Fault diagnosis of power transformers based on dissolved gas analysis and multi-kernel graph convolution network integrated with dual-channel classifiers

Xuebin  Lv; Fuzheng  Liu; Mingshun Jiang; Faye  Zhang; Lei  Jia

doi:10.6180/jase.202510_28(10).0014

Fault diagnosis of power transformers based on dissolved gas analysis and multi-kernel graph convolution network integrated with dual-channel classifiers

Xuebin Lv^1,2, Fuzheng Liu¹, Mingshun Jiang¹, Faye Zhang¹This email address is being protected from spambots. You need JavaScript enabled to view it., and Lei Jia¹

¹School of Control Science and Engineering, Shandong University, Jinan, 250061, China

²Equipment Department, State Grid Shandong Electric Power Company, Jinan, 250021, China

Received: September 11, 2024
Accepted: December 21, 2024
Publication Date: January 24, 2025

Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Download Citation: ||https://doi.org/10.6180/jase.202510_28(10).0014

Apower transformer fault diagnosis method based on dissolved gas analysis and multi-kernel graph con volution network integrated with dual-channel classifiers (DM-DC) is proposed to address the problems of insufficient accuracy and large deviation in recognition when dealing with imbalanced data. Firstly, construct multi-dimensional supplementary feature vectors adopting multi-feature dissolved gas ratio analysis to enrich the characterization features of transformers. Secondly, extract and model sample features deeply adopting graph generation network and multi-kernel graph convolution network to further explore the relationship between representation features and fault samples. Finally, a dual-channel classification network composed of binary classifier and multi-class classifier is introduced to alleviate the training bias towards the majority class and improve the model’s ability to handle imbalanced data through sample level reweighting. The various experiments shows that the proposed DM-DC can effectively solve the problem of low accuracy of minority class samples, with superior overall diagnosis accuracy performance, and is suitable for power transformer fault identification.

Keywords: Dissolved gas analysis (DGA); power transformer; fault diagnosis; multi-kernel graph convolution network (MGCN); dual-channel classifiers (DC)

[1] H.Shang,Z.Liu,Y.Wei,andS.Zhang,(2024)“Anovel faultdiagnosismethodforapowertransformerbasedon multi-scaleapproximateentropyandoptimizedconvolu tionalnetworks"Entropy26(3):186.DOI:10.3390/e26030186.
[2] O.Kherif,Y.Benmahamed,M.Teguar,A.Boubakeur, andS.S.Ghoneim,(2021)“Accuracyimprovementof powertransformerfaultsdiagnosticusingKNNclassi fierwithdecisiontreeprinciple"IEEEAccess9:81693 81701.DOI:10.1109/ACCESS.2021.3086135.
[3] B.Raja,G.Venkatakrishnan,andR.Rengaraj,(2024) “Powertransformerfaultdiagnosisandconditionmonitor ingusinghybridTDO-SNNtechnique"International JournalofHydrogenEnergy68:1370–1381.DOI:10.1016/j.ijhydene.2024.03.115.
[4] G.Wang,Y.Liu,X.Chen,Q.Yan,H. Sui,C.Ma, andJ.Zhang,(2021)“Powertransformerfaultdiagnosis systembasedonInternetofThings"EurasipJournalon WirelessCommunicationsandNetworking2021: 1–24.DOI:10.1186/s13638-020-01871-6.
[5] M.Demirci,H.Gözde, andM.C.Taplamacioglu, (2023)“Improvementofpowertransformerfaultdiagnosis byusingsequentialKalmanfiltersensorfusion"Interna tionalJournalofElectricalPower&EnergySystems 149:109038.DOI:10.1016/j.ijepes.2023.109038.
[6] A.Rangel Bessa, J. Farias Fardin, P. Marques Ciarelli, and L. Frizera Encarnação, (2023) “Conventional dis solved gases analysis in power transformers" Energies 16(21): 7219. DOI: 10.3390/en16217219.
[7] A.Wajid,A.U.Rehman,S.Iqbal, M.Pushkarna, S. M. Hussain, H. Kotb, M. Alharbi, and I. Zaitsev, (2023) “Comparative performance study of dissolved gas analysis (DGA) methods for identification of faults in power trans former" International Journal of Energy Research 2023(1): 9960743. DOI: 10.1155/2023/9960743.
[8] X.Han,S.Huang,X.Zhang,Y.Zhu,G.An,andZ.Du, (2023) “Atransformer condition recognition method based on dissolved gas analysis features selection and multiple models fusion" Engineering Applications of Artificial Intelligence 123: 106518. DOI: 10.1016/j.engappai.2023.106518.
[9] N.A.Bakar,I. S. Chairul, S. Ab Ghani, M. S. A. Khiar, and M. Z. C. Wanik, (2023) “Improvement of trans former dissolved gas analysis interpretation using J48 decision tree model" IAES International Journal of Artificial Intelligence 12(1): 48. DOI: 10.11591/ijai.v12.i1.pp48-56.
[10] Y. Jin, H. Wu, J. Zheng, J. Zhang, and Z. Liu, (2023) “Power transformer fault diagnosis based on improved BP neural network" Electronics 12(16): 3526. DOI: 10.3390/electronics12163526.
[11] Z. Xing and Y. He, (2023) “Multimodal Mutual Neural Network for Health Assessment of Power Transformer" IEEE Systems Journal 17(2): 2664–2673. DOI: 10.1109/JSYST.2023.3237225.
[12] I. B. Taha, S. Ibrahim, and D.-E. A. Mansour, (2021) “Power transformer fault diagnosis based on DGA using a convolutional neural network with noise in measure ments" IEEE Access 9: 111162–111170. DOI: 10.1109/ACCESS.2021.3102415.
[13] W. Zeng, Y. Cao, L. Feng, J. Fan, M. Zhong, W. Mo, and Z. Tan, (2023) “Hybrid CEEMDAN-DBN-ELM for online DGA serials and transformer status forecasting" Electric Power Systems Research 217: 109176. DOI: 10.1016/j.epsr.2023.109176.
[14] Y. Zhou, Y. Liu, N. Wang, X. Han, and J. Li, (2022) “Partial discharge ultrasonic signals pattern recognition in transformer using BSO-SVM based on microfiber coupler sensor" Measurement 201: 111737. DOI: 10.1016/j.measurement.2022.111737.
[15] X.Yang,W.Chen,A.Li,C.Yang,Z.Xie,andH.Dong, (2019) “BA-PNN-based methods for power transformer fault diagnosis" Advanced engineering informatics 39: 178–185. DOI: 10.1016/j.aei.2019.01.001.
[16] R. A. Prasojo, M. A. A. Putra, M. E. Apriyani, A. N. Rahmanto, S. S. Ghoneim, K. Mahmoud, M. Lehto nen, M. M. Darwish, et al., (2023) “Precise transformer fault diagnosis via random forest model enhanced by syn thetic minority over-sampling technique" Electric Power Systems Research 220: 109361. DOI: 10.1016/j.epsr.2023.109361.
[17] F. Lv, G. Liu, Q. Wang, X. Lu, S. Lei, S. Wang, and K. Ma, (2023) “Pattern recognition of partial discharge in power transformer based on infogan and cnn" Journal of Electrical Engineering & Technology 18(2): 829–841. DOI: 10.1007/s42835-022-01260-7.
[18] W. Liao, B. Bak-Jensen, J. R. Pillai, Y. Wang, and Y. Wang, (2021) “A review of graph neural networks and their applications in power systems" Journal of Modern Power Systems and Clean Energy 10(2): 345–360. DOI: 10.35833/MPCE.2021.000058.
[19] M.S. Ali, A. Omar, A. S. A. Jaafar, S. H. Mohamed, et al., (2023) “Conventional methods of dissolved gas analysis using oil-immersed power transformer for fault diagnosis: A review" Electric Power Systems Research 216: 109064. DOI: 10.1016/j.epsr.2022.109064.
[20] W.Liao, D. Yang, Y. Wang, and X. Ren, (2020) “Fault diagnosis of power transformers using graph convolu tional network" CSEE Journal of Power and Energy Systems 7(2): 241–249. DOI: 10.17775/CSEEJPES.2020.04120.
[21] T. Li, Z. Zhao, C. Sun, R. Yan, and X. Chen, (2020) “Multireceptive field graph convolutional networks for ma chine fault diagnosis" IEEE Transactions on Industrial Electronics 68(12): 12739–12749. DOI: 10.1109/TIE.2020.3040669.
[22] A. Nanfak, S. Eke, F. Meghnefi, I. Fofana, G. M. Ngaleu, and C. H. Kom, (2023) “Hybrid DGA method for power transformer faults diagnosis based on evolution ary k-means clustering and dissolved gas subsets analysis" IEEE Transactions on Dielectrics and Electrical In sulation 30(5): 2421–2428. DOI: 10.1109/TDEI.2023.3275119.
[23] I. B. Taha, A. Hoballah, and S. S. Ghoneim, (2020) “Optimal ratio limits of rogers’ four-ratios and IEC 60599 code methods using particle swarm optimization fuzzy logic approach" IEEE Transactions on Dielectrics and Electrical Insulation 27(1): 222–230. DOI: 10.1109/TDEI.2019.008395.
[24] S. A. Ward, A. El-Faraskoury, M. Badawi, S. A. Ibrahim, K. Mahmoud, M. Lehtonen, and M. M. Dar wish, (2021) “Towards precise interpretation of oil trans formers via novel combined techniques based on DGA and partial discharge sensors" Sensors 21(6): 2223. DOI: 10.3390/s21062223.
[25] S. Misbahulmunir, V. K. Ramachandaramurthy, and Y. H.M.Thayoob,(2020)“Improvedself-organizing map clustering of power transformer dissolved gas analysis using inputs pre-processing" IEEE Access 8: 71798 71811. DOI: 10.1109/ACCESS.2020.2986726.
[26] I. B. Taha, D.-E. A. Mansour, S. S. Ghoneim, and N. I. Elkalashy, (2017) “Conditional probability-based inter pretation of dissolved gas analysis for transformer incipi ent faults" IET Generation, Transmission&Distribu tion 11(4): 943–951. DOI: 10.1049/iet-gtd.2016.0886.