A One-Versus-All Class Binarization Strategy for Bearing Diagnostics of Concurrent Defects
In bearing diagnostics using a data-driven modeling approach, a concern is the need for data from all possible scenarios to build a practical model for all operating conditions. This paper is a study on bearing diagnostics with the concurrent occurrence of multiple defect types. The authors are not...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2014-01-01
|
Series: | Sensors |
Subjects: | |
Online Access: | http://www.mdpi.com/1424-8220/14/1/1295 |
id |
doaj-fea0a20978a64a2a9d130c4c12c308ee |
---|---|
record_format |
Article |
spelling |
doaj-fea0a20978a64a2a9d130c4c12c308ee2020-11-24T21:41:21ZengMDPI AGSensors1424-82202014-01-011411295132110.3390/s140101295s140101295A One-Versus-All Class Binarization Strategy for Bearing Diagnostics of Concurrent DefectsSelina S. Y. Ng0Peter W. Tse1Kwok L. Tsui2Department of Systems Engineering and Engineering Management (SEEM), City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, ChinaDepartment of Systems Engineering and Engineering Management (SEEM), City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, ChinaDepartment of Systems Engineering and Engineering Management (SEEM), City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, ChinaIn bearing diagnostics using a data-driven modeling approach, a concern is the need for data from all possible scenarios to build a practical model for all operating conditions. This paper is a study on bearing diagnostics with the concurrent occurrence of multiple defect types. The authors are not aware of any work in the literature that studies this practical problem. A strategy based on one-versus-all (OVA) class binarization is proposed to improve fault diagnostics accuracy while reducing the number of scenarios for data collection, by predicting concurrent defects from training data of normal and single defects. The proposed OVA diagnostic approach is evaluated with empirical analysis using support vector machine (SVM) and C4.5 decision tree, two popular classification algorithms frequently applied to system health diagnostics and prognostics. Statistical features are extracted from the time domain and the frequency domain. Prediction performance of the proposed strategy is compared with that of a simple multi-class classification, as well as that of random guess and worst-case classification. We have verified the potential of the proposed OVA diagnostic strategy in performance improvements for single-defect diagnosis and predictions of BPFO plus BPFI concurrent defects using two laboratory-collected vibration data sets.http://www.mdpi.com/1424-8220/14/1/1295bearingmultiple defectsfault diagnosticsclass binarizationsupport vector machine (SVM)decision tree |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Selina S. Y. Ng Peter W. Tse Kwok L. Tsui |
spellingShingle |
Selina S. Y. Ng Peter W. Tse Kwok L. Tsui A One-Versus-All Class Binarization Strategy for Bearing Diagnostics of Concurrent Defects Sensors bearing multiple defects fault diagnostics class binarization support vector machine (SVM) decision tree |
author_facet |
Selina S. Y. Ng Peter W. Tse Kwok L. Tsui |
author_sort |
Selina S. Y. Ng |
title |
A One-Versus-All Class Binarization Strategy for Bearing Diagnostics of Concurrent Defects |
title_short |
A One-Versus-All Class Binarization Strategy for Bearing Diagnostics of Concurrent Defects |
title_full |
A One-Versus-All Class Binarization Strategy for Bearing Diagnostics of Concurrent Defects |
title_fullStr |
A One-Versus-All Class Binarization Strategy for Bearing Diagnostics of Concurrent Defects |
title_full_unstemmed |
A One-Versus-All Class Binarization Strategy for Bearing Diagnostics of Concurrent Defects |
title_sort |
one-versus-all class binarization strategy for bearing diagnostics of concurrent defects |
publisher |
MDPI AG |
series |
Sensors |
issn |
1424-8220 |
publishDate |
2014-01-01 |
description |
In bearing diagnostics using a data-driven modeling approach, a concern is the need for data from all possible scenarios to build a practical model for all operating conditions. This paper is a study on bearing diagnostics with the concurrent occurrence of multiple defect types. The authors are not aware of any work in the literature that studies this practical problem. A strategy based on one-versus-all (OVA) class binarization is proposed to improve fault diagnostics accuracy while reducing the number of scenarios for data collection, by predicting concurrent defects from training data of normal and single defects. The proposed OVA diagnostic approach is evaluated with empirical analysis using support vector machine (SVM) and C4.5 decision tree, two popular classification algorithms frequently applied to system health diagnostics and prognostics. Statistical features are extracted from the time domain and the frequency domain. Prediction performance of the proposed strategy is compared with that of a simple multi-class classification, as well as that of random guess and worst-case classification. We have verified the potential of the proposed OVA diagnostic strategy in performance improvements for single-defect diagnosis and predictions of BPFO plus BPFI concurrent defects using two laboratory-collected vibration data sets. |
topic |
bearing multiple defects fault diagnostics class binarization support vector machine (SVM) decision tree |
url |
http://www.mdpi.com/1424-8220/14/1/1295 |
work_keys_str_mv |
AT selinasyng aoneversusallclassbinarizationstrategyforbearingdiagnosticsofconcurrentdefects AT peterwtse aoneversusallclassbinarizationstrategyforbearingdiagnosticsofconcurrentdefects AT kwokltsui aoneversusallclassbinarizationstrategyforbearingdiagnosticsofconcurrentdefects AT selinasyng oneversusallclassbinarizationstrategyforbearingdiagnosticsofconcurrentdefects AT peterwtse oneversusallclassbinarizationstrategyforbearingdiagnosticsofconcurrentdefects AT kwokltsui oneversusallclassbinarizationstrategyforbearingdiagnosticsofconcurrentdefects |
_version_ |
1725922495959138304 |