Clustering analysis for partial discharge detection in oil reactors

Velasquez R.A.; Munoz-Ccuro F.E.; Mamani M.J.A. · Proceedings of the 2021 IEEE Engineering International Research Conference, EIRCON 2021 (2021) · DOI: 10.1109/eircon52903.2021.9613454

Resumen

Reactors are the main equipment in the transmission systems, for voltage control in long overhead lines connected to the high voltage grid. The main problem is the one hundred percent of load or higher according the voltage level; usually in 500 kV grid, at the north of Peru, it is a nominal condition, therefore, the overload is a common practice. In this research article, it evaluates partial discharge with DGA analysis based on Support Vector Machine and clustering analysis associated to high voltage reactor. Suddenly increase of hydrogen; and its remaining life. Our findings in this paper is a new classification techniques for the reactors and a clustering analysis associated to partial discharge, it improves the accuracy compared with the power transformers. This paper is relevant due to knowledge engineering with the incorporation the new methodology for the partial discharge detection with high accuracy with 99.91% compared with transformer results of 97.55%.

Instituciones: Universidad Tecnológica del Perú; Universidad César Vallejo