Estimation of Audible-Noise Level in Transformers Aging with Artificial Neural Networks

Öz

In this study, the noises, temperatures and loads of 8 transformers in Istanbul İkitelli Organized Industrial Zone, as well as ambient conditions, were monitored at different moments over a period of 4 months and their data were recorded. It has been observed that the noise levels give different results according to the changing environmental conditions. The received data were determined as inputs and desired targets in MATLAB environment. The data were modeled with feed forward artificial neural network and recurrent artificial neural network types and different training percentages. The estimation results closest to the target were tried to be obtained. As a result of the study, it was seen that the recurrent neural network model was better and it gave better results when the training percentage needed for the network kept high.

Anahtar Kelimeler

• Neural Networks
• Transformer Noise
• Noise Estimation
• Feed Forward

Trafo Yaşlanmasında Duyulabilir Gürültü Seviyesinin Yapay Sinir Ağları ile Tahmini

Öz

Bu çalışmada İstanbul İkitelli Organize Sanayi Bölgesi'nde bulunan 8 adet trafonun gürültü, sıcaklık ve yükleri ile ortam koşulları 4 ay boyunca farklı anlarda izlenerek verileri kayıt altına alınmıştır. Gürültü düzeylerinin değişen çevre koşullarına göre farklı sonuçlar verdiği gözlemlenmiştir. Alınan veriler MATLAB ortamında girdi ve istenilen hedefler olarak belirlendi. Veriler ileri beslemeli yapay sinir ağı ve tekrarlayan yapay sinir ağı türleri ve farklı eğitim yüzdeleri ile modellenmiştir. Hedefe en yakın tahmin sonuçları elde edilmeye çalışıldı. Çalışma sonucunda tekrarlayan sinir ağı modelinin daha iyi olduğu ve eğitim yüzdesi yüksek tutulduğunda daha iyi sonuçlar verdiği görülmüştür.

Anahtar Kelimeler

• Sinir ağları
• İleri besleme
• Transformatör gürültüsü
• Gürültü kestrimi

Kaynakça

1. [1] P. Shuai and J. Biela, "Investigation of acoustic noise sources in medium frequency, medium voltage transformers," 2014 16th European Conference on Power Electronics and Applications, Lappeenranta, Finland, 2014, pp. 1-11, doi: 10.1109/EPE.2014.6910949
2. [2] R. B. George, «Power Transformer Noise Its Characteristics and Reduction,» Transactions of the American Institute of Electrical Engineers , cilt 50, no. 1, pp. 347 - 352, 1931.
3. [3] McDonald, J.D., Electric Power Substations Engineering, Taylor & Francis Group, 2012, pp. 191-194.
4. [4] Mizokami, M., & Kurosaki, Y. (2016). Variation of noise and magnetostriction associated with joint types of transformer core. Electrical Engineering in Japan, 194(2), 1-8. Https://Doi.Org/10.1002/Eej.2273
5. [5] Z. Lu et al., "Measurement and analysis of UHV transformer noise with sound intensity and vibration method," 2017 20th International Conference on Electrical Machines and Systems (ICEMS), Sydney, NSW, Australia, 2017, pp. 1-4, doi: 10.1109/ICEMS.2017.8056537.
6. [6] T. Yanada, S. Minowa, O. Ichinokura and S. Kikuchi, "Design and analysis of noise-reduction transformer based on equivalent circuit," in IEEE Transactions on Magnetics, vol. 34, no. 4, pp. 1351-1353, July 1998, doi: 10.1109/20.706545.
7. [7] Zhang, X. and Sun, Z. (2023). Application of improved pnn in transformer fault diagnosis. Processes, 11(2), 474. https://doi.org/10.3390/pr11020474
8. [8] M. S. Uddin, K. K. Halder, M. Tahtali, A. J. Lambert and M. R. Pickering, "Speckle reduction and deblurring of ultrasound images using artificial neural network," 2015 Picture Coding Symposium (PCS), Cairns, QLD, Australia, 2015, pp. 105-108, doi: 10.1109/PCS.2015.7170056.

9. [9] Ledesma, S., Ibarra-Manzano, M. A., Almanza-Ojeda, D. L., Fallavollita, P., & Steffener, J. (2021). Artificial Intelligence to Analyze the Cortical Thickness Through Age. Frontiers in artificial intelligence, 4, 549255. https://doi.org/10.3389/frai.2021.549255
10. [10] He, Y., Zhou, Q., Lin, S., & Zhao, L. (2020). Validity Evaluation Method Based on Data Driving for On-Line Monitoring Data of Transformer under DC-Bias. Sensors, 20(15), 4321. https://doi.org/10.3390/s20154321
11. [11] F. Aghaeipoor, M. Mohammadi and V. S. Naeini, "Target tracking in noisy wireless sensor network using artificial neural network," 7'th International Symposium on Telecommunications (IST'2014), Tehran, Iran, 2014, pp. 720-724, doi: 10.1109/ISTEL.2014.7000796.
12. [12] Yasin Günlemiş, Trafo ses seviyesinin yapay sinir ağları ile tahmini, Yüksek Lisans Tezi, Istanbul Üniversitesi-Cerrahpaşa, LEE, 2021 (in Turkish)
13. [13] A. S. Farag, M. Mohandes and A. Al-Shaikh, "Diagnosing failed distribution transformers using neural networks," in IEEE Transactions on Power Delivery, vol. 16, no. 4, pp. 631-636, Oct. 2001, doi: 10.1109/61.956749. [14] Elmas, Ç., Yapay Zeka Uygulamaları, Seçkin Yayıncılık, 2011.
14. [15] Çakır, Fatma Sönmez. Yapay Sinir Ağları Matlab Kodları ve Matlab Toolbox Çözümleri. Nobel Akademik Yayıncılık, 2018
15. [16] Öztemel, E., Yapay Sinir Ağları, İstanbul: Papatya Yayıncılık, 2012.
16. [17] S. Haykin, Neural Networks and Learning Machines, New Jersey: Pearson Education, 2009.
17. [18] Pascanu, R., Mikolov, T., Bengio, Y. (2013). On the difficulty of training recurrent neural networks. Proceedings of the 30th International Conference on Machine Learning, in Proceedings of Machine Learning Research, 28(3):1310-1318 https://proceedings.mlr.press/v28/pascanu13.html.
18. [19] P. Abhigna, S. Jerritta, R. Srinivasan and V. Rajendran, "Analysis of feed forward and recurrent neural networks in predicting the significant wave height at the moored buoys in Bay of Bengal," 2017 International Conference on Communication and Signal Processing (ICCSP), Chennai, India, 2017, pp. 1856-1860, doi: 10.1109/ICCSP.2017.8286717