Equalização e segmentação de imagens termográficas aplicadas à metodologia de inspeção de isoladores poliméricos.

Abstract

The analysis of the temperature variation along the insulator through thermography can be one of the most effective methods of evaluating insulators in service, for it considers the combined effect of pollution, electrical stress, internal defects and environmental conditions. However, analyses based only on visual inspection of images are not enough. In fact, only specialized engineers are able to make correct judgments. In this work a computational routine was developed, based on the digital image processing. The routine utilizes the signal nonlinear expansion technique, histogram equalization and principal component analysis. The digital processing of the thermographic images is used to obtain the temperature values that indicate the thermal behavior of the insulators under inspection. Laboratory tests were performed with 230 kV polymeric insulators with different levels of degradation in order to capture images using an infrared radiation detection camera. The results demonstrate the efficiency of the digital thermographic imaging technique to obtain the thermal profile of the insulators inspected. Based on the products of the digital processing of the thermographic images, we have a better evaluation of the degradation levels of the insulators, besides efficiently support to the classification of the operational states of the insulators.