Predicting flashover events is critical for safety in various electrical domains, particularly in designing medium and high-voltage systems. This paper presents an innovative approach to automating the prediction of flashover occurrences in medium-voltage connectors using machine learning techniques. The study involves the analysis of model-generated plots based on experimental flashover events, specifically focusing on the characterization of electrical parameters indicative of flashover. Leveraging machine learning models, such as convolutional neural networks and regression techniques, we demonstrate the ability to predict flashover and non-flashover at certain applied voltage level instances with high accuracy. The proposed methodology aims to streamline the identification process, eliminating the need for manual inspection and enhancing the efficiency of flashover prediction in diverse medium-voltage applications.
Predictive Modeling of Surface Flashover Using Deep Learning
2024-06-19
3057015 byte
Conference paper
Electronic Resource
English
Aerial flashover caused by surface flashover generated from microscopic pores of insulating solid
Online Contents | 2012
|Inhibiting surface flashover with magnetic fields
Tema Archive | 1989
|Surface Flashover Triggering of Pseudospark Switchers
British Library Online Contents | 2003
|Measurement of the effects of surface roughness on flashover
Tema Archive | 1988
|