Header menu link for other important links
A model based on bag of visual words to predict the category of damage in XLPE insulation under the application of combined AC and repeated lightning impulses of both polarities
M.S. Vidya, K. Sunitha, S. Ashok, , V. Chandra
Published in Springer Science and Business Media Deutschland GmbH
Cross-linked polyethylene (XLPE) is employed in high-voltage transmission cables, due to their distinguished insulation performance. Localized degradation due to AC voltages is a major cause of damage in these systems. These damages known as electrical trees vary in their shape and characteristics according to the cause of damage. This paper discusses the effect of lightning overvoltages in trees formed in these insulation systems. Initially experiments have been conducted to generate trees at 10 kV (tree-like tree) and 12 kV (bush branch tree) and the structure of the trees are visualized. This work is extended to study the effect of lightning overvoltages. It is crucial to detect the type of damage as accurately as possible. In this research work, a classification tree model, which is based on the Bag of Visual Words (BoVW), has been developed to predict the category of damage. The accuracy of the proposed model is evaluated using Receiver Operating Characteristics (ROCs) curves and confusion matrix. Finally, the validity of the developed model is verified by comparing it with the in-house collected experimental data and the other State-of-the-Art feature extraction techniques. © 2021, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.
About the journal
JournalData powered by TypesetElectrical Engineering
PublisherData powered by TypesetSpringer Science and Business Media Deutschland GmbH