Software tool to detect hot spots on electrical insulators

This article presents a software tool to automatically detect hot spots on electrical insulators of 220 KV transmission lines subjected to high-polluted environments in the city of Barranquilla, Colombia. The aim of this research was to register the visual effects of pollution on three types of insulators, and analyze a set of images to record the number of hot spots and flashover arc during 14-day periods. Identification of hot spots on insulators was conducted by segmenting the images with machine vision techniques, which helped to determine the states of insulators for maintenance activities. Finally, the software tool performance was tested for different insulators subjected to polluted conditions in the Colombian Caribbean Region. As results, we obtained a visual tool to evaluate the effects of pollution on electric insulators, the visual verification of the behavior of electric insulators in polluted-environments after the rain and manual washing, and the threshold to obtain video segmentation of insulators and other elements of the power grid for automatic recognition.