Deep Adaptive Adversarial Network-Based Method for Mechanical Fault Diagnosis under Different Working Conditions
The demand for transfer learning methods for mechanical fault diagnosis has considerably progressed in recent years. However, the existing methods always depend on the maximum mean discrepancy (MMD) in measuring the domain discrepancy. But MMD can not guarantee the different domain features to be si...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2020-01-01
|
| Series: | Complexity |
| Online Access: | http://dx.doi.org/10.1155/2020/6946702 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849408661306212352 |
|---|---|
| author | Jinrui Wang Shanshan Ji Baokun Han Huaiqian Bao Xingxing Jiang |
| author_facet | Jinrui Wang Shanshan Ji Baokun Han Huaiqian Bao Xingxing Jiang |
| author_sort | Jinrui Wang |
| collection | DOAJ |
| description | The demand for transfer learning methods for mechanical fault diagnosis has considerably progressed in recent years. However, the existing methods always depend on the maximum mean discrepancy (MMD) in measuring the domain discrepancy. But MMD can not guarantee the different domain features to be similar enough. Inspired by generative adversarial networks (GAN) and domain adversarial training of neural networks (DANN), this study presents a novel deep adaptive adversarial network (DAAN). The DAAN comprises a condition recognition module and domain adversarial learning module. The condition recognition module is constructed with a generator to extract features and classify the health condition of machinery automatically. The domain adversarial learning module is achieved with a discriminator based on Wasserstein distance to learn domain-invariant features. Then spectral normalization (SN) is employed to accelerate convergence. The effectiveness of DAAN is demonstrated through three transfer fault diagnosis experiments, and the results show that the DAAN can converge to zero after approximately 15 training epochs, and all the average testing accuracies in each case can achieve over 92%. It is expected that the proposed DAAN can effectively learn domain-invariant features to bridge the discrepancy between the data from different working conditions. |
| format | Article |
| id | doaj-art-7bcabb811e19459fbb52cab3991da6c7 |
| institution | Kabale University |
| issn | 1076-2787 1099-0526 |
| language | English |
| publishDate | 2020-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Complexity |
| spelling | doaj-art-7bcabb811e19459fbb52cab3991da6c72025-08-20T03:35:44ZengWileyComplexity1076-27871099-05262020-01-01202010.1155/2020/69467026946702Deep Adaptive Adversarial Network-Based Method for Mechanical Fault Diagnosis under Different Working ConditionsJinrui Wang0Shanshan Ji1Baokun Han2Huaiqian Bao3Xingxing Jiang4College of Mechanical and Electronic Engineering, Shandong University of Science and Technology, Qingdao, ChinaCollege of Mechanical and Electronic Engineering, Shandong University of Science and Technology, Qingdao, ChinaCollege of Mechanical and Electronic Engineering, Shandong University of Science and Technology, Qingdao, ChinaCollege of Mechanical and Electronic Engineering, Shandong University of Science and Technology, Qingdao, ChinaSchool of Rail Transportation, Soochow University, Suzhou, ChinaThe demand for transfer learning methods for mechanical fault diagnosis has considerably progressed in recent years. However, the existing methods always depend on the maximum mean discrepancy (MMD) in measuring the domain discrepancy. But MMD can not guarantee the different domain features to be similar enough. Inspired by generative adversarial networks (GAN) and domain adversarial training of neural networks (DANN), this study presents a novel deep adaptive adversarial network (DAAN). The DAAN comprises a condition recognition module and domain adversarial learning module. The condition recognition module is constructed with a generator to extract features and classify the health condition of machinery automatically. The domain adversarial learning module is achieved with a discriminator based on Wasserstein distance to learn domain-invariant features. Then spectral normalization (SN) is employed to accelerate convergence. The effectiveness of DAAN is demonstrated through three transfer fault diagnosis experiments, and the results show that the DAAN can converge to zero after approximately 15 training epochs, and all the average testing accuracies in each case can achieve over 92%. It is expected that the proposed DAAN can effectively learn domain-invariant features to bridge the discrepancy between the data from different working conditions.http://dx.doi.org/10.1155/2020/6946702 |
| spellingShingle | Jinrui Wang Shanshan Ji Baokun Han Huaiqian Bao Xingxing Jiang Deep Adaptive Adversarial Network-Based Method for Mechanical Fault Diagnosis under Different Working Conditions Complexity |
| title | Deep Adaptive Adversarial Network-Based Method for Mechanical Fault Diagnosis under Different Working Conditions |
| title_full | Deep Adaptive Adversarial Network-Based Method for Mechanical Fault Diagnosis under Different Working Conditions |
| title_fullStr | Deep Adaptive Adversarial Network-Based Method for Mechanical Fault Diagnosis under Different Working Conditions |
| title_full_unstemmed | Deep Adaptive Adversarial Network-Based Method for Mechanical Fault Diagnosis under Different Working Conditions |
| title_short | Deep Adaptive Adversarial Network-Based Method for Mechanical Fault Diagnosis under Different Working Conditions |
| title_sort | deep adaptive adversarial network based method for mechanical fault diagnosis under different working conditions |
| url | http://dx.doi.org/10.1155/2020/6946702 |
| work_keys_str_mv | AT jinruiwang deepadaptiveadversarialnetworkbasedmethodformechanicalfaultdiagnosisunderdifferentworkingconditions AT shanshanji deepadaptiveadversarialnetworkbasedmethodformechanicalfaultdiagnosisunderdifferentworkingconditions AT baokunhan deepadaptiveadversarialnetworkbasedmethodformechanicalfaultdiagnosisunderdifferentworkingconditions AT huaiqianbao deepadaptiveadversarialnetworkbasedmethodformechanicalfaultdiagnosisunderdifferentworkingconditions AT xingxingjiang deepadaptiveadversarialnetworkbasedmethodformechanicalfaultdiagnosisunderdifferentworkingconditions |