Abdulla Lutfi

The utilization of ceramic insulators in overhead lines has been a prevalent practice for more than a century. A significant portion of these insulators have either approached or exceeded their intended lifetime, posing a risk of sudden failure. Therefore, there is a growing need for reliable, fast, and cost-effective condition monitoring systems. Defective insulators can experience partial discharge (PD) and dry band arcing (DBA), which release detectable radiation in the form of heat, electromagnetic waves and ultrasonic emissions. This research introduces a non-contact method for assessing ceramic insulators using vision- and radiation-based sensors. It employs deep learning algorithms to classify and quantify defects from images and acoustic emissions.