A probabilistic model for power transformer condition assessment based on integrated diagnostic features

Accurate assessment of the technical condition of power transformers is a fundamental problem in modern power systems. Conventional diagnostic approaches are often based on heuristic thresholds or expert interpretation of individual indicators, which limits their ability to account for uncertainty and measurement variability.

This paper proposes a probabilistic model for power transformer condition assessment based on an integrated set of diagnostic features. The transformer condition is modeled as a hidden random variable inferred from observable diagnostic data within a Bayesian framework. The proposed approach provides a mathematically rigorous representation of diagnostic uncertainty and enables probabilistic interpretation of condition states. Analytical properties of the model are investigated, and numerical experiments using synthetic data demonstrate robustness with respect to noise and feature correlation. The presented framework is intended for theoretical analysis and electronic modeling of transformer diagnostics.

1. . Bishop, C. M. Pattern Recognition and Machine Learning. Springer.

2. . Murphy, K. P. Machine Learning: A Probabilistic Perspective. MIT Press.

3. . Kundur, P. Power Transformer Engineering. McGraw-Hill.

4. . IEC 60599. Guide to the Interpretation of Dissolved Gas Analysis.

5. . IEEE Std C57.104. Guide for the Interpretation of Gases Generated in Oil-Immersed Transformers.

6. . Zhou, H., Zhang, Y., & Li, P. Smart IoT Applications in Transformer Condition Monitoring // IEEE Access. – 2021; 9:124557-124568.

7. . Kumar, V., & Singh, R. Artificial Intelligence for Predictive Transformer Maintenance // Electric Power Systems Research. – 2022; 208:107864.

8. . Gupta, A. Statistical Modelling of Transformer Health in IoT-Integrated Networks // Energies. – 2023; 16(4):1442.

9. . Patel, R., Desai, M., & Khan, S. IoT-Enabled SCADA Systems for Smart Substations // IEEE Transactions on Industrial Informatics. – 2021; 17(9):6045-6057.

10. . Aliyev, F., & Yusupov, D. Real-Time Transformer Monitoring Using Edge Computing // Sensors. – 2024; 24(1):52-66.

11. . Li, J., Wang, X., & Zhang, Y. Intelligent condition monitoring of power transformers using IoT-enabled sensors and machine learning models // IEEE Transactions on Industrial Informatics. – 2023; 19(8):6543-6555. https://doi.org/10.1109/TII.2023.3248761

12. . Sharma, P., & Singh, R. Integration of IoT and SCADA for real-time monitoring and fault diagnosis in electrical substations // Electric Power Systems Research. – 2022; 212:108389. https://doi.org/10.1016/j.epsr.2022.108389

13. . Kumar, S., & Khan, M. A. A hybrid AI model for transformer health prediction based on ANN and SVM classifiers // Energy Reports. – 2021; 7:876-884. https://doi.org/10.1016/j.egyr.2021.01.039

14. . Bakar, A., Mohamad, H., & Yusuf, M. Internet of Things-based monitoring architecture for oil-immersed transformers. Sensors. – 2023; 23(5):2541. https://doi.org/10.3390/s23052541

15. . Tabrizi, M., & Gholami, M. Condition assessment of power transformers using health index and fuzzy inference systems // IEEE Access. – 2022; 10:100234100245. https://doi.org/10.1109/ACCESS.2022.3204523

16. . Rahman, M. A., & Islam, S. Machine learning and deep learning approaches for transformer fault diagnosis: A comparative study // Energies. – 2021; 14(17):5329. https://doi.org/10.3390/en14175329

17. . Ahmed, M. F., & Ali, Z. Thermal modeling and lifetime prediction of power transformers under IoT-based monitoring systems // Applied Energy. – 2023; 337:120929. https://doi.org/10.1016/j.apenergy.2023.120929

18. . Zhou, H., & Lin, J. Development of cloudbased predictive maintenance systems for smart grids using IoT data analytics // Renewable and Sustainable Energy Reviews. – 2022); 169:113013. https://doi.org/10.1016/j.rser.2022.113013

19. . Raj, A., & Mehta, R. Smart transformer monitoring through wireless sensor networks and SCADA integration // Measurement. – 2023; 215:112889. https://doi.org/10.1016/j.measurement.2023.112889

20. . Chen, D., & Zhao, L. Reliability estimation of power transformers using real-time IoT data and Bayesian networks // Electric Power Components and Systems. – 2024; 52(4-5):312-326. https://doi.org/10.1080/15325008.2024.2331123

21. . Aizpurua, J. I. Power transformer dissolved gas analysis through Bayesian networks and hypothesis testing / J. I. Aizpurua, V. M. Catterson, B. G. Stewart, S. D. J. McArthur, B. Lambert, J. G. Cross // IEEE Transactions on Dielectrics and Electrical Insulation. – 2018. – Vol. 25, No. 2. – Pp. 494-506. – DOI: 10.1109/TDEI.2018.006766.

22. . Ali, M. S. A comprehensive review on dissolved gas analysis methods for power transformer condition monitoring / M. S. Ali, S. H. Bakar, N. Rahim, K. Y. Lau // IEEE Access. – 2023. – Vol. 11. – Pp. 6968869710. – DOI: 10.1109/ACCESS.2023.3291866.

23. . Arias Velásquez, R. M. Root cause analysis for power transformers failures using machine learning / R. M. Arias Velásquez, J. V. Mejía Lara // Engineering Failure Analysis. – 2020. – Vol. 115. – Art. 104684. – DOI: 10.1016/j.engfailanal.2020.104684.

24. . Bacha, K. Power transformer fault diagnosis based on dissolved gas analysis by support vector machine / K. Bacha, S. Souahlia, M. Gossa // Electric Power Systems Research. – 2012. – Vol. 83, No. 1. – Pp. 73-79. – DOI: 10.1016/j.epsr.2011.09.012.

25. . Bakar, N. A. A review of dissolved gas analysis measurement and interpretation techniques / N. A. Bakar, A. Abu-Siada, S. Islam // IEEE Electrical Insulation Magazine. – 2014. – Vol. 30, No. 3. – Pp. 39-49. – DOI: 10.1109/MEI.2014.6804740.

26. . Bengio, Y. Learning long-term dependencies with gradient descent is difficult / Y. Bengio, P. Simard, P. Frasconi // IEEE Transactions on Neural Networks. – 1994. – Vol. 5, No. 2. – Pp. 157-166.

27. . Benmahamed, Y. Accuracy improvement of transformer faults diagnostic based on DGA data using SVM-BA classifier / Y. Benmahamed, M. Teguar, A. Boubakeur // Energies. – 2017. – Vol. 10, No. 12. – Art. 2061. – DOI: 10.3390/en10122061.

28. . Benmahamed, Y. Application of SVM and KNN to Duval Pentagon 1 for transformer oil diagnosis / Y. Benmahamed, Y. Kemari, M. Teguar, A. Boubakeur // IEEE Transactions on Dielectrics and Electrical Insulation. – 2017. – Vol. 24, No. 6. – Pp. 3443-3451.

29. . Bhalla, D. Applications of artificial intelligence techniques in dissolved gas analysis: A review / D. Bhalla, R. K. Bansal, H. O. Gupta // International Journal of Electrical Power & Energy Systems. – 2010. – Vol. 32, No. 8. – Pp. 871-884.

30. . Breiman, L. Random forests / L. Breiman // Machine Learning. – 2001. – Vol. 45, No. 1. – Pp. 5-32. – DOI: 10.1023/A:1010933404324.

31. . Chen, T. XGBoost: A scalable tree boosting system / T. Chen, C. Guestrin // Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. – San Francisco, CA, USA, 2016. – Pp. 785-794. – DOI: 10.1145/2939672.2939785.

32. . CIGRE Technical Brochure 227. Life Management Techniques for Power Transformers. – Paris: CIGRE, 2003. – 129 p.

33. . CIGRE Technical Brochure 349. Moisture Equilibrium and Moisture Migration Within Transformer Insulation Systems. – Paris: CIGRE, 2008. – 62 p.

34. . Khalyasmaa, A. I. Decision Support System for Assessing the Technical Condition of Substation Electrical Equipment: Dis. ... Candidate of Technical Sciences: 05.14.02 / Khalyasmaa Alexandra Ilyinichna. – Yekaterinburg, 2015. – 188 p.

35. . Khalyasmaa, A. I. Assessment of the Technical Condition of Electrical Equipment in Distribution Networks: Textbook. / A. I. Khalyasmaa, S. A. Dmitriev. – Yekaterinburg: Ural University Press, 2015. – 160 p.

36. . Khrennikov, A. Yu. Diagnostics of Power Autotransformers: Application Experience and Development Prospects / A. Yu. Khrennikov // Electrical Stations. – 2018. – No. 2. – Pp. 27-34.

37. . Khrennikov, A. Yu. Power Autotransformer Diagnostics Systems Based on Expert and Neural Network Methods / A. Yu. Khrennikov, A. V. Mayorov, and N. M. Shakurin. Moscow: NTF Energoprogress, 2016. 208 p.

38. . Tsirel, Ya. A. Operation of Power Autotransformers at Power Plants / Ya. A. Tsirel. – M.: Energoatomizdat, 1990. – 256 p.

39. . Shakirzyanov, F. N. Application of Artificial Neural Networks for Diagnostics of Autotransformers / F. N. Shakirzyanov, A. R. Safin // Power Engineering of Tatarstan. – 2019. – No. 2 (54). – Pp. 23-29.

40. . Shutenko, O. V. Methods of diagnostics of power autotransformers based on the analysis of gases dissolved in oil: a comparative analysis / O. V. Shutenko // Technical Electrodynamics. – 2018. – No. 3. – Pp. 77-86.

41. . GOST 11677-85. Power autotransformers. General technical conditions. – Introduced on 1986-0701. – Moscow: Publishing House of Standards, 1985. – 30 p.