Estimation of Power Transformer Insulation Ageing Using Optical Coherence Tomography

Abstract

Oil-impregnated cellulose material, such as Kraft paper, is an essential part of the insulation system of a power transformer. The mechanical strength of Kraft paper directly impacts the safety of operating a power transformer. Obtaining structural subsurface images of the transformer paper would offer significantly more information than from current optical methods. This subsurface imaging could allow accurate estimation of the operational age of the transformer insulation, beyond only classifying it as belonging to a broad age group. In this work, optical coherence tomography (OCT), a subsurface imaging technique, is used to acquire subsurface structural images of the thermally-aged Kraft paper samples. Statistical texture analysis is applied to these OCT images to convert each image into a spatial grayscale level dependence matrix (SGLDM), from which three Haralick texture features are calculated. These three features are used to estimate the ageing time of thermally-aged Kraft paper.