Feature Extraction of Chalking in Silicone Rubber Insulators Using Digital Image Processing

In the present work, quantitative evaluation of chalking phenomenon in silicone rubber (SiR) insulators sub- jected to aging has been carried out. Chalking is observed in SiR composite insulators due to aging which get accelerated under the influence of UV exposure at high temperature and humidity. This leads to the changes in the surface properties such as discoloured surface, reduced hydrophobicity, increase in hardness, potentially resulting in surface cracking and erosion. The sheds of polymeric insulators exhibit various stages of chalking in service life, making it challenging for utilities to estimate the extent of this process and its impact on long-term performance. To address this issue, artificially chalking is created in High Temperature Vulcanized (HTV) SiR samples in the laboratory coinditions using solvents like methylcyclohexane (MCH), diethylamine (DEA), and catalysts such as tetramethylammonium hydroxide (TMAH). This study explores a less investigated quantitative evaluation of chalking using Digital Image Processing (DIP) techniques. The methodology involves preprocessing of images for adaptive thresholding and image segmentation techniques then applied to identify chalking regions. Basic morphological operations such as erosion and dilation further refine the segmentation process, facilitating the identification of chalked regions. The reported results are obtained using image processing as a non-contact technique for the periodic evaluation, operation, and maintenance of inservice composite insulators, offering insights into their chalking progression and aiding in optimizing their long-term performance and reliability.