Résumé

Dissolved gas analysis (DGA) is still one of the most popular techniques for diagnosing transformer faults by adopting it as a primary technique for early detection of faults and increasing the reliability of the electrical system. This paper proposes an efficient fusion method based on DGA data using the two best algorithms accuracy, the neural network (MLP), the naïve bayes (NB) through ppm input vectors and percentages. The fusion method has predictively combined the two classifiers and obtained a higher accuracy than both of them, which reached 95.83%. This indicates that the proposed method is effective and promising for diagnosing faults in power transformers.

Information

Item Type:	Conference
Divisions:	» Laboratoire de Recherche en Électromécanique et Sûreté de Fonctionnement » Faculté des Sciences et de la Technologie
ePrint ID:	3746
Date Deposited:	2022-12-22
Further Information:	Google Scholar
URI:	https://www.univ-soukahras.dz/fr/publication/article/3746

BibTex