Pre-breakdown and breakdown Currents and correlated light emissions analysis in liquid dielectrics
Résumé
Condition monitoring and fault diagnostics plays a very important role in the production security and the product quality. The demand for monitoring and fault diagnosis of power systems has increased the efforts to develop new analysis techniques. A wide variety of techniques were used for this purpose. In this work, a systematic survey on positive and negative streamers currents and corresponding emitted lights has been led in three liquid dielectrics: Mineral oil, tetraester and toluene, and the results are presented and discussed with regard to their energy and frequency parameters.This study will mainly investigate real time techniques for signal identification of electrical discharges of different energy levels mainly in transformer mineral oil in order to identify the steamers propagation that can transit to high level energy arcing discharges. The energy and inter-correlation between electrical and light signals are also investigated. The use of artificial neural networks (ANN) for real time discharge signals analysis (currents or/and emitted lights) has been proved to be a fast and reliable solution for the protection of sensitive power equipments. With a database of 40 samples of each type of signal used for training the ANN, acceptable results with a rate of 75% of success have been obtained.