Depot Institutionnel de l'UMBB >
Publications Scientifiques >
Communications Internationales >
Veuillez utiliser cette adresse pour citer ce document :
http://dlibrary.univ-boumerdes.dz:8080/handle/123456789/11332
|
Titre: | A new transformer condition monitoring based on infrared thermography imaging and machine learning |
Auteur(s): | Mahami, Amine Bettahar, Toufik Rahmoune, Chemseddine Amrane, Foudil Touati, Mohamed Benazzouz, Djamel |
Mots-clés: | Electrical transformer Faults classification stability Faults diagnosis Feature extraction Machine learning methods Infrared thermography images |
Date de publication: | 2023 |
Editeur: | Springer |
Collection/Numéro: | Lecture Notes in Networks and Systems/ Vol.591 LNNS (2023);pp. 408-418 |
Résumé: | Electrical systems maintenance is becoming a crucial and an important part in the economic policies and that’s due their deep implication in the majority of the industrial installations. Electrical transmission and distribution relay mainly on transformers. Electrical transformers condition monitoring plays a major role in increasing their availability, enhancing their reliability and preventing further major failures and high cost maintenance. A new non-contact and non-intrusive method is adopted in this paper to monitor electrical transformers and diagnose their faults based on infrared thermography imaging techniques (IRT). When thermographs are obtained using an infrared camera for different states of the studied transformer, a dataset is then prepared for the following step. Features extraction was applied on the considered infrared images to be used later as input indicators for an automatic classification and identification of transformer’s healthy and several faulty states based machine learning methods (LS-SVM). This method was applied and compared with several IA techniques in order to select the most efficient one in term of accuracy and stability to be relied on in this purpose. The proposed technique, which is mainly based on IRT, features extraction and machine learning, has shown a remarkable efficiency in transformers condition monitoring and an accurate faults diagnosis, and can be generalized as a reliable and powerful tool in such problematics |
URI/URL: | DOI 10.1007/978-3-031-21216-1_43 https://link.springer.com/chapter/10.1007/978-3-031-21216-1_43 http://dlibrary.univ-boumerdes.dz:8080/handle/123456789/11332 |
ISBN: | 978-303121215-4 |
ISSN: | 23673370 |
Collection(s) : | Communications Internationales
|
Fichier(s) constituant ce document :
Il n'y a pas de fichiers associés à ce document.
|
Tous les documents dans DSpace sont protégés par copyright, avec tous droits réservés.
|