th Signal Processing and Communications Applications Conference (SIU), IEEE, 1-4, (2017).
[20] Karasu S., Altan A., Saraç Z. and Hacıoğlu R., “Estimation of fast varied wind speed based on NARX neural network by using curve fitting”, International Journal of Energy Applications and Technologies, 4(3): 137-146, (2017).