ADAPTIVE MULTI-STAGE FUZZY LOGIC APPROACH WITH DYNAMIC WEIGHT ADJUSTMENT FOR ROBUST POWER TRANSFORMER DIAGNOSTICS

Authors

• Dilafruz Rustamovna Abdullabekova, Tashkent University of Information Technologies named after Muhammad al-Khwarizmi, abdullabekova_94@mail.ru, https://orcid.org/0009-0003-8887-6831Follow
• Odiljon Muhammadjonovich Kutbidinov, Tashkent State Transport University, Tashkent city, Republic of Uzbekistan

Abstract

An adaptive multi-level fuzzy logic framework with dynamic weight adjustment for power transformer fault diagnosis and health index assessment was proposed in this study. A comprehensive analysis of existing transformer diagnostic approaches was performed, and their limitations related to static weighting schemes and uncertainty handling were identified. A hierarchical fuzzy inference structure was introduced, integrating multi-source diagnostic data, including dissolved gas analysis, transformer oil quality indicators, thermal parameters, and electrical measurements. At the first level, individual fuzzy subsystems were developed to evaluate partial condition indices associated with insulation degradation, oil aging, and thermal–electrical stress. At the second level, a global Transformer Health Index was formulated using a dynamic weighting mechanism that accounted for data reliability, operating conditions, and aging effects. Reliability-aware weighting coefficients were defined and incorporated into the aggregation process, and their influence on diagnostic robustness was investigated. Simulation and experimental analyses were carried out using operational transformer datasets. Comparative evaluation was conducted against conventional threshold-based methods and static fuzzy logic models. The obtained results demonstrated a significant improvement in fault classification accuracy and health index stability, with an increase of up to 35% compared to baseline approaches. The effectiveness of the proposed framework under conditions of measurement uncertainty and incomplete data was shown. The proposed adaptive fuzzy logic framework was demonstrated to be suitable for condition-based maintenance, predictive diagnostics, and lifecycle management of power transformers. The results can be applied in SCADA systems, digital substations, and intelligent asset management platforms in modern power systems.

First Page

41

Last Page

46

References

1. Ghoneim, S. S. M., Taha, I. B. M., Elkalashy, N.I.Integrated fuzzy logic and dissolved gas analysis for power transformer fault diagnosis. IET Generation, Transmission & Distribution, 10(1), 2016, pp. 77–85.
2. Abdelmalik, A. A., Youssef, T., Shaban, K. B. An adaptive multi-fuzzy logic model for diagnosing transformer faults. Electric Power Systems Research, 181, 2020, 106189.
3. Jalbert, J., Duchesne, S., Rodriguez-Celis, E., et al. Health index for power transformers using fuzzy logic. IEEE Electrical Insulation Magazine, 30(6), 2014, pp. 28–35.
4. Shang, H., Xu, J., Li, Y. Health index-based condition assessment of power transformers considering multi-source information. IEEE Transactions on Power Delivery, 35(1), 2020, pp. 334–343.
5. Islam, M. M., Lee, G., Hettiwatte, S., Williams, D. A review of condition monitoring techniques and diagnostic tests for power transformers. Electric Power Systems Research, 128, 2015, pp. 35–48.
6. García, B., Burgos, J. C., Alonso, Á. M. Transformer oil degradation assessment using fuzzy logic. IEEE Transactions on Dielectrics and Electrical Insulation, 15(2), 2008, pp. 496–504.
7. Bakar, N. A., Abu-Siada, A. Fuzzy logic approach for transformer condition assessment. IET Generation, Transmission & Distribution, 10(12), 2016, pp. 2961–2968.
8. da Silva, L. E. B., Veloso, G. F. C., Honório, L. M., et al. A multi-criteria decision method for transformer health index estimation. Measurement, 120, 2018, pp. 146–155.
9. Prasojo, R. A., Suwarno. Power transformer condition assessment using fuzzy logic based on oil test and DGA. Energies, 11(9), 2018, 2398.
10.  Khan, S. A., Islam, S. M. Predictive diagnostics of power transformers using fuzzy logic and DGA. IEEE Transactions on Dielectrics and Electrical Insulation, 25(3), 2018, pp. 974–982.
11.  Macedo, J. L., da Silva, J. G., Duque, C. A. An adaptive fuzzy inference system for power transformer diagnosis. Electric Power Systems Research, 189, 2020, 106703.
12.  Liu, Y., Zhang, Y., Li, J. Condition assessment of power transformers based on multi-attributes using fuzzy logic. International Journal of Electrical Power & Energy Systems, 118, 2020, 105787.
13.  Morsalin, S., Rahman, M. A. Health index-based condition monitoring of transformers under uncertainty. IEEE Access, 8, 2020, pp. 161724–161735.
14.  Zhang, X., Gockenbach, E. Asset management of transformers based on condition monitoring and diagnostics. IEEE Electrical Insulation Magazine, 24(2), 2008, pp. 26–40.
15.  Ekanayake, C., Saha, T. K. Fuzzy logic-based insulation assessment of power transformers. IEEE Transactions on Power Delivery, 21(4), 2006, pp. 2151–2159.
16.  Bakar, N. A., Abu-Siada, A., Islam, S. A review of dissolved gas analysis measurement and interpretation techniques. IEEE Electrical Insulation Magazine, 30(3), 2014, pp. 39–49.
17.  Tang, W., Wu, Q., Richardson, Z. Equivalent aging models of transformer insulation for health index estimation. IEEE Transactions on Power Delivery, 28(3), 2013, pp. 1538–1546.

Recommended Citation

Abdullabekova, Dilafruz Rustamovna and Kutbidinov, Odiljon Muhammadjonovich (2026) "ADAPTIVE MULTI-STAGE FUZZY LOGIC APPROACH WITH DYNAMIC WEIGHT ADJUSTMENT FOR ROBUST POWER TRANSFORMER DIAGNOSTICS," Technical science and innovation: Vol. 2026: Iss. 1, Article 11.
DOI: https://doi.org/10.59048/2181-1180.1795
Available at: https://btstu.researchcommons.org/journal/vol2026/iss1/11