Binary classification of operating state of power transformers using Performance Index and Machine Learning Models

Authors
DOI:
https://doi.org/10.47456/bjpe.v12i1.49144
Keywords:
Classifier, Machine Learning, Performance Index, Power Transformer, Predictive Technique
Abstract

Power transformers are strategic and valuable assets in electrical systems, as their unexpected failures can lead to significant operational and financial losses for power sector companies and consumers. Although there have been advancements in monitoring their operational conditions, some methodologies still require specialized interpretation, lack standardization, or adopt models whose complexity can hinder integration with the usual operational practices of maintenance professionals. In this context, the objective of this work is to develop binary classifiers based on machine learning algorithms for fast and efficient prediction of the operational state of power transformers, labeled as Satisfactory or Unsatisfactory, using data derived from physicochemical tests, Dissolved Gas Analysis (DGA), and performance indices, based on real equipment samples. The methodology involves the development of supervised machine learning models, such as Random Forest, HistGradientBoosting, Balanced Logistic Regression, and XGBoost, implemented with stratified cross-validation. The results indicate that the classifiers can satisfactorily identify transformers in critical condition, even in a scenario with considerable data dispersion. Therefore, the proposed approach represents a promising tool for technical decision-making in preventive maintenance strategies, combining reliability, scalability, and ease of application in field environments.

Author Biographies
  1. Vinícius Faria Costa Mendanha, Federal University of Goiás

    Vinícius Faria Costa Mendanha nasceu em 19 de maio de 1998. Atualmente, é mestrando no Programa de Pós-Graduação em Engenharia Elétrica e de Computação da Universidade Federal de Goiás. Graduou-se em Engenharia Elétrica pela Universidade Federal de Goiás (UFG) em 2023. Em 2018-2019, integrou a Agremiação Politécnica da UFG, na área de Administração. Em 2019, ingressou no PET (Programa de Educação Tutorial) Engenharias Conexão de Saberes, realizando também Iniciação Científica. Estudou análise de sinais e fenômeno de Gibbs (2019-2020). Dedica-se ao estudo de Inteligência Computacional aplicada à área de manutenção de transformadores de potência e é membro da Equipe do Laboratório de Pesquisa em Engenharia de Alta Tensão - LAPEAT UFG. https://orcid.org/0000-0003-0982-9328

  2. André Pereira Marques, Federal University of Goiás

    André Pereira Marques nasceu em 25 de fevereiro de 1961 em Araguari, Minas Gerais, Brasil. Doutor (2018) e Mestre (2004) em Engenharia Elétrica pela Escola de Engenharia Elétrica, Mecânica e de Computação (EMC) da Universidade Federal de Goiás (UFG). Ele é Professor Titular no Instituto Federal de Goiás – Campus Goiânia (IFG) desde 1990, trabalhando nos cursos de Engenharia Elétrica, Engenharia de Automação e Curso Técnico em Eletrotécnica. Ele também é Gerente Técnico e proprietário da empresa APMarques Consultoria e Capacitação em Engenharia Elétrica desde 2019, especializado em estudos de carregamento e diagnósticos em transformadores de potência e é membro da Equipe do Laboratório de Pesquisa em Engenharia de Alta Tensão - LAPEAT UFG https://orcid.org/0000-0001-6641-7835

  3. Cacilda de Jesus Ribeiro, Federal University of Goiás

    Cacilda de J. Ribeiro nasceu em 8 de agosto de 1971 em Matão, São Paulo, Brasil. Doutora (2002) e Pós-Doutora (2004) em Engenharia Elétrica pela Escola de Engenharia de São Carlos, da Universidade de São Paulo. Atualmente, é professora titular na Escola de Engenharia Elétrica, Mecânica e de Computação (EMC) na Universidade Federal de Goiás (UFG), e Coordenadora do Grupo de Trabalho Mulheres nas Engenharias GTME EMC UFG e do Laboratório de Pesquisa em Engenharia de Alta Tensão - LAPEAT UFG. https://orcid.org/0000-0002-8725-3443

References

Arias, R. (2020). Data for: Root cause analysis improved with machine learning for failure analysis in power transformers [Dataset]. Mendeley. https://doi.org/10.17632/RZ75W3FKXY.1

Barbosa, F. R., Almeida, O. M., Braga, A. P. S., Amora, M. A. B., & Cartaxo, S. J. M. (2012). Application of an artificial neural network in the use of physicochemical properties as a low-cost proxy of power transformers DGA data. IEEE Transactions on Dielectrics and Electrical Insulation, 19(1), 239-246. https://doi.org/10.1109/TDEI.2012.6148524

Bechara, R. (2010). Análise de falhas de transformadores de potência [Dissertação (Mestrado)]. Universidade de São Paulo.

Bishop, C. M. (2006). Pattern recognition and machine learning. Springer.

Campi, R. L. (2014). Modelagem fuzzy da concentração dos gases dissolvidos em óleo mineral isolante de transformadores baseada em resultados de ensaios físico-químicos [Dissertação (Mestrado)]. Universidade de São Paulo.

Castro, A. R. G. & Miranda, V. (2005). Knowledge discovery in neural networks with application to transformer failure diagnosis. IEEE Transactions on Power Systems, 20(2), 717-724. https://doi.org/10.1109/TPWRS.2005.846074

Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic Minority Over-sampling Technique. Journal of Artificial Intelligence Research, 16, 321-357. https://doi.org/10.1613/jair.953

Cherif, I. L., & Kortebi, A. (2019). On using extreme gradient boosting (XGBoost) machine learning algorithm for home network traffic classification. Wireless Days (WD), 1-6. https://doi.org/10.1109/WD.2019.8734193

Dempster, A., Webb, G. I., & Schmidt, D. F. (2025). Prevalidated ridge regression is a highly efficient drop-in replacement for logistic regression for high-dimensional data (No. arXiv:2401.15610). arXiv. https://doi.org/10.48550/arXiv.2401.15610

Dias, Y. A. (2023). Method for predicting the performance indices of power transformers immersed in mineral insulating oil and medium and high-voltage circuit breakers [Tese (Doutorado)]. Universidade Federal de Goiás.

Duraisamy, V., Devarajan, N., Somasundareswari, D., Vasanth, A. A. M., & Sivanandam, S. N. (2007). Neuro fuzzy schemes for fault detection in power transformer. Applied Soft Computing, 7(2), 534-539. https://doi.org/10.1016/j.asoc.2006.10.001

Duval, M. (2002). A review of faults detectable by gas-in-oil analysis in transformers. IEEE Electrical Insulation Magazine, 18(3), 8-17. https://doi.org/10.1109/MEI.2002.1014963

Duval, M. & dePabla, A. (2001). Interpretation of gas-in-oil analysis using new IEC publication 60599 and IEC TC 10 databases. IEEE Electrical Insulation Magazine, 17(2), 31-41. https://doi.org/10.1109/57.917529

Han, J., Kamber, M., & Pei, J. (2011). Data Mining: Concepts and Techniques (3o ed). Elsevier.

IEEE. (2019). IEEE Guide for the interpretation of gases generated in mineral oil-immersed transformers. IEEE. https://doi.org/10.1109/IEEESTD.2019.8890040

Shen, Z., Hassani, H., Kale, S., & Karbasi, A. (2022). Federated functional gradient boosting. Proceedings of the 25th International Conference on Artificial Intelligence and Statistics, 151, 7814-7840. Recuperado de https://proceedings.mlr.press/v151/shen22a.html

Li, J., Heap, A. D., Potter, A., & Daniell, J. J. (2011). Application of machine learning methods to spatial interpolation of environmental variables. Environmental Modelling & Software, 26(12), 1647-1659. https://doi.org/10.1016/j.envsoft.2011.07.004

Marques, A. P. (2018). Optimized diagnosis of power transformers through the integration of predictive techniques [Tese (Doutorado)]. Universidade Federal de Goiás.

Marques, A. P., Azevedo, C. H. B., Santos, J. A. L., Machado, S. G., Ribeiro, C. J. R., Moura, N. K., Dias, Y. A., & Brito, L. C. (2017). Metodologia para Reenergização de Transformadores de Potência após Interrupções não Programadas no Sistema Elétrico. XXIV Seminário Nacional de Produção e Transmissão de Energia Elétrica (SNPTEE).

Marques, A. P., Azevedo, C. H. B., Santos, J. A.L. dos., Sousa, F. R. de C., Moura, N. K., Ribeiro, C. J., Dias, Y. A., Rodrigues, A., Rocha, A. S., & da C. Brito, L. da C. (2017). Insulation resistance of power transformers - method for optimized analysis. 2017 IEEE 19th International Conference on Dielectric Liquids (ICDL), Manchester, UK, pp. 1-4, https://doi.org/10.1109/ICDL.2017.8124629

Marques, A. P., Moura, N. K. de., Dias, Y. A., Jesus Ribeiro, C. de., Rocha, A. S., Brito, L. da C., Azevedo, C. H. B., & Santos, J. A. L. dos. (2017). Method for the evaluation and classification of power transformer insulating oil based on physicochemical analyses. IEEE Electrical Insulation Magazine, 33(1), 39-49. https://doi.org/10.1109/MEI.2017.7804315

Miller, C., Hao, L., & Fu, C. (2022). Gradient boosting machines and careful pre-processing work best: ASHRAE Great Energy Predictor III lessons learned. https://doi.org/10.48550/ARXIV.2202.02898

Moura, N. K. de. (2018). Otimização Computacional da Avaliação de Resultados de Ensaios Físico-químicos em Transformadores de Potência [Dissertação (Mestrado)]. Universidade Federal de Goiás.

Naderian, A., Cress, S., Piercy, R., Wang, F., & Service, J. (2008). An approach to determine the health index of power transformers. Conference Record of the 2008 IEEE International Symposium on Electrical Insulation, 192-196. https://doi.org/10.1109/ELINSL.2008.4570308

Naresh, R., Sharma, V., & Vashisth, M. (2008). An integrated neural fuzzy approach for fault diagnosis of transformers. IEEE Transactions on Power Delivery, 23(4), 2017-2024. https://doi.org/10.1109/TPWRD.2008.2002652

Richardson, Z. J., Fitch, J., Tang, W. H., Goulermas, J. Y., & Wu, Q. H. (2008). A Probabilistic classifier for transformer dissolved gas analysis with a particle swarm optimizer. IEEE Transactions on Power Delivery, 23(2), 751-759. https://doi.org/10.1109/TPWRD.2008.915812

Rogers, R. (1978). IEEE and IEC Codes to Interpret incipient faults in transformers, using gas in oil analysis. IEEE Transactions on Electrical Insulation, EI-13(5), 349-354. https://doi.org/10.1109/TEI.1978.298141

Sekulić, A., Kilibarda, M., Heuvelink, G. B. M., Nikolić, M., & Bajat, B. (2020). Random forest spatial interpolation. Remote Sensing, 12(10), 1687. https://doi.org/10.3390/rs12101687

Senoussaoui, M. E. A., Brahami, M., & Fofana, I. (2018). Combining and comparing various machine‐learning algorithms to improve dissolved gas analysis interpretation. IET Generation, Transmission & Distribution, 12(15), 3673-3679. https://doi.org/10.1049/iet-gtd.2018.0059

Sheridan, R. P., Wang, W. M., Liaw, A., Ma, J., & Gifford, E. M. (2016). Extreme gradient boosting as a method for quantitative structure-activity relationships. Journal of Chemical Information and Modeling, 56(12), 2353-2360. https://doi.org/10.1021/acs.jcim.6b00591

Shi, H., Wang, H., Huang, Y., Zhao, L., Qin, C., & Liu, C. (2019). A hierarchical method based on weighted extreme gradient boosting in ECG heartbeat classification. Computer Methods and Programs in Biomedicine, 171, 1-10. https://doi.org/10.1016/j.cmpb.2019.02.005

Shi, Y., Ke, G., Chen, Z., Zheng, S., & Liu, T.-Y. (2023). Quantized training of gradient boosting decision trees, (No. arXiv:2207.09682). arXiv. https://doi.org/10.48550/arXiv.2207.09682

Velarde, G., Weichert, M., Deshmunkh, A., Deshmane, S., Sudhir, A., Sharma, K., & Joshi, V. (2024). Tree boosting methods for balanced and imbalanced classification and their robustness over time in risk assessment. Intelligent Systems with Applications, 22, 200354. https://doi.org/10.1016/j.iswa.2024.200354

Velásquez, R. M. A., Lara, J. V. M., & Melgar, A. (2019). Converting data into knowledge for preventing failures in power transformers. Engineering Failure Analysis, 101, 215-229. https://doi.org/10.1016/j.engfailanal.2019.03.027

Velásquez, R. M. A. & Lara, J. V. M. (2020). Root cause analysis improved with machine learning for failure analysis in power transformers. Engineering Failure Analysis, 115, 104684. https://doi.org/10.1016/j.engfailanal.2020.104684

Velásquez, R. M. A. & Mejía Lara, J. V. (2018). Corrosive Sulphur effect in power and distribution transformers failures and treatments. Engineering Failure Analysis, 92, 240-267. https://doi.org/10.1016/j.engfailanal.2018.05.018

Wang, J., Wu, K., Zhu, W., & Gu, C. (2017). Condition assessment for power transformer using Health Index. IOP Conference Series: Materials Science and Engineering, 199, 012046. https://doi.org/10.1088/1757-899X/199/1/012046

Wang, J., Xie, X., Wang, P., Sun, J., Liu, Y., & Zhang, L. (2025). Incorporating symmetric smooth regularizations into sparse logistic regression for classification and feature extraction. Symmetry, 17(2), 151. https://doi.org/10.3390/sym17020151

Wattakapaiboon, W. & Pattanadech, N. (2016). The new developed Health Index for transformer condition assessment. International Conference on Condition Monitoring and Diagnosis (CMD), 32-35. https://doi.org/10.1109/CMD.2016.7757760

Yadaiah, N. & Ravi, N. (2011). Internal fault detection techniques for power transformers. Applied Soft Computing, 11(8), 5259-5269. https://doi.org/10.1016/j.asoc.2011.05.034

Zeinoddini‐Meymand, H. & Vahidi, B. (2016). Health index calculation for power transformers using technical and economical parameters. IET Science, Measurement & Technology, 10(7), 823-830. https://doi.org/10.1049/iet-smt.2016.0184

Zhang, Y., Wei, C., & Liu, X. (2022). Group logistic regression models with lp,q regularization. Mathematics, 10(13), 2227. https://doi.org/10.3390/math10132227

Cover Image
Image featuring a blue sky with light clouds in the background and a yellow road-style warning sign in the foreground displaying the words “Machine Learning.” The sign is slightly tilted, symbolizing attention or emphasis on the topic of machine learning. At the top of the image appears the article title “Binary Classification of the Operational State of Power Transformers Using a Performance Index and Machine Learning Models,” along with the authors Marques, A. P., Mendanha, V. F. C., and Ribeiro, C. de J. (2026). In the lower corner, the identification of the Brazilian Journal of Production Engineering and the journal’s ISSN are displayed.
Downloads
Published
2026-03-11
Issue
Vol. 12 No. 1 (2026): Regular Issue (January - March)
Section
ENERGY
License

Copyright (c) 2026 Mendanha, V. F. C., Marques, A. P., & Ribeiro, C. de J.

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

All works published in the Brazilian Journal of Production Engineering (BJPE) are licensed under Creative Commons Attribution 4.0 International (CC BY 4.0). This means that: Anyone can copy, distribute, display, adapt, remix, and even commercially use the content published in the journal; Provided that due credit is given to the authors and to BJPE as the original source; No additional permission is required for reuse, as long as the license terms are respected. This policy complies with the principles of open access, promoting the broad dissemination of scientific knowledge. 🔗 Click here to access the full license

How to Cite

Mendanha, V. F. C., Marques, A. P., & Ribeiro, C. de J. (2026). Binary classification of operating state of power transformers using Performance Index and Machine Learning Models. Brazilian Journal of Production Engineering, 12(1), 120-133. https://doi.org/10.47456/bjpe.v12i1.49144