Determine the aging status of silicone rubber insulators base on smartphone hyperspectral cameras

Abstract

In the power industry, silicone rubber composite insulators are widely used. This article introduces a set of methods for judging the aging degree of silicone rubber composite insulators on site, including our handheld compact and low-cost hyperspectral camera and classification algorithm. The hyperspectral camera has the advantages of compactness, stable working performance, and Broad spectral detection range (400nm-1000nm). After connecting the hyperspectral camera to the smart mobile terminal,We also through the APP we developed, which completes the visualization of spectral data, can display the real-time spectral of the silicone rubber composite insulator sample on the screen of the smart terminal. In addition, we build a composite insulator aging prediction model based on the three algorithms of SVM(support vector machines), KNN(K-NearestNeighbor) and decision tree. The results show that the SVM model has the best effect. And the prediction accuracy of this model is 96.09%. it means that the model has high feasibility for predicting the aging degree of composite insulators.