Archives of Acoustics, 46, 3, pp. 409–417, 2021
10.24425/aoa.2021.138134

The Application of Selected Hierarchical Clustering Methods for Classification the Acoustic Emission Signals Generated by Partial Discharges

Sebastian BORUCKI
Opole University of Technology
Poland

Jacek ŁUCZAK
Opole University of Technology
Poland

Marcin LORENC
Opole University of Technology
Poland

The paper presents the results of the application of the hierarchical clustering methods for the classification of the acoustic emission (AE) signals generated by eight basic forms of partial discharges (PD), which can occur in paper-oil insulation of power transformers. Based on the registered AE signals from the particular PD forms, using a frequency descriptor in the form of the power spectral density (PSD) of the signal, their representation in the form of the set of points on plane XY was created. Next, these sets were subjected to analysis using research algorithms consisting of selected clustering methods. Based on the suggested numeric performance indicators, the analysis of the degree of reproduction of the actual distribution of points showing the particular time waveforms of the AE signals from eight adopted PD forms (PD classes) in the obtained clusters was carried out. As a result of the analyses carried out, the clustering algorithms of the highest effectiveness in the identification of all eight PD classes, classified simultaneously, where indicated. Within the research carried out, an attempt to draw general conclusions as to the selection of the most effective hierarchical clustering method studied and the similarity function to be used for classification of the selected basic PD forms.
Keywords: acoustic emission method; acoustic signals; partial discharges; power transformer; clustering method
Full Text: PDF
Copyright © The Author(s). This is an open-access article distributed under the terms of the Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0).

References

Akbari A., Setayeshmehr A., Borsi H., Gockenbach E. (2010), Intelligent agent-based system using dissolved gas analysis to detect incipient faults in power transformers, IEEE Electrical Insulation Magazine, 26(6): 27–40, doi: 10.1109/MEI.2010.5599977.

Boczar T. (2001), Identification of a specific type of PD form acoustics emission frequency spectra, IEEE Transaction on Dielectric and Electrical Insulation, 8(4): 598–606, doi: 10.1109/94.946712.

Boczar T., Borucki S., Cichoń A., Zmarzły D. (2009), Application Possibilities of Artificial Neural Networks for Recognizing Partial Discharges Measured by the Acoustic Emission Method, IEEE Transaction on Dielectric and Electrical Insulation, 16(1): 214–223, doi: 10.1109/TDEI.2009.4784570.

Boczar T., Cichoń A., Borucki S. (2014), Diagnostic expert system of transformer insulation systems using the acoustic emission method, IEEE Transaction on Dielectric and Electrical Insulation, 21(2): 854–865, doi: 10.1109/TDEI.2013.004126.

Borucki S., Boczar T., Cichoń A., Lorenc M. (2007), The evaluation of neural networks application for recognizing single-source PD forms generated in paper-oil insulation systems based on the AE signal analysis, European Physical Journal Special Topics, 154: 23–29, doi: 10.1140/epjst/e2008-00512-7.

Borucki S., Łuczak J. (2017), Assessment of the impact of an acoustic signal power spectral density frequency selection on partial discharges basic forms classification efficiency with the use of data clustering method [in Polish: Ocena wpływu doboru częstotliwości widmowej gęstości mocy sygnału akustycznego na efektywność klasyfikacji podstawowych form wyładowań niezupełnych z użyciem metody klasteryzacji], Energetyka, 7: 448–452.

Borucki S., Łuczak J., Zmarzły D. (2018), Using Clustering Methods for the Identification of Acoustic Emission Signals Generated by the Selected Form of Partial Discharge in Oil-Paper Insulation, Archives of Acoustics, 43(2): 207–215, doi: 10.24425/122368.

Castro Heredia L.C., Rodrigo Mor A. (2019), Density-based clustering methods for unsupervised separation of partial discharge sources, International Journal of Electrical Power & Energy Systems, 107: 224–230, doi: doi: 10.1016/j.ijepes.2018.11.015.

Chia-Hung L., Chien-Hsien W., Ping-Zan H. (2009), Grey clustering analysis for incipient fault diagnosis in oil-immersed transformers, Expert Systems with Applications, 36(2, part 1): 1371–1379, doi: 10.1016/j.eswa.2007.11.019.

Cichoń A. (2013), Assessment of technical condition of on-load tap-changers by the method of acoustic emission, [in Polish: Ocena stanu technicznego podobciążeniowych przełączników zaczepów metodą emisji akustycznej], Studia i Monografie, No. 352, Ofic. Wyd. Politechniki Opolskiej.

Fuhr J. (2005), Procedure for identification and localization of dangerous partial discharge sources in power transformers, IEEE Transaction on Dielectric and Electrical Insulation, 12(5): 1005–1014, doi: 10.1109/TDEI.2005.1522193.

Han J., Kamber M., Pei J. (2012), Data Mining. Concepts and Techniques, 3rd ed., Morgan Kaufmann Publishers, Waltham.

Kapinos J., Glinka T., Drak B. (2014), Typical causes of operational failures in power transformers working in National Grid [in Polish: Typowe przyczyny uszkodzeń eksploatacyjnych transformatorów energetycznych], Przegląd Elektrotechniczny, 90(1): 186–189, doi: 10.12915/pe.2014.01.45.

Kazmierski M., Olech W. (2013), Technical Diagnostics and Monitoring of Transformers [in Polish: Diagnostyka techniczna i monitoring transformatorów], Printing house of ZPBE Energopomiar-Elektryka Sp. z o.o., Gliwice.

Krzyśko M., Wołyński W., Górecki T. Skorzybut M. (2008), Learning Systems – Pattern Recognition, Cluster Analysis and Dimensional Reduction [in Polish: Systemy uczące się – rozpoznawanie wzorców, analiza skupień i redukcja wymiarowości], Wydawnictwa Naukowo-Techniczne, Warszawa.

Kurtasz P. (2011), Application of a multi-comparative algorithm to classify acoustic emission signals generated by partial discharges [in Polish: Zastosowanie algorytmu multikomparacyjnego do klasyfikacji sygnałów emisji akustycznej generowanych przez wyładowania niezupełne], Ph.D. Dissertation, Opole University of Technology.

Lalitha E.M., Satish L. (2002), Wavelet analysis for classification of multi-source PD patterns, IEEE Transaction on Dielectric and Electrical Insulation, 7(1): 40–47, doi: 10.1109/94.839339.

Ming-Shou S., Chung-Chu C., Chien-Yi C., Jiann-Fuh C. (2014), Classification of partial discharge events in GILBS using probabilistic neural networks and the fuzzy c-means clustering approach, International Journal of Electrical Power & Energy Systems, 61: 173–179, doi: 10.1016/j.ijepes.2014.03.054.

Mohan Rao U., Sood Y.R., Jarial R.K. (2015), Subtractive Clustering Fuzzy Expert System for Engineering Applications, Procedia Computer Science, 48: 77–83, doi: 10.1016/j.procs.2015.04.153.

Morzy T. (2013), Data mining. Methods and Algorithms [in Polish: Eksploracja danych. Metody i algorytmy], Wydawnictwo Naukowe PWN, Warszawa.

Olszewska A., Witos F. (2012), Location of partial discharge sources and analysis of signals in chosen power oil transformers by means of acoustic emission method, Acta Physica Polonica A, 122(5): 921–926.

Radionov A.A., Evdokimov S.A., Sarlybaev A.A., Karandaeva O.I. (2015), Application of Subtractive Clustering for Power Transformer Fault Diagnostics, Procedia Engineering, 129: 22–28, doi: 10.1016/j.proeng.2015.12.003.

Rodrigo Mor A., Castro Heredia L.C., Muñoz F.A., Effect of acquisition parameters on equivalent time and equivalent bandwidth algorithms for partial discharge clustering, International Journal of Electrical Power & Energy Systems, 88: 141–149, doi: 10.1016/j.ijepes.2016.12.017.

Rubio-Serrano J., Rojas-Moreno M., Posada J., Martienez-Tarifa J., Robles G., Garcia-Souto J. (2012), Electro-acoustic detection, identification and location of PD sources in oil-paper insulation systems, IEEE Transaction on Dielectric and Electrical Insulation, 19(5): 1569–1578, doi: 10.1109/TDEI.2012.6311502.

Soltani A.A., Haghjoo F., Shahrtash S.M. (2012), Compensation of the effects of electrical sensors in measuring PD signals, IET Science, Measurement &Technology, 6(6): 494–501, doi: 10.1049/iet-smt.2012.0001.




DOI: 10.24425/aoa.2021.138134