Intelligent Diagnosis of Incipient Fault in Power Distribution Lines based on Corona Detection in UV-Visible Videos

As one of the non-contact methods for incipient fault diagnosis of power distribution lines, UV-Visible imaging has become more popular due to its good performance and robustness against environmental parameters. This paper presents a method based on UV-Visible video processing, deep learning, and experimental information. First, some videos are acquired from power distribution lines using CoroCam 6D2 considering observation distance, the UV gain of the camera, air pressure, and humidity as effective parameters on the discharge area in images. Frames are extracted from acquired video with the rate according to the nominal voltage of the line. Power equipment is detected in each frame using Faster R-CNN, and then, it is tracked through the whole video frames to compensate the camera movement. Then, color thresholding is used to identify corona discharges in the image for each device separately. Finally, based on the ratio of the detected discharge area to the area of the equipment, the incipient fault severity level is determined. The proposed method not only performs better than state-of-the-art but also it is a practical method without the need for user skill, and it can automatically identify defects in distribution lines, even in videos containing several possible defective devices.