| Titre : |
Comparative analysis of machine learning methods for power transformer oil diagnosis using dissolved gas analysis |
| Type de document : |
document électronique |
| Auteurs : |
Imed-Eddine Boukhari, Auteur ; Youcef Benmahamed, Directeur de thèse |
| Editeur : |
[S.l.] : [s.n.] |
| Année de publication : |
2024 |
| Importance : |
1 fichier PDF (6 Mo) |
| Présentation : |
ill. |
| Note générale : |
Mode d'accès : accès au texte intégral par intranet.
Mémoire de Projet de Fin d’Études : Électrotechnique : Alger, École Nationale Polytechnique : 2024
Bibliogr. p. 101 - 107 |
| Langues : |
Français (fre) |
| Mots-clés : |
Power transformer
Insulating oil
Diagnosis
Dissolved gas analysis
Electrical and thermal faults
Machine learning |
| Index. décimale : |
PA01424 |
| Résumé : |
This work focuses on diagnosing the condition of power transformer oil through dissolved gas analysis consisting of H2, CH4, C2H2, C2H4, and C2H6. For this purpose, many machine learning algorithms have been developed. eight input vectors have been considered, and several pre-processing techniques were used. The database used contains 666 samples, of which 506 are selected for training and 160 for testing. Inspired by international standards such as IEC and IEEE, six electrical and thermal faults have been considered, namely PD, D1, D2, T1, T2, and T3. The best diagnostic rate of 99.375% was achieved using a custom-built decision tree. |
Comparative analysis of machine learning methods for power transformer oil diagnosis using dissolved gas analysis [document électronique] / Imed-Eddine Boukhari, Auteur ; Youcef Benmahamed, Directeur de thèse . - [S.l.] : [s.n.], 2024 . - 1 fichier PDF (6 Mo) : ill. Mode d'accès : accès au texte intégral par intranet.
Mémoire de Projet de Fin d’Études : Électrotechnique : Alger, École Nationale Polytechnique : 2024
Bibliogr. p. 101 - 107 Langues : Français ( fre)
| Mots-clés : |
Power transformer
Insulating oil
Diagnosis
Dissolved gas analysis
Electrical and thermal faults
Machine learning |
| Index. décimale : |
PA01424 |
| Résumé : |
This work focuses on diagnosing the condition of power transformer oil through dissolved gas analysis consisting of H2, CH4, C2H2, C2H4, and C2H6. For this purpose, many machine learning algorithms have been developed. eight input vectors have been considered, and several pre-processing techniques were used. The database used contains 666 samples, of which 506 are selected for training and 160 for testing. Inspired by international standards such as IEC and IEEE, six electrical and thermal faults have been considered, namely PD, D1, D2, T1, T2, and T3. The best diagnostic rate of 99.375% was achieved using a custom-built decision tree. |
|
PA00100 
Ajouter le résultat dans votre panier
Visionner les documents numériques
Faire une suggestion Affiner la rechercheComparative analysis of machine learning methods for power transformer oil diagnosis using dissolved gas analysis (2024)
