Energy transmission lines are of great importance in the economical and efficient transportation of electrical energy from the places where it is produced to the consumption areas. Overhead transmission line insulators are one of the most important elements of electrical energy transmission and distribution systems. Insulators are network materials that fix the conductors in power transmission lines, switchyards and distribution centers to poles, carry them and isolate them from each other and from the ground. The basic principles of insulators are insulation. For this reason, insulators must show great resistance to electric current and must not lose their insulating properties in overvoltages, must withstand high temperatures, perform their functions without deterioration in different environmental conditions, and must be in a structure and on the surface that will not allow electrical jumps. Insulators may vary in their durability rates according to the material they are made of. In case of any insulator failure, the insulator cannot fulfill its duties, especially its own insulation function, and this may cause serious problems in power transmission lines. In case of breakage or failure of the insulator, it should be noticed immediately and replaced with a new one soon. Since the insulators are on the pole, sometimes their faults cannot be noticed from a distance. In this study, in order to solve this problem, a deep learning model that can detect insulator failures both at a certain distance and up close has been developed and implemented. As a result of the study, by applying convolutional neural networks, which are the most preferred among deep learning methods and can be successfully applied to every problem today, defective isolators are detected with a high accuracy rate. ORCID NO: 0000-0001-9105-508X
Anahtar Kelimeler: CNN, TL, Insulators, Overhead Transmission Line
|
