Institutional Repository
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/72FDEF98-9BDE-4959-B1A2-83C4DEFC806C | ||
| Year | 2020 | ||
| Type of Item | Master Thesis | ||
| License |
|
||
| Bibliographic Citation | Alexandros Zormpas, "Power transmission lines fault detection algorithm using properly equipped unmanned aerial vehicle (UAV)", Master Thesis, School of Electrical and Computer Engineering, Technical University of Crete, Chania, Greece, 2020 https://doi.org/10.26233/heallink.tuc.84840 | ||
| Appears in Collections |
The inspection and maintenance of the electricity transmission and distribution network has been and continues to be a major issue to ensure its stability. This paper examines the effectiveness of modern methods of image processing and artificial intelligence and also examines their limitations for facilitating inspection. The proposed methodology, written in Python, consists of three stages. The first locates the powerlines through a neural network trained for the specific application. In the second stage, an attempt is made to detect the crossarms, exclusively in the parts of the image where powerlines have been detected. The Gabor filter allows the crossarm to be located in different directions by filtering the image with a corresponding kernel. In the third and final stage, two separate methodologies are proposed to detect the insulators. One is based on morphological filtering and the other on training a neural network to locate the body of the insulator. Finally, according to the detected pixels, the infrared image is searched to extract the temperature of each insulator and to detect errors if any.
Copyright © DIAS 2013
