The ageing level of ZnO materials in gapless surge arresters can be determined by using either the traditional leakage current measurements or recently introduced thermal images of the arrester. However, a direct correlation between arrester thermal images and its leakage current (and hence the ageing level) is yet to be established. This paper attempts to find such a correlation using an artificial neural network (ANN). Experimental work was carried out to capture both the thermal images and leakage current of 120kV rated polymeric housed gapless arresters. Critical parameters were then extracted from both the thermal image and the leakage current, before being exported into the artificial neural network tool. Using the leakage current level, the conditions of the arrester are classified as normal, caution, and faulty. The ANN correctly classifies the ageing level using only the thermal image information with an accuracy of 97%, which is highly encouraging.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.