Logo Repositorio Institucional

Please use this identifier to cite or link to this item: http://dspace.ucuenca.edu.ec/handle/123456789/33140
Title: Gearbox Fault Classification Using Dictionary Sparse Based Representations of Vibration Signals
Other Titles: Clasificación de fallas de la caja de engranajes usando representaciones dispersas basadas en diccionario de señales de vibración
Authors: Medina, Rubén
Jadan Aviles, Diana Carolina
Alvarez Palomeque, Lourdes Ximena
Macancela, Jean Carlo
Sánchez, René Vinicio
Cerrada, Mariela
metadata.dc.ucuenca.correspondencia: Medina, Rubén, ruben.djmedina@ieee.org
Keywords: Dictionary Learning
Sparse Representation
Vibration Signal
Gearbox Fault
Feature Extraction
metadata.dc.ucuenca.areaconocimientofrascatiamplio: 2. Ingeniería y Tecnología
metadata.dc.ucuenca.areaconocimientofrascatidetallado: 2.11.2 Otras Ingenierias y Tecnologías
metadata.dc.ucuenca.areaconocimientofrascatiespecifico: 2.11 Otras Ingenierias y Tecnologías
metadata.dc.ucuenca.areaconocimientounescoamplio: 07 - Ingeniería, Industria y Construcción
metadata.dc.ucuenca.areaconocimientounescodetallado: 0711 - Ingeniería y Procesos Químicos
metadata.dc.ucuenca.areaconocimientounescoespecifico: 071 - Ingeniería y Profesiones Afines
Issue Date: 2018
metadata.dc.ucuenca.volumen: volumen 34
metadata.dc.source: IOS Press
metadata.dc.type: ARTÍCULO
Abstract: 
Fault detection in rotating machinery is important for optimizing maintenance chores and avoiding severe damages to other parts. Signal processing based fault detection is usually performed by considering classical techniques for alternative representation of significant signals in time domain, frequency domain or time-frequency domain. An approach based on dictionary learning for sparse representations of vibration signals aiming at gearbox fault detection and classification is proposed. A gearbox signal dataset with 900 records considering the normal case and nine fault classes is analyzed. A dictionary is learned by using a training set of signals from the normal case. This dictionary is used for obtaining the sparse representation of signals in the test set and the norm metric is used to measure the residual from the sparse representation. The extracted features are useful for machine learning based fault detection. The analysis is performed considering different load conditions. ANOVA statistical analysis shows that there are significant differences between features in the normal case and each of the faulty classes, and best ranked features form well separated clusters. An experiment of fault classification is developed using a support vector machine for multi-class classification of faults. The accuracy obtained is 95.1% in the cross-validation testing.
Description: 
Fault detection in rotating machinery is important for optimizing maintenance chores and avoiding severe damages to other parts. Signal processing based fault detection is usually performed by considering classical techniques for alternative representation of significant signals in time domain, frequency domain or time-frequency domain. An approach based on dictionary learning for sparse representations of vibration signals aiming at gearbox fault detection and classification is proposed. A gearbox signal dataset with 900 records considering the normal case and nine fault classes is analyzed. A dictionary is learned by using a training set of signals from the normal case. This dictionary is used for obtaining the sparse representation of signals in the test set and the norm metric is used to measure the residual from the sparse representation. The extracted features are useful for machine learning based fault detection. The analysis is performed considering different load conditions. ANOVA statistical analysis shows that there are significant differences between features in the normal case and each of the faulty classes, and best ranked features form well separated clusters. An experiment of fault classification is developed using a support vector machine for multi-class classification of faults. The accuracy obtained is 95.1% in the cross-validation testing.
URI: http://dspace.ucuenca.edu.ec/handle/123456789/33140
https://content.iospress.com/articles/journal-of-intelligent-and-fuzzy-systems/ifs169537?resultNumber=0&totalResults=1&start=0&q=author%3A%28%22Medina%2C+Ruben%22%29&resultsPageSize=10&rows=10
metadata.dc.ucuenca.urifuente: https://content.iospress.com
ISSN: 1064-1246
Appears in Collections:Artículos

Files in This Item:
File Description SizeFormat 
documento.pdfdocument98.12 kBAdobe PDFThumbnail
View/Open


This item is protected by original copyright



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Centro de Documentacion Regional "Juan Bautista Vázquez"

Biblioteca Campus Central Biblioteca Campus Salud Biblioteca Campus Yanuncay
Av. 12 de Abril y Calle Agustín Cueva, Telf: 4051000 Ext. 1311, 1312, 1313, 1314. Horario de atención: Lunes-Viernes: 07H00-21H00. Sábados: 08H00-12H00 Av. El Paraíso 3-52, detrás del Hospital Regional "Vicente Corral Moscoso", Telf: 4051000 Ext. 3144. Horario de atención: Lunes-Viernes: 07H00-19H00 Av. 12 de Octubre y Diego de Tapia, antiguo Colegio Orientalista, Telf: 4051000 Ext. 3535 2810706 Ext. 116. Horario de atención: Lunes-Viernes: 07H30-19H00