Analysis of Discharge Image and Particle Characteristics of Needle Plate in Transformer Oil Under AC Voltage

Abstract

For the development of converter transformer applications, research on the discharge characteristics in transformer oil is required. This article presents an analysis of the streamer discharge characteristics caused by a needle-plate electrode in oil-pressboard insulation under ac voltage and discharge images on the time scale of microseconds. Based on the discharge image, a Gaussian filter, image segmentation of the Otsu algorithm, and Canny edge detection are used to detect the edge of the discharge image. By extracting the discharge profile image, the discharge characteristics are effectively identified. The size and quantity of particles in transformer oil during the discharge are analyzed. Research on the characteristics of streamer discharge propagation is not only helpful for understanding the process of discharge but also has important significance for converter transformer design from the perspective of inhibiting streamer discharge.