Identificación de fallas en los aisladores de una línea de transmisión mediante visión artificial.
This research is based on the superficial fault’s identification in the porcelain insulators of the 13.8 kV transmission line of the cities of Latacunga and Salcedo through artificial vision, to know the status of these elements in an easy and simple way. Development began with a database of 3000 im...
| Autor: | |
|---|---|
| Formato: | tesis de maestría |
| Estado: | Versión publicada |
| Fecha de publicación: | 2022 |
| País: | Ecuador |
| Recursos: | Universidad Técnica de Cotopaxi |
| Repositorio: | Repositorio Universidad Técnica de Cotopaxi |
| Idioma: | español |
| OAI Identifier: | oai:oai:repositorio.utc.edu.ec:27000:27000/8801 |
| Acesso em linha: | http://repositorio.utc.edu.ec/handle/27000/8801 |
| Access Level: | acceso abierto |
| Palavra-chave: | VISIÓN ARTIFICIAL ALGORITMO AISLADORES REDES NEURONALES ELECTRICIDAD |
| Resumo: | This research is based on the superficial fault’s identification in the porcelain insulators of the 13.8 kV transmission line of the cities of Latacunga and Salcedo through artificial vision, to know the status of these elements in an easy and simple way. Development began with a database of 3000 images of insulators divides into three states: dirty/flamed, broken and in good condition. With these images, Yolo was trained as identification algorithm, which is based on neural networks, the identification algorithm works both in real time and on pre-recorded images and videos. To make it’s handling as easy as possible, a graphical interface was designed to be used by a single operator. The field tests carried out showed that there is a high percentage of insulators that need maintenance since they are broken and/or dirty/flamed with a percentage of 50% of the samples, additionally the algorithm has an accuracy of 87.5% when works in real time, while the algorithm applied to images had an efficiency of 94% finally in videos of 89%. In conclusion, its implementation is feasible within the maintenance plans of the transmission lines, lowering costs, because it is not necessary to have a crane car for the inspection, in fact, it is just necessary an operator with a drone and a personal computer to make the decision of whether or not to perform maintenance. |
|---|