Research on Fault Diagnosis Method Based on Improved CNN
Traditional fault diagnosis methods require complex signal processing and expert experience, and the accuracy of fault identification is low. To solve these problems, a fault diagnosis method based on an improved convolutional neural network (CNN) is proposed. Based on the traditional CNN model, a v...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Hindawi Limited
2022-01-01
|
| Series: | Shock and Vibration |
| Online Access: | http://dx.doi.org/10.1155/2022/9312905 |
| _version_ | 1798005113223118848 |
|---|---|
| author | Hu Hao Feng Fuzhou Zhu Junzhen Zhou Xun Jiang Pengcheng Jiang Feng Xue Jun Li Yazhi Sun Guanghui |
| author_facet | Hu Hao Feng Fuzhou Zhu Junzhen Zhou Xun Jiang Pengcheng Jiang Feng Xue Jun Li Yazhi Sun Guanghui |
| author_sort | Hu Hao |
| collection | DOAJ |
| description | Traditional fault diagnosis methods require complex signal processing and expert experience, and the accuracy of fault identification is low. To solve these problems, a fault diagnosis method based on an improved convolutional neural network (CNN) is proposed. Based on the traditional CNN model, a variety of convolution stride modes were added to extract features of different scales of signals and expand the feature dimension. Firstly, the vibration signals were collected and grouped. Then, the data were divided into a training set and a test set and input into improved CNN for feature extraction and model training to realize fault identification. The proposed model achieved a classification accuracy of 99.3% when testing the vibration data of the armored vehicle. Finally, the proposed model was used to classify different fault types of planetary gearboxes. The gradient-weighted class activation mapping (Grad-CAM) method was used to visualize the classification weight of samples. The results showed that the classification accuracy reaches 98% under various working conditions of the planetary gearbox. |
| first_indexed | 2024-04-11T12:34:04Z |
| format | Article |
| id | doaj.art-04bc5778015342dabb679c7c4d12503e |
| institution | Directory Open Access Journal |
| issn | 1875-9203 |
| language | English |
| last_indexed | 2024-04-11T12:34:04Z |
| publishDate | 2022-01-01 |
| publisher | Hindawi Limited |
| record_format | Article |
| series | Shock and Vibration |
| spelling | doaj.art-04bc5778015342dabb679c7c4d12503e2022-12-22T04:23:40ZengHindawi LimitedShock and Vibration1875-92032022-01-01202210.1155/2022/9312905Research on Fault Diagnosis Method Based on Improved CNNHu Hao0Feng Fuzhou1Zhu Junzhen2Zhou Xun3Jiang Pengcheng4Jiang Feng5Xue Jun6Li Yazhi7Sun Guanghui8Department of Vehicle EngineeringDepartment of Vehicle EngineeringDepartment of Vehicle EngineeringDepartment of Vehicle EngineeringDepartment of Vehicle EngineeringDepartment of Battlefield RepairDepartment of Battlefield RepairDepartment of Battlefield RepairDepartment of Battlefield RepairTraditional fault diagnosis methods require complex signal processing and expert experience, and the accuracy of fault identification is low. To solve these problems, a fault diagnosis method based on an improved convolutional neural network (CNN) is proposed. Based on the traditional CNN model, a variety of convolution stride modes were added to extract features of different scales of signals and expand the feature dimension. Firstly, the vibration signals were collected and grouped. Then, the data were divided into a training set and a test set and input into improved CNN for feature extraction and model training to realize fault identification. The proposed model achieved a classification accuracy of 99.3% when testing the vibration data of the armored vehicle. Finally, the proposed model was used to classify different fault types of planetary gearboxes. The gradient-weighted class activation mapping (Grad-CAM) method was used to visualize the classification weight of samples. The results showed that the classification accuracy reaches 98% under various working conditions of the planetary gearbox.http://dx.doi.org/10.1155/2022/9312905 |
| spellingShingle | Hu Hao Feng Fuzhou Zhu Junzhen Zhou Xun Jiang Pengcheng Jiang Feng Xue Jun Li Yazhi Sun Guanghui Research on Fault Diagnosis Method Based on Improved CNN Shock and Vibration |
| title | Research on Fault Diagnosis Method Based on Improved CNN |
| title_full | Research on Fault Diagnosis Method Based on Improved CNN |
| title_fullStr | Research on Fault Diagnosis Method Based on Improved CNN |
| title_full_unstemmed | Research on Fault Diagnosis Method Based on Improved CNN |
| title_short | Research on Fault Diagnosis Method Based on Improved CNN |
| title_sort | research on fault diagnosis method based on improved cnn |
| url | http://dx.doi.org/10.1155/2022/9312905 |
| work_keys_str_mv | AT huhao researchonfaultdiagnosismethodbasedonimprovedcnn AT fengfuzhou researchonfaultdiagnosismethodbasedonimprovedcnn AT zhujunzhen researchonfaultdiagnosismethodbasedonimprovedcnn AT zhouxun researchonfaultdiagnosismethodbasedonimprovedcnn AT jiangpengcheng researchonfaultdiagnosismethodbasedonimprovedcnn AT jiangfeng researchonfaultdiagnosismethodbasedonimprovedcnn AT xuejun researchonfaultdiagnosismethodbasedonimprovedcnn AT liyazhi researchonfaultdiagnosismethodbasedonimprovedcnn AT sunguanghui researchonfaultdiagnosismethodbasedonimprovedcnn |