Modélisation et diagnostic du comportement dynamique des transformateurs de puissance par les méthodes complémentaires : AGD, RNA et Essais FRA.

Abstract

The monitoring and diagnosis of electrical equipment, in particular power transformers which constitute the heart of the production, transmission and distribution of electricity have long been the subject of great interest. Any failure or unpredictable shutdown of this element can cause damage to the electrical networks and lead to significant financial losses. Major faults inside the transformer can be related to reduced dielectric strength, mechanical integrity and thermal faults. The detection of incipient faults in transformers can thus provide valuable information that will help to take the necessary measures to avoid these failures. Dissolved Gas Analysis (DGA) and Frequency Response Analysis (FRA) and artificial neural networks (ANN) are methods that provide parts of important information regarding transformer life condition, and each information (which may be corroborating, complementary and even contradictory) taken together constitute the evidence which, through the analysis of an expert, allow to make a reliable and precise diagnosis and which does not require any devatting of the transformer because of its high sensitivity

Description

Citation

Endorsement

Review

Supplemented By

Referenced By