Entropy-Optimized Fault Diagnosis Based on Unsupervised Domain Adaptation
In practice, the cross-domain transfer of data distribution and the sample imbalance of fault status are inevitable, but one or both are often ignored, which restricts the adaptability and classification accuracy of the generated fault diagnosis (FD) model. Accordingly, an entropy-optimized method i...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-04-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/11/9/2110 |
| _version_ | 1827742876703916032 |
|---|---|
| author | Fuqiang Liu Yandan Chen Wenlong Deng Mingliang Zhou |
| author_facet | Fuqiang Liu Yandan Chen Wenlong Deng Mingliang Zhou |
| author_sort | Fuqiang Liu |
| collection | DOAJ |
| description | In practice, the cross-domain transfer of data distribution and the sample imbalance of fault status are inevitable, but one or both are often ignored, which restricts the adaptability and classification accuracy of the generated fault diagnosis (FD) model. Accordingly, an entropy-optimized method is proposed in this paper based on an unsupervised domain-adaptive technique to enhance FD model training. For the training, pseudosamples and labels corresponding to the target samples are generated through data augmentation and self-training strategies to diminish the distribution discrepancy between the source and target domains. Meanwhile, an adaptive conditional entropy loss function is developed to improve the data quality of the semisupervised learning, with which reliable samples are generated for the training. According to the experiment results, compared with other state-of-the-art algorithms, our method can achieve significant accuracy improvement in rolling bearing FD. Typically, the accuracy improvement compared with the baseline Convolutional Neural Network (CNN) is achieved by over 13.23%. |
| first_indexed | 2024-03-11T04:13:34Z |
| format | Article |
| id | doaj.art-d0478118f32e45c1a3eeef32e322725d |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-11T04:13:34Z |
| publishDate | 2023-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-d0478118f32e45c1a3eeef32e322725d2023-11-17T23:20:08ZengMDPI AGMathematics2227-73902023-04-01119211010.3390/math11092110Entropy-Optimized Fault Diagnosis Based on Unsupervised Domain AdaptationFuqiang Liu0Yandan Chen1Wenlong Deng2Mingliang Zhou3College of Mechanical and Vehicle Engineering, Chongqing University, Chongqing 400044, ChinaCollege of Mechanical and Vehicle Engineering, Chongqing University, Chongqing 400044, ChinaCollege of Mechanical and Vehicle Engineering, Chongqing University, Chongqing 400044, ChinaCollege of Computer Science, Chongqing University, Chongqing 400044, ChinaIn practice, the cross-domain transfer of data distribution and the sample imbalance of fault status are inevitable, but one or both are often ignored, which restricts the adaptability and classification accuracy of the generated fault diagnosis (FD) model. Accordingly, an entropy-optimized method is proposed in this paper based on an unsupervised domain-adaptive technique to enhance FD model training. For the training, pseudosamples and labels corresponding to the target samples are generated through data augmentation and self-training strategies to diminish the distribution discrepancy between the source and target domains. Meanwhile, an adaptive conditional entropy loss function is developed to improve the data quality of the semisupervised learning, with which reliable samples are generated for the training. According to the experiment results, compared with other state-of-the-art algorithms, our method can achieve significant accuracy improvement in rolling bearing FD. Typically, the accuracy improvement compared with the baseline Convolutional Neural Network (CNN) is achieved by over 13.23%.https://www.mdpi.com/2227-7390/11/9/2110class imbalancedomain adaptationentropy optimizationfault diagnosis (FD) |
| spellingShingle | Fuqiang Liu Yandan Chen Wenlong Deng Mingliang Zhou Entropy-Optimized Fault Diagnosis Based on Unsupervised Domain Adaptation Mathematics class imbalance domain adaptation entropy optimization fault diagnosis (FD) |
| title | Entropy-Optimized Fault Diagnosis Based on Unsupervised Domain Adaptation |
| title_full | Entropy-Optimized Fault Diagnosis Based on Unsupervised Domain Adaptation |
| title_fullStr | Entropy-Optimized Fault Diagnosis Based on Unsupervised Domain Adaptation |
| title_full_unstemmed | Entropy-Optimized Fault Diagnosis Based on Unsupervised Domain Adaptation |
| title_short | Entropy-Optimized Fault Diagnosis Based on Unsupervised Domain Adaptation |
| title_sort | entropy optimized fault diagnosis based on unsupervised domain adaptation |
| topic | class imbalance domain adaptation entropy optimization fault diagnosis (FD) |
| url | https://www.mdpi.com/2227-7390/11/9/2110 |
| work_keys_str_mv | AT fuqiangliu entropyoptimizedfaultdiagnosisbasedonunsuperviseddomainadaptation AT yandanchen entropyoptimizedfaultdiagnosisbasedonunsuperviseddomainadaptation AT wenlongdeng entropyoptimizedfaultdiagnosisbasedonunsuperviseddomainadaptation AT mingliangzhou entropyoptimizedfaultdiagnosisbasedonunsuperviseddomainadaptation |