Microwave reflectometry for imaging diagnosis of internal defects in composite-insulator trunks

In virtue of the exceptional performance regarding the electric insulation, composite insulators are intensively utilized in the power transmission and distribution system. However, the intricate production process and hostile environments may leave the trunk of the composite insulator vulnerable to the internal defects such as the air gap, the carbonization defect and the groove. These defects undermine the integrity of the composite insulator, and would lead to the failure of the electrical system. In light of this, Microwave Non-destructive Testing (MNT) is investigated in this paper for imaging and evaluation of internal defects in composite-insulator trunks. The characteristics of the testing signal are scrutinized via 3D finite-element simulations. The feasibility of microwave reflectometry for detection of the internal-defects is identified. In parallel, an experimental MNT system is established for imaging and assessment of internal defects in composite-insulator trunks. In an effort to improve the quality of the defect image, an adaptive interference suppression method based on singular value decomposition is proposed. An image segmentation technique based on the pixel statistics is subsequently exploited to identify the defect region. It is indicated from the results that the localization and quantitative evaluation regarding the area of internal defects buried within the composite-insulator trunk can be realized via microwave reflectometry alongside the proposed methods.