A Partial Discharge Fault Detection Method Based on Wavelet Packet Transform, ReliefF Feature Selection and Ensemble Learning Algorithm

Öz

Many faults can occur in power transmission lines. Especially in power transmission lines, faults occur frequently as a result of phase conductors coming into contact with environmental vegetation and each other. Insulated conductors are widely used, especially in power transmission lines, in order to prevent malfunctions that may occur in this way. However, deformations that may occur in the insulating material of insulated conductors may cause events called partial discharge (PD) in these conductors. PD need to be detected quickly in order to prevent much larger failures that may occur. In this paper, an effective detection approach based on wavelet packet transform (WPT), ReliefF feature selection approach and ensemble learning algorithm classifier is proposed for the detection of PD in the transmission line. The most important advantage of this approach is to obtain features based on effective frequency bands by using WPT of PD data and to select features with high detection performance among these features by using ReliefF approach. The proposed detection system is tested using the VSB real dataset and an accuracy rate of 89,22% is obtained. When compared with similar studies using VSB dataset in the literature, it has been seen that the performance is quite high and the proposed approach has an effective performance for PD detection.

Anahtar Kelimeler

• Partial discharge
• fault detection
• wavelet packet transform
• ReliefF
• ensemble learning algorithm.

Dalgacık Paket Dönüşümü, ReliefF Özellik Seçimi ve Topluluk Öğrenme Algoritması Tabanlı Bir Kısmi Deşarj Arızası Tespit Yöntemi

Öz

Enerji nakil hatlarında birçok arıza olayı meydana gelebilmektedir. Özellikle hatlarda faz iletkenlerinin çevresel bitki örtüleriyle ve birbirleriyle temas etmeleri sonucunda oluşan arızalar sıklıkla meydana gelir. Bu şekilde oluşabilecek arızaların önüne geçebilmek için özellikle enerji nakil hatlarında izolasyonlu iletkenler yaygın olarak kullanılmaktadır. Ancak izolasyonlu iletkenlerin yalıtım malzemesinde meydana gelebilecek deformasyonlar bu iletkenlerde kısmi deşarj (KD) adı verilen olaylara sebep olabilirler. Oluşabilecek çok daha büyük arızaların önüne geçebilmek için KD’lerin hızlı bir şekilde tespit edilmesi gerekir. Bu çalışmada, iletim hatlarında meydana gelen KD’lerin tespiti için dalgacık paket dönüşümü (DPD), ReliefF özellik seçim yaklaşımı ve topluluk öğrenme algoritma sınıflandırıcı tabanlı etkili bir tespit yaklaşımı önerilmiştir. Bu yaklaşımın en önemli özelliği, KD verilerinin DPD kullanarak etkili frekans bantlarına dayanan özellikler elde edilmesi ve ReliefF yaklaşımı kullanılarak bu özellikler içerisinden tespit performansı yüksek özelliklerin seçilmesidir. Önerilen tespit sistemi VSB gerçek veri seti kullanılarak test edilmiş ve 89.22% doğruluk oranı elde edilmiştir. Literatürde VSB veri seti kullanan benzer çalışmalarla karşılaştırıldığında başarımın oldukça yüksek olduğu ve önerilen yaklaşımın KD tespiti için etkili bir performans sergilediği görülmüştür.

Anahtar Kelimeler

• Kısmi deşarj
• arıza tespiti
• dalgacık paket dönüşümü
• ReliefF
• topluluk öğrenme algoritması

Kaynakça

1. Elmasry, W., Wadi, M. EDLA-EFDS: A novel ensemble deep learning approach for electrical fault detection systems. Electric Power Systems Research 2022; 207: 107834.
2. Elmasry, W., Wadi, M. Detection of faults in electrical power grids using an enhanced anomaly-based method. Arabian Journal for Science and Engineering 2022; 47(11): 14899–14914.
3. Wang, W., Yu, N. Partial discharge detection with convolutional neural networks. In 2020 International Conference on Probabilistic Methods Applied to Power Systems (PMAPS) August 2020; EEE. pp. 1–6.
4. Xu, N. , Gooi, H. B., Wang, L., Zheng, Y., Wang, W., Yang, J. Loop Optimization Noise-Reduced LSTM Based Classifier for PD Detection, IEEE Transactions on Industry Applications 2023; 59(1): 392–402.
5. Xi, Y., Tang, X., Li, Z., Shen, Y., Zeng, X. Fault detection and classification on insulated overhead conductors based on MCNN‐LSTM. IET Renewable Power Generation 2022; 16(7): 1425–1433.
6. Huang, C., Ding, S., Li, S., Liu, R. LMFE: Learning-Based Multiscale Feature Engineering in Partial Discharge Detection. IEEE Transactions on Neural Networks and Learning Systems 2022; doi: 10.1109/TNNLS.2022.3222671.
7. Bajwa, B., Butani, C., Patel, C. A novel approach towards predicting faults in power systems using machine learning. Electrical Engineering 2022; 104: 363–368.
8. Michau, G., Hsu, C. C., Fink, O. Interpretable detection of partial discharge in power lines with deep learning. Sensors 2021; 21(6): 2154.

9. Tehrani, P., Levorato, M. Frequency-based multi task learning with attention mechanism for fault detection in power systems. In 2020 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm) November, 2020; IEEE. pp. 1–6.
10. Ahmad, D., Wang, S., Alam, M. Long short term memory based deep learning method for fault power line detection in a MV overhead lines with covered conductors. In 2020 21st National Power Systems Conference (NPSC) December 2020; IEEE. pp. 1–4.
11. Kalanidhi, K., Baskar, D., Kumar, V. Transmission Power Line Fault Detection using Convolutional Neural Networks. In Proceedings of the First International Conference on Computing, Communication and Control System, I3CAC 2021, 7-8 June 2021; Bharath University, Chennai, India.
12. Qu, N., Li, Z., Zuo, J., Chen, J. Fault detection on insulated overhead conductors based on DWT-LSTM and partial discharge. IEEE Access 2020; 8: 87060–87070.
13. Wadi, M. Fault detection in power grids based on improved supervised machine learning binary classification. Journal of Electrical Engineering 2021;72(5): 315–322.
14. VSB Power Line Fault Detection, Kaggle, 2018, (https://www.kaggle.com/c/vsb-power-line-fault-detection/data).
15. ENET Centre, VSB, 2020, (https://cenet.vsb.cz/en/).
16. Chui, C. K.. An introduction to wavelets (Vol. 1). Academic press, 1992.
17. Daubechies, I. Ten lectures on wavelets. Society for industrial and applied mathematics, 1992.
18. Desai, R., Porob, P., Rebelo, P., Edla, D. R., Bablani, A. EEG data classification for mental state analysis using wavelet packet transform and Gaussian process classifier. Wireless Personal Communications 2020;115(3): 2149–2169.
19. Khushaba, R. N., Kodagoda, S., Lal, S., Dissanayake, G. Driver drowsiness classification using fuzzy wavelet-packet-based feature-extraction algorithm. IEEE transactions on biomedical engineering 2010; 58(1): 121–131.
20. Kira, K., Rendell, L. A. (1992, July). The feature selection problem: Traditional methods and a new algorithm. In Aaai 1992: 2(1992a); 129–134.
21. Kononenko, I. Estimating attributes: Analysis and extensions of RELIEF. In ECML, April 1994: 94, pp. 171–182.
22. Tuncer, T., Ertam, F. Neighborhood component analysis and reliefF based survival recognition methods for Hepatocellular carcinoma. Physica A: Statistical Mechanics and its Applications 2020; 540: 123143.
23. Özyurt, F. Efficient deep feature selection for remote sensing image recognition with fused deep learning architectures. The Journal of Supercomputing 2020; 76(11) : 8413–8431.
24. Kapucu, C., Cubukcu, M. A supervised ensemble learning method for fault diagnosis in photovoltaic strings. Energy 2021; 227: 120463.
25. Dietterich, T. G. Ensemble methods in machine learning. In Multiple Classifier Systems: First International Workshop 2000, MCS 2000 Cagliari, Italy, June 21–23, pp. 1–15.
26. He, L., Cheng, Y., Li, Y., Li, F., Fan, K., Li, Y. An improved method for soil moisture monitoring with ensemble learning methods over the Tibetan plateau. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2021; 14: 2833–2844.
27. [Yu, L., Wang, S., Lai, K. K. Credit risk assessment with a multistage neural network ensemble learning approach. Expert systems with applications 2008; 34(2) : 1434–1444.
28. Yu, Z., Chen, H., You, J., Wong, H. S., Liu, J., Li, L., Han, G. Double selection based semi-supervised clustering ensemble for tumor clustering from gene expression profiles. IEEE/ACM transactions on computational biology and bioinformatics 2014; 11(4) : 727–740.
29. Wang, H., Ma, J. Wang G., Hao J., Ma J., Jiang H. A comparative assessment of ensemble learning for credit scoring, Expert Systems with Applications 2011; 38(1) : 223–230.
30. Akram, V. K., Taşer, P. Y. Telsiz Duyarga Ağlarda Bizans Saldırılarının Topluluk Öğrenme-tabanlı Tespiti. Dokuz Eylül Üniversitesi Mühendislik Fakültesi Fen ve Mühendislik Dergisi 2020; 22(6) : 905–918.
31. Akcan, F., Sertbaş, A. Topluluk Öğrenmesi Yöntemleri ile Göğüs Kanseri Teşhisi. Electronic Turkish Studies( 2021; 16(2): 511–528.
32. Hossin, M., Sulaiman, M. N. A review on evaluation metrics for data classification evaluations. International journal of data mining & knowledge management process 2015; 5(2) : 1–11.
33. N. V. Chawla, N. Japkowicz and A. Kotcz. Special issue on learning from imbalanced data sets. ACM SIGKDD Explorations Newsletter 2004; 6(1): 1–6.
34. Dong, M., Sun, J. Partial discharge detection on aerial covered conductors using time-series decomposition and long short-term memory network. Electric Power Systems Research 2020; 184: 106318.
35. Wadi, M., Elmasry, W. An anomaly-based technique for fault detection in power system networks. In 2021 International Conference on Electric Power Engineering–Palestine (ICEPE–P) 2021; IEEE. pp. 1–6.
36. Vantuch, T., Prílepok, M., Fulneček, J., Hrbáč, R., Mišák, S. Towards the text compression based feature extraction in high impedance fault detection. Energies 2019;12(11): 2148.
37. Li, Z., Qu, N., Li, X., Zuo, J., Yin, Y. Partial discharge detection of insulated conductors based on CNN-LSTM of attention mechanisms. Journal of Power Electronics 2021; 21: 1030–1040.