Smart Trafo: A Random Forest and LLM-Based Decision Support System for Power Transformer Fault Diagnosis via Dissolved Gas Analysis
DOI:
https://doi.org/10.69533/informatech.volume3number1.511Keywords:
Decision Support System, Dissolved Gas Analysis, Large Language Model, Power Transformer, Random ForestAbstract
Manual management of Dissolved Gas Analysis (DGA) data for power transformers at PT. PLN (Persero) UPT Manado has been identified as a critical operational bottleneck, with existing spreadsheet-based workflows susceptible to human error, poor historical traceability, and limited scalability. Prior studies on DGA-based transformer diagnosis have been predominantly confined to standalone classification models without integration into operational management systems, leaving a significant gap in practical field deployment. This research contributes a novel integrated Decision Support System named Smart Trafo, which is the first to combine a Random Forest classification model, Duval Pentagon visualization, historical trending analysis, and an LLM-based conversational assistant (Volty AI) within a unified full-stack web platform. The Random Forest model was trained on 375 DGA samples across six fault classes using five gas parameters conforming to IEEE C57.104, achieving an overall accuracy of 84% and a macro-average F1-score of 0.83. Feature importance analysis revealed Hydrogen (H₂) as the dominant diagnostic indicator at 26.2%. The system successfully automates DGA fault classification, eliminates manual calculation errors, and provides real-time technical recommendations, thereby enabling more efficient and data-driven preventive maintenance decisions at PT. PLN (Persero) UPT Manado.
Downloads
References
A. B. Vernandez, A. Jamaah, D. S. Pasisarha, A. Subagyo, and P. Rahardjo, “Pengaruh Electricity Saving Box Terhadap Penggunaan Energi Dan Tagihan Listrik Bulanan Pada Kelas Pelanggan Tegangan Rendah,” Orbith: Majalah Ilmiah Pengembangan Rekayasa dan Sosial, vol. 20, no. 2, pp. 143–151, Jul. 2024, doi: 10.32497/ORBITH.V20I2.5776.
M. Huwae, M. A. F. Haurissa, H. L. Latupeirissa, T. Elektro, and N. Ambon, “Analisis Pengaruh Pembebanan Terhadap Efisiensi Transformator Daya 30 MVA PLTMG Ambon Peaker 30 MW,” Jurnal FORTECH, vol. 6, no. 1, pp. 1–8, Jan. 2025, doi: 10.56795/FORTECH.V6I1.6101.
A. Abduvokhid et al., “A review on power transformer failures: analysis of failure types and causative factors,” Indonesian Journal of Electrical Engineering and Computer Science, vol. 38, no. 2, pp. 713–722, May 2025, doi: 10.11591/IJEECS.V38.I2.PP713-722.
P. Prayitno, U. S. Serang, and B. Corresponding, “Diagnosis Minyak Isolasi pada Trafo dengan Metode Dissolved Gas Analysis (DGA),” Asian Journal of Mechatronics and Electrical Engineering, vol. 1, no. 1, pp. 47–52, Sep. 2022, doi: 10.55927/AJMEE.V1I1.1124.
K. N. V. P. S. Rajesh, U. Mohan Rao, I. Fofana, P. Rozga, and A. Paramane, “Influence of Data Balancing on Transformer DGA Fault Classification With Machine Learning Algorithms,” IEEE Transactions on Dielectrics and Electrical Insulation, vol. 30, no. 1, pp. 385–392, Feb. 2023, doi: 10.1109/TDEI.2022.3230377.
S. A. M. Abdelwahab, I. B. M. Taha, R. Fahim, and S. S. M. Ghoneim, “Transformer fault diagnose intelligent system based on DGA methods,” Scientific Reports 2025 15:1, vol. 15, no. 1, pp. 8263-, Mar. 2025, doi: 10.1038/s41598-024-78293-7.
R. Furqaranda, A. Bintoro, A. Asri, W. K. A. Al-Ani, and A. Shrestha, “Analysis Oil Condition of Transformer PT-8801-A by Using the Method TDCG, Rogers Ratio, Key Gas, and Duval Triangle: A Case Study at PT. Perta Arun Gas,” Journal of Renewable Energy, Electrical, and Computer Engineering, vol. 2, no. 2, pp. 47–54, Sep. 2022, doi: 10.29103/JREECE.V2I2.8567.
N. Al and A. Munir, “Analisis Interpretasi Hasil Uji Dissolved Gas Analysis Minyak Transformator Dengan Metode Kombinasi Duval Pentagon,” Mutiara: Multidiciplinary Scientifict Journal, vol. 2, no. 3, pp. 1056–1064, Mar. 2024, doi: 10.57185/MUTIARA.V2I3.165.
H. Yu, X. Wang, Q. Chen, D. Yang, X. Gao, and H. Du, “Evaluation and optimization of the Duval Pentagon method for diagnosing various types of transformer faults,” Electrical Engineering, vol. 107, no. 9, pp. 11431–11439, Sep. 2025, doi: 10.1007/s00202-025-03099-3.
H. Risatayn, E. Ekojono, and D. S. Hormansyah, “Metode Random Forest Untuk Klasifikasi Kerusakan Transformator Daya Berdasarkan Gas Terlarut Pada Duval Triangle Dan Duval Pentagon,” JATI (Jurnal Mahasiswa Teknik Informatika), vol. 7, no. 6, pp. 3464–3471, Feb. 2023, doi: 10.36040/JATI.V7I6.7743.
R. M. Hidayat, G. Media, N. Sartika, and L. Kamelia, “Diagnosis Kegagalan Transformator Daya Menggunakan Algoritma K-Nearest Neighbors Dan Artificial Neural Network Berbasis Dissolved Gas Analysis,” Transmisi: Jurnal Ilmiah Teknik Elektro, vol. 27, no. 3, pp. 140–148, Jul. 2025, doi: 10.14710/Transmisi.27.3.140-148.
Suwarno, H. Sutikno, R. A. Prasojo, and A. Abu-Siada, “Machine learning based multi-method interpretation to enhance dissolved gas analysis for power transformer fault diagnosis,” Heliyon, vol. 10, no. 4, p. e25975, Feb. 2024, doi: 10.1016/J.Heliyon.2024.E25975.
V. M. N. Dladla and B. A. Thango, “Fault Classification in Power Transformers via Dissolved Gas Analysis and Machine Learning Algorithms: A Systematic Literature Review,” Applied Sciences 2025, Vol. 15, Page 2395, vol. 15, no. 5, p. 2395, Feb. 2025, doi: 10.3390/APP15052395.
M. Fariz, S. Lazuardy, and D. Anggraini, “Modern Front End Web Architectures with React.Js and Next.Js,” International Research Journal of Advanced Engineering and Science, vol. 7, no. 1, pp. 132–141, 2022.
U. U. Sufandi, “Analisis Kebutuhan dan Dokumentasi Sistem Informasi Tiras dan Transaksi Bahan Ajar Universitas Terbuka,” Jurnal Nasional Pendidikan Teknik Informatika (JANAPATI), vol. 11, no. 2, pp. 112–122, Aug. 2022, doi: 10.23887/janapati.v11i2.42966.
R. J. G, “A Comparative Study On Full Stack Web Frameworks For Modern Web Application Development,” International Research Journal of Engineering and Technology, 2025, [Online]. Available: www.irjet.net
F. P. E. Putra, H. Hasbullah, F. Muslim, and R. Paradina, “Technical Performance Comparison of Modern Frontend Frameworks Study on Svelte, React, and Vue,” Brilliance: Research of Artificial Intelligence, vol. 5, no. 1, pp. 355–364, Jul. 2025, doi: 10.47709/brilliance.v5i1.6133.
R. Yusuf Azhari, “Web Service Framework : Flask Dan Fastapi,” Technology and Informatics Insight Journal, vol. 1, no. 1, pp. 58–65, Feb. 2022, doi: 10.32639/TIIJ.V1I1.54.
Imarticus Learning, “Flask vs FastAPI: Which is Better for Deploying ML Models? - Finance, Tech & Analytics Career Resources | Imarticus Blog.” Accessed: Jan. 19, 2026. [Online]. Available: https://imarticus.org/blog/flask-vs-fastapi-which-is-better-for-deploying-ml-models/
B. El Bakkouri, S. Raki, and T. Belgnaoui, “The Role of Chatbots in Enhancing Customer Experience: Literature Review,” Procedia Comput. Sci., vol. 203, pp. 432–437, 2022, doi: 10.1016/J.PROCS.2022.07.057.
L. Aldhafeeri et al., “Generative AI Chatbots Across Domains: A Systematic Review,” Applied Sciences 2025, Vol. 15, Page 11220, vol. 15, no. 20, p. 11220, Oct. 2025, doi: 10.3390/APP152011220.
S. R. Al-Sakini et al., “Dissolved Gas Analysis for Fault Prediction in Power Transformers Using Machine Learning Techniques,” Applied Sciences 2025, Vol. 15, vol. 15, no. 1, Dec. 2024, doi: 10.3390/APP15010118.
W. C. Kurniawan and B. Sudiarto, “Optimizing Power Transformer Failure Identification: A Multi-Method Framework Based on Normalized Energy Intensity According to IEEE C57.104-2019 Standards Adapted to Indonesian Power Transformer Characteristics,” International Journal of Electrical, Computer, and Biomedical Engineering, vol. 3, no. 2, pp. 254-282–254–282, Jun. 2025, doi: 10.62146/IJECBE.V3I2.121.
Rizal Furqan Ramadhan, “Implementation Weighting Method for Selection Online Buying and Selling Platforms in the Digital Era,” Int. J. Innov. Enterp. Syst., vol. 8, no. 2, pp. 45–53, Dec. 2024.
R. F. Ramadhan, K. Eliyen, and F. R. Aulia, “SSOBS: Online buying and selling platform selection system based on store characteristics,” in 2023 International Conference on Electrical and Information Technology (IEIT), 2023, pp. 348–353.
S. Mohamed, “AI and Blockchain in Cybersecurity: A Sustainable Approach to Protecting Digital Assets,” J. Ilm. Inform. dan Komput., vol. 2, no. 1, pp. 1–8, 2025.
M. S. Rumetna, T. N. Lina, J. Karay, A. B. Santoso, and W. Ferdinandus, “Application of the Hungarian Algorithm for Workforce Task Optimization in Mobile Device Repair Operations,” J. Ilm. Inform. dan Komput., vol. 2, no. 2, pp. 94–101, 2025.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Jeremia David Anthony Paduli, Jonathan Gabrillio Kaligis, Ade Yusupa, Yaulie Rindengan (Author)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
