Rolling Bearing Fault Diagnosis Based on Domain Adaptation and Preferred Feature Selection under Variable Working Conditions
In real industrial scenarios, with the use of conventional machine learning techniques, data-driven diagnosis models have a limitation that it is difficult to achieve the desirable fault diagnosis performance, and the reason is that the training and testing datasets are assumed to have the same feat...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-01-01
|
| Series: | Shock and Vibration |
| Online Access: | http://dx.doi.org/10.1155/2021/8843124 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832552485886099456 |
|---|---|
| author | Xiao Yu Wei Chen Chuanlong Wu Enjie Ding Yuanyuan Tian Haiwei Zuo Fei Dong |
| author_facet | Xiao Yu Wei Chen Chuanlong Wu Enjie Ding Yuanyuan Tian Haiwei Zuo Fei Dong |
| author_sort | Xiao Yu |
| collection | DOAJ |
| description | In real industrial scenarios, with the use of conventional machine learning techniques, data-driven diagnosis models have a limitation that it is difficult to achieve the desirable fault diagnosis performance, and the reason is that the training and testing datasets are assumed to have the same feature distributions. To address this problem, a novel bearing fault diagnosis framework based on domain adaptation and preferred feature selection is proposed, in that the model trained by the labeled data collected from a working condition can be applied to diagnose a new but similar target data collected from other working conditions. In this framework, an improved domain adaptation method, transfer component analysis with preserving local manifold structure (TCAPLMS), is proposed to reduce the differences in the data distributions between different domain datasets and, at the same time, take the label information of feature dataset and the local manifold structure of feature data into consideration. Furthermore, preferred feature selection by fault sensitivity and feature correlation (PSFFC) is embedded into this framework for selecting features which are more beneficial to fault pattern recognition and reduce the redundancy of feature set. Finally, vibration datasets collected from two test platforms are used for experimental analysis. The experimental results validate that the proposed method can obviously improve diagnosis accuracy and has significant potential benefits towards actual industrial scenarios. |
| format | Article |
| id | doaj-art-29dfa3e763f645a0983c5b3ddd85238a |
| institution | Kabale University |
| issn | 1070-9622 1875-9203 |
| language | English |
| publishDate | 2021-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Shock and Vibration |
| spelling | doaj-art-29dfa3e763f645a0983c5b3ddd85238a2025-02-03T05:58:30ZengWileyShock and Vibration1070-96221875-92032021-01-01202110.1155/2021/88431248843124Rolling Bearing Fault Diagnosis Based on Domain Adaptation and Preferred Feature Selection under Variable Working ConditionsXiao Yu0Wei Chen1Chuanlong Wu2Enjie Ding3Yuanyuan Tian4Haiwei Zuo5Fei Dong6IOT Perception Mine Research Center, China University of Mining and Technology, Xuzhou 221000, ChinaIOT Perception Mine Research Center, China University of Mining and Technology, Xuzhou 221000, ChinaIOT Perception Mine Research Center, China University of Mining and Technology, Xuzhou 221000, ChinaIOT Perception Mine Research Center, China University of Mining and Technology, Xuzhou 221000, ChinaIOT Perception Mine Research Center, China University of Mining and Technology, Xuzhou 221000, ChinaSchool of Medicine Information and Engineering, Xuzhou Medical University, Xuzhou 221000, ChinaIOT Perception Mine Research Center, China University of Mining and Technology, Xuzhou 221000, ChinaIn real industrial scenarios, with the use of conventional machine learning techniques, data-driven diagnosis models have a limitation that it is difficult to achieve the desirable fault diagnosis performance, and the reason is that the training and testing datasets are assumed to have the same feature distributions. To address this problem, a novel bearing fault diagnosis framework based on domain adaptation and preferred feature selection is proposed, in that the model trained by the labeled data collected from a working condition can be applied to diagnose a new but similar target data collected from other working conditions. In this framework, an improved domain adaptation method, transfer component analysis with preserving local manifold structure (TCAPLMS), is proposed to reduce the differences in the data distributions between different domain datasets and, at the same time, take the label information of feature dataset and the local manifold structure of feature data into consideration. Furthermore, preferred feature selection by fault sensitivity and feature correlation (PSFFC) is embedded into this framework for selecting features which are more beneficial to fault pattern recognition and reduce the redundancy of feature set. Finally, vibration datasets collected from two test platforms are used for experimental analysis. The experimental results validate that the proposed method can obviously improve diagnosis accuracy and has significant potential benefits towards actual industrial scenarios.http://dx.doi.org/10.1155/2021/8843124 |
| spellingShingle | Xiao Yu Wei Chen Chuanlong Wu Enjie Ding Yuanyuan Tian Haiwei Zuo Fei Dong Rolling Bearing Fault Diagnosis Based on Domain Adaptation and Preferred Feature Selection under Variable Working Conditions Shock and Vibration |
| title | Rolling Bearing Fault Diagnosis Based on Domain Adaptation and Preferred Feature Selection under Variable Working Conditions |
| title_full | Rolling Bearing Fault Diagnosis Based on Domain Adaptation and Preferred Feature Selection under Variable Working Conditions |
| title_fullStr | Rolling Bearing Fault Diagnosis Based on Domain Adaptation and Preferred Feature Selection under Variable Working Conditions |
| title_full_unstemmed | Rolling Bearing Fault Diagnosis Based on Domain Adaptation and Preferred Feature Selection under Variable Working Conditions |
| title_short | Rolling Bearing Fault Diagnosis Based on Domain Adaptation and Preferred Feature Selection under Variable Working Conditions |
| title_sort | rolling bearing fault diagnosis based on domain adaptation and preferred feature selection under variable working conditions |
| url | http://dx.doi.org/10.1155/2021/8843124 |
| work_keys_str_mv | AT xiaoyu rollingbearingfaultdiagnosisbasedondomainadaptationandpreferredfeatureselectionundervariableworkingconditions AT weichen rollingbearingfaultdiagnosisbasedondomainadaptationandpreferredfeatureselectionundervariableworkingconditions AT chuanlongwu rollingbearingfaultdiagnosisbasedondomainadaptationandpreferredfeatureselectionundervariableworkingconditions AT enjieding rollingbearingfaultdiagnosisbasedondomainadaptationandpreferredfeatureselectionundervariableworkingconditions AT yuanyuantian rollingbearingfaultdiagnosisbasedondomainadaptationandpreferredfeatureselectionundervariableworkingconditions AT haiweizuo rollingbearingfaultdiagnosisbasedondomainadaptationandpreferredfeatureselectionundervariableworkingconditions AT feidong rollingbearingfaultdiagnosisbasedondomainadaptationandpreferredfeatureselectionundervariableworkingconditions |