An Approach to Condition Assessment of High-Voltage Insulators by Ultrasound and an Ensemble of Fourier Neural Networks

This paper proposes an approach for evaluating the condition of high-voltage insulators of overhead distribution lines using Neural Networks and pre-processing techniques to analyze the results of ultrasound tests provided by a probe equipment. It presents the methodology of the laboratory tests in...

Full description

Saved in:
Bibliographic Details
Main Authors: de Campos, Arthur Franklim Marques, de Souza Pereira Gomes, Gabriel, Araujo, Daniel Carrijo Polonio, da Silva, Frederico Dourado, Fehlberg, Rafael Prux, Sardinha, Bruno Fernandes, Flauzino, Rozerio Andrade
Format: Conference Proceeding
Language:English
Subjects:
Condition Assessment
Distribution Networks
Ensemble
Inspection
Insulators
Neural networks
Testing
Training
Ultrasonic imaging
Ultrasonic variables measurement
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper proposes an approach for evaluating the condition of high-voltage insulators of overhead distribution lines using Neural Networks and pre-processing techniques to analyze the results of ultrasound tests provided by a probe equipment. It presents the methodology of the laboratory tests in which the ultrasound signals of insulator chains were collected in several situations, besides the treatment of the obtained data. Next, the neural network architectures and the flowchart of the insulation condition analysis methodology are detailed. Finally, the results of the diagnostics resulting from the network with the training sets and test sets are presented and discussed. The proposed methodology achieved 100 % accuracy in validation and test data. The future goal is to implement a system that allows field crews to easily identify insulator failures using simple ultrasound sensors.
ISSN:2160-8563
DOI:10.1109/TD39804.2020.9300030