Al-Shaikh, Omar Ahmed Mohammed Hamood and Ahmad, Arfah and Haja Mohideen, Ahmad Jazlan and Yahaya, Muhammad Shahril (2026) Transformer health index prediction using static and temporal learning models. International Journal of Advanced Computer Science and Applications, 17 (8). pp. 135-149. ISSN 2158-107X E-ISSN 2156-5570
|
PDF
- Published Version
Download (986kB) | Preview |
Abstract
Abstract—Transformer health index (THI) prediction sup-ports condition-based maintenance by mapping dissolved-gas, oil-quality, and furan indicators to an interpretable asset-condition score. This study develops a supervised benchmarking framework using 3,392 transformer oil diagnostic records from 510 trans-formers. Thirteen diagnostic features from dissolved gas analysis, oil quality analysis, and furan analysis are evaluated across three modelling groups: static machine learning, static deep learning, and variable-length temporal deep learning. Transformer-level splitting is applied to reduce data leakage from repeated trans-former histories. Static gradient-boosted tree models achieved the strongest results. XGBoost obtained the lowest test root mean square error of 3.4254 with a coefficient of determination of 0.9780 and health-index class accuracy of 88.20 percent. The best tem-poral model, the liquid time-constant neural network (LTC-LNN), achieved a test root mean square error of 4.8691 and a coefficient of determination of 0.9567. Permutation feature importance iden-tified 2FAL, C2H2, and dielectric breakdown as the most influen-tial predictors. For the present dataset, static gradient-boosted tree models produced the strongest observed point-estimate per-formance, while LTC-LNN remained a promising temporal-learn-ing alternative for repeated diagnostic histories. However, the un-certainty analysis indicates that small numerical differences among the leading static models should not be interpreted as de-finitive evidence of model superiority.
| Item Type: | Article (Journal) |
|---|---|
| Uncontrolled Keywords: | Transformer health index; machine learning; deep learning; liquid neural networks; oil diagnostic data; conditionbased maintenance; smart grid asset management |
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7800 Electronics. Computer engineering. Computer hardware. Photoelectronic devices |
| Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button): | Kulliyyah of Engineering > Department of Mechatronics Engineering Kulliyyah of Engineering |
| Depositing User: | Dr Ahmad Jazlan Haja Mohideen |
| Date Deposited: | 22 Sep 2026 16:38 |
| Last Update: | 22 Sep 2026 16:38 |
| Queue Number: | 2026-09-Q5196 |
| URI: | http://irep.iium.edu.my/id/eprint/131408 |
| Indexed In: | ERA |
Actions (login required)
 |
View Item |