A Novel Unsupervised Learning Method Based on Cross-Normalization for Machinery Fault Diagnosis
Sparse representation is the important principle of unsupervised learning method. In order to accurately identify the fault condition of machines, the desired feature distribution should show population sparsity and lifetime sparsity. In this paper, to improve the accuracy and robustness of the clas...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2020-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9086129/ |
id |
doaj-8834c0c744ff4902b8418d3647bc59e5 |
---|---|
record_format |
Article |
spelling |
doaj-8834c0c744ff4902b8418d3647bc59e52021-03-30T01:35:37ZengIEEEIEEE Access2169-35362020-01-018924079241710.1109/ACCESS.2020.29922019086129A Novel Unsupervised Learning Method Based on Cross-Normalization for Machinery Fault DiagnosisZongzhen Zhang0https://orcid.org/0000-0002-2022-2116Shunming Li1https://orcid.org/0000-0002-1271-6036Jiantao Lu2https://orcid.org/0000-0002-9674-165XJinrui Wang3https://orcid.org/0000-0001-8690-0672Xingxing Jiang4https://orcid.org/0000-0003-2987-6930College of Energy and Power Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, ChinaCollege of Energy and Power Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, ChinaCollege of Energy and Power Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, ChinaCollege of Mechanical and Electronic Engineering, Shandong University of Science and Technology, Qingdao, ChinaSchool of Rail Transportation, Soochow University, Suzhou, ChinaSparse representation is the important principle of unsupervised learning method. In order to accurately identify the fault condition of machines, the desired feature distribution should show population sparsity and lifetime sparsity. In this paper, to improve the accuracy and robustness of the classification, a novel fault diagnosis method named Cross-sparse Filtering (Cr-SF) is proposed based on the cross l<sub>1/2</sub>-norms of the feature matrix, which mean the population sparsity and lifetime sparsity terms. After the weights training process, a novel nonlinear activation function is used for feature extraction in the test process. Cr-SF can learn discriminative features from the raw data and accurately identify the fault condition. Rolling bearing fault and gear-box fault datasets are employed to validate the performance of the proposed method. The verification results confirm that Cr-SF is an effective tool for handling big data. The robustness and accuracy of the classification results using Cr-SF are comparable to convolutional networks with a much faster training process.https://ieeexplore.ieee.org/document/9086129/Cross-normalizationunsupervised learningintelligent fault diagnosisrotating machinerybig data |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Zongzhen Zhang Shunming Li Jiantao Lu Jinrui Wang Xingxing Jiang |
spellingShingle |
Zongzhen Zhang Shunming Li Jiantao Lu Jinrui Wang Xingxing Jiang A Novel Unsupervised Learning Method Based on Cross-Normalization for Machinery Fault Diagnosis IEEE Access Cross-normalization unsupervised learning intelligent fault diagnosis rotating machinery big data |
author_facet |
Zongzhen Zhang Shunming Li Jiantao Lu Jinrui Wang Xingxing Jiang |
author_sort |
Zongzhen Zhang |
title |
A Novel Unsupervised Learning Method Based on Cross-Normalization for Machinery Fault Diagnosis |
title_short |
A Novel Unsupervised Learning Method Based on Cross-Normalization for Machinery Fault Diagnosis |
title_full |
A Novel Unsupervised Learning Method Based on Cross-Normalization for Machinery Fault Diagnosis |
title_fullStr |
A Novel Unsupervised Learning Method Based on Cross-Normalization for Machinery Fault Diagnosis |
title_full_unstemmed |
A Novel Unsupervised Learning Method Based on Cross-Normalization for Machinery Fault Diagnosis |
title_sort |
novel unsupervised learning method based on cross-normalization for machinery fault diagnosis |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2020-01-01 |
description |
Sparse representation is the important principle of unsupervised learning method. In order to accurately identify the fault condition of machines, the desired feature distribution should show population sparsity and lifetime sparsity. In this paper, to improve the accuracy and robustness of the classification, a novel fault diagnosis method named Cross-sparse Filtering (Cr-SF) is proposed based on the cross l<sub>1/2</sub>-norms of the feature matrix, which mean the population sparsity and lifetime sparsity terms. After the weights training process, a novel nonlinear activation function is used for feature extraction in the test process. Cr-SF can learn discriminative features from the raw data and accurately identify the fault condition. Rolling bearing fault and gear-box fault datasets are employed to validate the performance of the proposed method. The verification results confirm that Cr-SF is an effective tool for handling big data. The robustness and accuracy of the classification results using Cr-SF are comparable to convolutional networks with a much faster training process. |
topic |
Cross-normalization unsupervised learning intelligent fault diagnosis rotating machinery big data |
url |
https://ieeexplore.ieee.org/document/9086129/ |
work_keys_str_mv |
AT zongzhenzhang anovelunsupervisedlearningmethodbasedoncrossnormalizationformachineryfaultdiagnosis AT shunmingli anovelunsupervisedlearningmethodbasedoncrossnormalizationformachineryfaultdiagnosis AT jiantaolu anovelunsupervisedlearningmethodbasedoncrossnormalizationformachineryfaultdiagnosis AT jinruiwang anovelunsupervisedlearningmethodbasedoncrossnormalizationformachineryfaultdiagnosis AT xingxingjiang anovelunsupervisedlearningmethodbasedoncrossnormalizationformachineryfaultdiagnosis AT zongzhenzhang novelunsupervisedlearningmethodbasedoncrossnormalizationformachineryfaultdiagnosis AT shunmingli novelunsupervisedlearningmethodbasedoncrossnormalizationformachineryfaultdiagnosis AT jiantaolu novelunsupervisedlearningmethodbasedoncrossnormalizationformachineryfaultdiagnosis AT jinruiwang novelunsupervisedlearningmethodbasedoncrossnormalizationformachineryfaultdiagnosis AT xingxingjiang novelunsupervisedlearningmethodbasedoncrossnormalizationformachineryfaultdiagnosis |
_version_ |
1724186804977926144 |