Transfer Learning Method Based on Adversarial Domain Adaption for Bearing Fault Diagnosis

At present, most of the intelligent fault diagnosis methods of rolling element bearings require sufficient labeled data for training. However, collecting labeled data is usually expensive and time-consuming, and when the distribution of the test data is different from the distribution of the trainin...

Full description

Bibliographic Details
Main Authors:	Jiajie Shao, Zhiwen Huang, Jianmin Zhu
Format:	Article
Language:	English
Published:	IEEE 2020-01-01
Series:	IEEE Access
Subjects:	Transfer learning fault diagnosis domain adaption deep learning
Online Access:	https://ieeexplore.ieee.org/document/9126795/

id	doaj-da686838d2c6410e8d09be48d0be310b
record_format	Article
spelling	doaj-da686838d2c6410e8d09be48d0be310b2021-03-30T01:54:59ZengIEEEIEEE Access2169-35362020-01-01811942111943010.1109/ACCESS.2020.30052439126795Transfer Learning Method Based on Adversarial Domain Adaption for Bearing Fault DiagnosisJiajie Shao0https://orcid.org/0000-0001-9998-676XZhiwen Huang1https://orcid.org/0000-0001-7325-3920Jianmin Zhu2https://orcid.org/0000-0002-7544-9721College of Mechanical Engineering, University of Shanghai for Science and Technology, Shanghai, ChinaCollege of Mechanical Engineering, University of Shanghai for Science and Technology, Shanghai, ChinaCollege of Mechanical Engineering, University of Shanghai for Science and Technology, Shanghai, ChinaAt present, most of the intelligent fault diagnosis methods of rolling element bearings require sufficient labeled data for training. However, collecting labeled data is usually expensive and time-consuming, and when the distribution of the test data is different from the distribution of the training data, the diagnostic performance will decrease. In order to solve the problem of unlabeled cross-domain diagnosis of bearings, this paper proposes an adversarial domain adaption method based on deep transfer learning. The short-time Fourier transform is used to transform the original data into a time-frequency image. The feature extractor is used to extract its deep features. The maximum mean discrepancy and domain confusion function are used for domain adaptation to extract domain-invariant features between two domains for cross-domain fault diagnosis. Experiments on two bearing datasets are carried out for validations. The results prove that the method in this paper is superior to other deep transfer learning methods. It shows the advantages of the improved method and can be used as an effective tool for cross-domain fault diagnosis.https://ieeexplore.ieee.org/document/9126795/Transfer learningfault diagnosisdomain adaptiondeep learning
collection	DOAJ
language	English
format	Article
sources	DOAJ
author	Jiajie Shao Zhiwen Huang Jianmin Zhu
spellingShingle	Jiajie Shao Zhiwen Huang Jianmin Zhu Transfer Learning Method Based on Adversarial Domain Adaption for Bearing Fault Diagnosis IEEE Access Transfer learning fault diagnosis domain adaption deep learning
author_facet	Jiajie Shao Zhiwen Huang Jianmin Zhu
author_sort	Jiajie Shao
title	Transfer Learning Method Based on Adversarial Domain Adaption for Bearing Fault Diagnosis
title_short	Transfer Learning Method Based on Adversarial Domain Adaption for Bearing Fault Diagnosis
title_full	Transfer Learning Method Based on Adversarial Domain Adaption for Bearing Fault Diagnosis
title_fullStr	Transfer Learning Method Based on Adversarial Domain Adaption for Bearing Fault Diagnosis
title_full_unstemmed	Transfer Learning Method Based on Adversarial Domain Adaption for Bearing Fault Diagnosis
title_sort	transfer learning method based on adversarial domain adaption for bearing fault diagnosis
publisher	IEEE
series	IEEE Access
issn	2169-3536
publishDate	2020-01-01
description	At present, most of the intelligent fault diagnosis methods of rolling element bearings require sufficient labeled data for training. However, collecting labeled data is usually expensive and time-consuming, and when the distribution of the test data is different from the distribution of the training data, the diagnostic performance will decrease. In order to solve the problem of unlabeled cross-domain diagnosis of bearings, this paper proposes an adversarial domain adaption method based on deep transfer learning. The short-time Fourier transform is used to transform the original data into a time-frequency image. The feature extractor is used to extract its deep features. The maximum mean discrepancy and domain confusion function are used for domain adaptation to extract domain-invariant features between two domains for cross-domain fault diagnosis. Experiments on two bearing datasets are carried out for validations. The results prove that the method in this paper is superior to other deep transfer learning methods. It shows the advantages of the improved method and can be used as an effective tool for cross-domain fault diagnosis.
topic	Transfer learning fault diagnosis domain adaption deep learning
url	https://ieeexplore.ieee.org/document/9126795/
work_keys_str_mv	AT jiajieshao transferlearningmethodbasedonadversarialdomainadaptionforbearingfaultdiagnosis AT zhiwenhuang transferlearningmethodbasedonadversarialdomainadaptionforbearingfaultdiagnosis AT jianminzhu transferlearningmethodbasedonadversarialdomainadaptionforbearingfaultdiagnosis
_version_	1724186266009862144

Transfer Learning Method Based on Adversarial Domain Adaption for Bearing Fault Diagnosis

Similar Items