Railway Fastener Fault Diagnosis Based on Generative Adversarial Network and Residual Network Model
The present work aimed at the problems of less negative samples and more positive samples in rail fastener fault diagnosis and low detection accuracy of heavy manual patrol inspection tasks. Exploiting the capacity of a Convolution Neural Network (CNN) to process unbalanced data to solve tedious and...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2020-01-01
|
| Series: | Shock and Vibration |
| Online Access: | http://dx.doi.org/10.1155/2020/8823050 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832565661905190912 |
|---|---|
| author | Dechen Yao Qiang Sun Jianwei Yang Hengchang Liu Jiao Zhang |
| author_facet | Dechen Yao Qiang Sun Jianwei Yang Hengchang Liu Jiao Zhang |
| author_sort | Dechen Yao |
| collection | DOAJ |
| description | The present work aimed at the problems of less negative samples and more positive samples in rail fastener fault diagnosis and low detection accuracy of heavy manual patrol inspection tasks. Exploiting the capacity of a Convolution Neural Network (CNN) to process unbalanced data to solve tedious and inefficient manual processing, a fault diagnosis method based on a Generative Adversarial Network (GAN) and a Residual Network (ResNet) was developed. First, GAN was used to track the distribution of rail fastener failure data. To study the noise distribution, the mapping relationship between image data was established. Additional real fault samples were then generated to balance and extend the existing data sets, and these data sets were used as input to ResNet for recognition and detection training. Finally, the average accuracy of multiple experiments was used as the evaluation index. The experimental results revealed that the fault diagnosis of rail fastener based on GAN and ResNet could improve the fault detection accuracy in the case of a serious shortage of fault data. |
| format | Article |
| id | doaj-art-9a4ab7341fdd4755bac84d446456b6a8 |
| institution | Kabale University |
| issn | 1070-9622 1875-9203 |
| language | English |
| publishDate | 2020-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Shock and Vibration |
| spelling | doaj-art-9a4ab7341fdd4755bac84d446456b6a82025-02-03T01:07:08ZengWileyShock and Vibration1070-96221875-92032020-01-01202010.1155/2020/88230508823050Railway Fastener Fault Diagnosis Based on Generative Adversarial Network and Residual Network ModelDechen Yao0Qiang Sun1Jianwei Yang2Hengchang Liu3Jiao Zhang4School of Mechanical-Electronic and Vehicle Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, ChinaSchool of Mechanical-Electronic and Vehicle Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, ChinaSchool of Mechanical-Electronic and Vehicle Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, ChinaSchool of Mechanical-Electronic and Vehicle Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, ChinaBeijing Mass Transit Railway Operation Corporation, Ltd., Beijing 100044, ChinaThe present work aimed at the problems of less negative samples and more positive samples in rail fastener fault diagnosis and low detection accuracy of heavy manual patrol inspection tasks. Exploiting the capacity of a Convolution Neural Network (CNN) to process unbalanced data to solve tedious and inefficient manual processing, a fault diagnosis method based on a Generative Adversarial Network (GAN) and a Residual Network (ResNet) was developed. First, GAN was used to track the distribution of rail fastener failure data. To study the noise distribution, the mapping relationship between image data was established. Additional real fault samples were then generated to balance and extend the existing data sets, and these data sets were used as input to ResNet for recognition and detection training. Finally, the average accuracy of multiple experiments was used as the evaluation index. The experimental results revealed that the fault diagnosis of rail fastener based on GAN and ResNet could improve the fault detection accuracy in the case of a serious shortage of fault data.http://dx.doi.org/10.1155/2020/8823050 |
| spellingShingle | Dechen Yao Qiang Sun Jianwei Yang Hengchang Liu Jiao Zhang Railway Fastener Fault Diagnosis Based on Generative Adversarial Network and Residual Network Model Shock and Vibration |
| title | Railway Fastener Fault Diagnosis Based on Generative Adversarial Network and Residual Network Model |
| title_full | Railway Fastener Fault Diagnosis Based on Generative Adversarial Network and Residual Network Model |
| title_fullStr | Railway Fastener Fault Diagnosis Based on Generative Adversarial Network and Residual Network Model |
| title_full_unstemmed | Railway Fastener Fault Diagnosis Based on Generative Adversarial Network and Residual Network Model |
| title_short | Railway Fastener Fault Diagnosis Based on Generative Adversarial Network and Residual Network Model |
| title_sort | railway fastener fault diagnosis based on generative adversarial network and residual network model |
| url | http://dx.doi.org/10.1155/2020/8823050 |
| work_keys_str_mv | AT dechenyao railwayfastenerfaultdiagnosisbasedongenerativeadversarialnetworkandresidualnetworkmodel AT qiangsun railwayfastenerfaultdiagnosisbasedongenerativeadversarialnetworkandresidualnetworkmodel AT jianweiyang railwayfastenerfaultdiagnosisbasedongenerativeadversarialnetworkandresidualnetworkmodel AT hengchangliu railwayfastenerfaultdiagnosisbasedongenerativeadversarialnetworkandresidualnetworkmodel AT jiaozhang railwayfastenerfaultdiagnosisbasedongenerativeadversarialnetworkandresidualnetworkmodel |