Abstract
An artificial intelligence system has been developed for diagnosing the occurrence of a partial discharge in the insulation of overhead power lines. In order to improve the accuracy of fault detection, several stages of preprocessing are used, such as separation into frequency components, balancing, segmentation and normalization of the initial data, which are measurements of the high-frequency components of the current and voltage spectrum in each phase of the power line. The paper shows that the main role in the identification of a partial discharge is played by electromagnetic waves that occur during a partial breakdown of insulation with a frequency of 10-20 MHz. To do this, the source data is divided by filtering into two components - high-frequency and low-frequency. The maximum accuracy of neural networks trained on the full signal and its high-frequency component differ by less than 1%, while the maximum accuracy of the network trained on the low-frequency component is 17% lower. To test the proposed system, a real data set was used, which makes it possible to speak about the possibility of practical application of the developed system. The developed artificial intelligence system showed a relatively high diagnostic accuracy of about 88%. Thus, with the help of data mining, it was shown that the main role in the identification of the PD is played by high-frequency components, that is, electromagnetic waves that lie in a certain frequency range and occur during a partial breakdown of the insulation. This opens the way to the creation of new methods for diagnosing and detecting the location of a PD source based on the use of this range.
Keywords
partial discharge, diagnostic system, neural network machine learning, data mining, neural network model, decameter waves.
1. Ushakov V.Ya. Sovremennaya i perspektivnaya energetika: tekhnologicheskie, sotsialno-ekonomicheskie i ekologicheskie aspekty. [Modern and prospective energy: technological, socio-economic and environmental aspects]. Tomsk, Pub-lishing House of TPU, 2008. 469 p. (In Russian)
2. IEA (2020), Power Systems in Transition, IEA, Paris. Avail-able at: https://www.iea.org/reports/power-systems-in-transition, License: CC BY 4.0. (accessed 19 May 2023)
3. Menthula J. A Review on Power System Faults and Protec-tion, International journal of engineering research & tech-nology (IJERT). 2022, vol. 11, Iss. 04, pp. 395-398. doi: 10.17577/IJERTV11IS040201
4. Solyonaya O.Ya., Rysin A.V., Soleny S.V., Kuzmenko V.P., Kvas E.S. Characteristics and parameters of the technical condition of overhead power lines. Izvestiya vuzov. Priboro-stroenie. [Journal of Instrument Engineering], 2021, vol. 64, no. 7, pp. 583-588. (In Russian)
5. Arbuzov R.S., Ovsyannikov A.G. Sovremennye metody di-agnostiki vozdushnykh liniy elektroperedachi [Modern methods of diagnostics of overhead power lines]. Novosi-birsk, Nauka, 2009. 135 p. (In Russian)
6. Suhas Aithal. VSB Power Line Fault Detection. 2021. Available at: https://suhaskvaithal.medium.com/vsb-power-line-fault-detection-47f9e94d11d2 (accessed 19 May 2023)
7. Polyakov D.A., Nikitin K.I., Tereshchenko N.A., Novoselov A.S., Bilevich Ya.P. Study of partial discharges in support insulators. Omskiy nauchnyi vestnik [Omsk Scientific Bulle-tin], 2020, no. 1 (169), pp. 32-38. doi: 10.25206/1813-8225-2020-169-32-38 (In Russian)
8. Vdoviko V.P. Chatsichnye razryady v diagnostirovanii vysokovoltnogo oborudovaniya [Partial discharges in the di-agnosis of high-voltage equipment]. Novosibirsk, Nauka, 2007. 155 p. (In Russian)
9. Foorthuis R. On the nature and types of anomalies: a review of deviations in data. Int J Data Sci Anal. 2021, Iss. 12(4), pp. 297-331. doi:10.1007/s41060-021-00265-1
10. Katkov V.I. Analytical review of methods and systems of control of electric discharges in high-voltage electric trans-mission insulators. XXIV Tupolevskie chteniya (shkola mo-lodykh uchenykh): Materialy Mezhdunarodnoy molodezhnoy nauchnoy konferentsii, 7-8 noyabrya 2019 goda, Kazan. [XXIV Tupolev Readings (school of young scientists): Pro-ceedings of the international youth scientific conference, November 7-8, 2019, Kazan]. Kazan, Sagieva A.R. Publ., 2019, vol. 3, pp. 274-277. (In Russian)
11. Gataullin A.M., Badretdinov M.N., Gubaev D.F., Matu-khin V.L. Registration and analysis of the parameters of par-tial discharges on a high-voltage power line with a voltage of 35 kV and higher. Izvestiya vuzov. Problemy energetiki [Power engineering: research, equipment, technology], 2007, no. 7-8, pp. 44-50. (In Russian)
12. Qiu S., Cui X., Ping Z., Shan N., Li Z., Bao X., Xu X. Deep Learning Techniques in Intelligent Fault Diagnosis and Prognosis for Industrial Systems: A Review. Sensors. 2023, no. 23, pp. 1305. doi:10.3390/s23031305
13. Obaldin M.D., Kolobanov P.A. Intelligent algorithm for unilateral fault location on power lines . Intellektualnaya el-ektronika [Smart Electrical Engineering], 2018, no. 1, pp. 84-92. (In Russian)
14. Dementiev S.S., Blokhina M.I. Diagnosis of overhead power lines using an intelligent pulse reflectometry system. Energo- i resursosberezhenie: Promyshlennost i transport [Energy and Resource Saving: Industry and Transport], 2020, no. 2(31), pp. 6-10. (In Russian)
15. Verzunov S. N. Artificial intelligence system for online monitoring of underground power cable lines based on deep learning technologies. Problemy avtomatiki i upravleniya [Automation and control problems], 2021, no. 3(42), pp. 83-94. (In Russian)
16. Abd-Alkader M.Y., Ebid A.M. Dr., Mahdi I., Nosseir I.A. Application of using fault detection techniques in different components in power systems. Future Engineering Journal. 2021, vol. 2, Iss. 2, pp. 1-8.
17. Yadav A., Thoke A.S., Patel R. N. Fault classification of double circuit transmission line using artificial neural network. International Journal of Systems Science. 2008, no. 1(4), pp. 750-755.
18. Abboush M., Bamal D., Knieke C., Rausch A. Intelligent Fault Detection and Classification Based on Hybrid Deep Learning Methods for Hardware-in-the-Loop Test of Auto-motive Software Systems. Sensors. 2022, no. 22, p. 4066. doi: 10.3390/s22114066
19. Wang S., Dehghanian P. On the Use of Artificial Intelligence for High Impedance Fault Detection and Electrical Safety. IEEE Transactions on Industry Applications. 2020, no. 6(56), pp. 7208-7216. doi: 10.1109/TIA.2020.3017698
20. Jarndal A., Alsyouf I. On the Usefulness of Pre-processing Methods in Rotating Machines Faults Classification using Artificial Neural Network. Journal of Applied and Computa-tional Mechanics. 2020, no. 7(1). doi: 10.22055/JACM.2020.35354.2639
21. Adhikari S., Sinha N., Dorendrajit T. Fuzzy logic based on-line fault detection and classification in transmission line. Springer Plus. 2016, no. 5, p. 1002. doi: 10.1186/s40064-016-2669-4
22. Verzunov S.N. Wavelet transform as a tool for analyzing magnetovariational data. Problemy avtomatiki i upravleniya [Automation and control problems], 2014, no. 2(27), pp. 52-61. (In Russian)
23. Sharma P., Saini D., Saxena A. Fault Detection and Classifi-cation in Transmission Line Using Wavelet Transform and ANN. Bulletin of Electrical Engineering and Informatics. 2016, no. 4(5), pр. 456-465. doi: 10.11591/eei.v5i3.537
24. Zen H., Tokuda K., Alan W. Black A.W. Statistical parametric speech synthesis. Speech Communication. 2009, vol. 51, Iss. 11, pp. 1039-1064. doi: 10.1016/j.specom.2009.04.004
25. Dong M., Sun Z., Wang C. A Pattern Recognition Method for Partial Discharge Detection on Insulated Overhead Con-ductors. Canadian Conference of Electrical and Computer Engineering (CCECE).IEEE, 2019, pp. 1-4. doi: 10.1109/CCECE.2019.8861809
26. Wadi M., Elmasry W. An anomaly-based technique for fault detection in power system networks. Proceedings of the In-ternational Conference on Electric Power Engineering – Pal-estine (ICEPE-P). IEEE, 2021, pp. 1–6. doi:10.1109/ICEPE-P51568.2021.9423479
27. Sciocchetti F., Caprara A., Puletti F., Montanari G.C., Caval-lini A. Innovative Partial Discharge measurement and analysis technique for the evaluation of the reliability of insulation systems. Proceedings of Grid and Cooperative Computing (GCC). 2005. Available at: https://studylib.net/doc/5917238/innovative-partial-discharge-measurement (accessed 19 May 2023)
28. Polyakov D.A., Pugach V.N., Nikitin K.I., Yurchuk D.A. Data collection system for monitoring the state of XLPE cable insulation. Omskiy nauchnyi vestnik [Omsk Scientific Bulletin], 2017, no. 4 (154), pp. 70-73. (In Russian)
29. Kaggle.com. Available at: https://www.kaggle.com/c/vsb-power-line-fault-detection/data (accessed 19 May 2023)
30. Misak S., Pokorný V. Testing of a Covered Conductor’s Fault Detectors. IEEE Transactions on Power Delivery. 2015, no. 3(30), pp. 1096-1103. doi: 10.1109/TPWRD.2014.2357072
31. Chollet F. Glubokoe obuchenie na Python [Deep learning with python]. Saint Petersburg, Piter Publ., 2018. 400 p. (In Russian)
32. Verzunov S.N., Raimzhanov Kh.A. Comparison of deep neural networks based on various pre-trained CNNs for di-agnosing COVID-19 using X-rays. Problemy avtomatiki i upravleniya [Automation and control problems], 2021, no. 1(40), pp. 12-25. (In Russian)
33. Strathprints.strath.ac.uk. Available at: https://strathprints.strath.ac.uk/56089/. (accessed 19 May 2023)
34. Perotoni M.В., Marconi K. SDR-Based Spectrum Analyzer Based in Open-Source GNU Radio. Journal of Microwaves, Optoelectronics and Electromagnetic Applications. 2021, no. 20(3), pp. 542-555. doi: 10.1590/2179-10742021v20i31194
35. IEC TS 62478. High Voltage Test Techniques – Measurement of Partial Discharges by Electromagnetic and Acoustic Methods. International Electrotechnical Commission, 2016. 54 р.
Verzunov S.N., Bochkarev I.V. System for Diagnosing the Occurrence of a Partial Discharge in Overhead Power Networks Based on Data Mining in the HF Band. Elektrotekhnicheskie sistemy i kompleksy [Electrotechnical Systems and Complexes], 2023, no. 3(60), pp. 4-14. (In Russian). https://doi.org/10.18503/2311-8318-2023-3(60)-4-14