Inspection of Power Line Insulators
Rogério Sales Gonçalves, Guilherme Salomão Agostini, Reinaldo A. C. Bianchi, Rafael Zimmermann Homma, Daniel Edgardo Tió Sudbrack, Paulo Victor Trautmann, Bruno Clasen
Abstract
Insulators are power transmission line components responsible for two key tasks: the first one is to support the mechanical stress originated by the weight of the cables and devices, and the second one is to avoid electrical dissipation from the cables to the tower structure. Even though the shape and material of the insulator is made in such a way as to avoid the conduction of electrical current on its surface, if some types of dirty accumulates excessively, the insulator can still conduct an electric arc to the tower, causing damage to the power grid. This chapter first presents the state-of-the-art power line insulator cleaning methods and the techniques used to identify insulators that require cleaning. Then, this chapter describes an algorithm that makes use of machine learning, deep learning, and computer vision technics, which can be used embedded in an unmanned aerial vehicle, to support the energy company in the assessment of the levels of dirt on the insulators. Finally, experimental results are presented showing the challenges and the open problems.