Accurate Identification of Transformer Faults From Dissolved Gas Data Using Recursive Feature Elimination Method
Bibliographic record
Abstract
Dissolved gas analysis (DGA) of insulating oils is one of the most popular methods to detect incipient faults in power transformers. However, appropriate feature selection is crucial for accurately detecting incipient faults using DGA data. Another issue is the unavailability of a balanced DGA dataset, which can hamper the fault classification accuracy. Considering these two issues, this article proposes a novel and accurate fault classification framework using gas ratios as features obtained from the DGA data of power transformers. The obtained unbalanced DGA data was initially balanced using the synthetic minority oversampling technique (SMOTE) in the data pre-processing stage. Following this, an efficient feature selection algorithm, namely, recursive feature elimination (RFE) was used to select the best possible features prior to the fault classification using three benchmark machine learning (ML) classifiers, namely,${k}$-nearest neighbor (KNN), multiclass support vector machines (SVMs), and extreme gradient boost (XGBoost). The proposed classification model was tested on the DGA data obtained from the local power utility and on the benchmark IEC TC-10 database. Investigations revealed that the proposed classification model delivered detection accuracy of 98.84% and 97.43%, respectively. The proposed method may be reliably used to diagnose incipient faults in power transformers.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".