Corona Effect Detection in Energized Polymeric Insulators Using Machine Learning and Ultrasonic Emissions

Corona Effect Detection in Energized Polymeric Insulators Using Machine Learning and Ultrasonic Emissions

The occurrence of corona effect is an important indication that an insulator probably has malfunction. Thus, the corona detection can help to diagnose problems in insulators, which might cause serious damage to the power distribution system, such as the sudden interruption of service. Artificial neu...

Full description

Saved in:
Bibliographic Details
Published in:Revista IEEE América Latina Vol. 16; no. 6; pp. 1587 - 1594
Main Authors: da Silva Junior, E.T., de Aquino, F.J.A., Silva, D.A., Rocha Neto, A.R., Gurgel, K.J.A., de Oliveira, A.E.R.M., de Araujo, A.L.C.
Format: Journal Article
Language:English
Portuguese
Published: Los Alamitos IEEE 01-06-2018
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Acoustics
Activation
Artificial intelligence
artificial neural network
Artificial neural networks
Circuits
Corona
corona effect
diagnostics
Electric power distribution
Electronic circuits
High voltages
Insulators
Learning theory
Machine learning
monitoring
Neural networks
polymeric insulator
Polymers
Random access memory
Support vector machines
ultrasonic sensor
Ultrasonic testing
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The occurrence of corona effect is an important indication that an insulator probably has malfunction. Thus, the corona detection can help to diagnose problems in insulators, which might cause serious damage to the power distribution system, such as the sudden interruption of service. Artificial neural networks (ANN) are machine learning techniques for classifying patterns into categories. This paper evaluated the application of neural networks for corona detection in high voltage polymeric insulators using ultrasonic emissions. To perform the capture of these data, a specific electronic circuit was developed to capture the ultrasonic signals tuned in the 40 kHz. The results permit to infer that neural networks showed high generalizability and looks to be a promising technique for this type of problem, using ultrasonic frequencies as the data source.
ISSN:1548-0992
1548-0992
DOI:10.1109/TLA.2018.8444153