GADF-VGG16 based fault diagnosis method for HVDC transmission lines.
Transmission lines are most prone to faults in the transmission system, so high-precision fault diagnosis is very important for quick troubleshooting. There are some problems in current intelligent fault diagnosis research methods, such as difficulty in extracting fault features accurately, low faul...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2022-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0274613 |
| _version_ | 1811206410141171712 |
|---|---|
| author | Hao Wu Yuping Yang Sijing Deng Qiaomei Wang Hong Song |
| author_facet | Hao Wu Yuping Yang Sijing Deng Qiaomei Wang Hong Song |
| author_sort | Hao Wu |
| collection | DOAJ |
| description | Transmission lines are most prone to faults in the transmission system, so high-precision fault diagnosis is very important for quick troubleshooting. There are some problems in current intelligent fault diagnosis research methods, such as difficulty in extracting fault features accurately, low fault recognition accuracy and poor fault tolerance. In order to solve these problems, this paper proposes an intelligent fault diagnosis method for high voltage direct current transmission lines (HVDC) based on Gramian angular difference field (GADF) domain and improved convolutional neural network (VGG16). This method first performs variational modal decomposition (VMD) on the original fault voltage signal, and then uses the correlation coefficient method to select the appropriate intrinsic mode function (IMF) component, and converts it into a two-dimensional image using the Gramian Angular Difference Field(GADF). Finally, the improved VGG16 network is used to extract and classify fault features adaptively to realize fault diagnosis. In order to improve the performance of the VGG16 fault diagnosis model, batch normalization, dense connection and global average pooling techniques are introduced. The comparative experimental results show that the model proposed in this paper can further identify fault features and has a high fault diagnosis accuracy. In addition, the method is not affected by fault type, transitional resistance and fault distance, has good anti-interference ability, strong fault tolerance, and has great potential in practical applications. |
| first_indexed | 2024-04-12T03:47:00Z |
| format | Article |
| id | doaj.art-f38107040c30464da5fa15c9c4f740ba |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-04-12T03:47:00Z |
| publishDate | 2022-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-f38107040c30464da5fa15c9c4f740ba2022-12-22T03:49:07ZengPublic Library of Science (PLoS)PLoS ONE1932-62032022-01-01179e027461310.1371/journal.pone.0274613GADF-VGG16 based fault diagnosis method for HVDC transmission lines.Hao WuYuping YangSijing DengQiaomei WangHong SongTransmission lines are most prone to faults in the transmission system, so high-precision fault diagnosis is very important for quick troubleshooting. There are some problems in current intelligent fault diagnosis research methods, such as difficulty in extracting fault features accurately, low fault recognition accuracy and poor fault tolerance. In order to solve these problems, this paper proposes an intelligent fault diagnosis method for high voltage direct current transmission lines (HVDC) based on Gramian angular difference field (GADF) domain and improved convolutional neural network (VGG16). This method first performs variational modal decomposition (VMD) on the original fault voltage signal, and then uses the correlation coefficient method to select the appropriate intrinsic mode function (IMF) component, and converts it into a two-dimensional image using the Gramian Angular Difference Field(GADF). Finally, the improved VGG16 network is used to extract and classify fault features adaptively to realize fault diagnosis. In order to improve the performance of the VGG16 fault diagnosis model, batch normalization, dense connection and global average pooling techniques are introduced. The comparative experimental results show that the model proposed in this paper can further identify fault features and has a high fault diagnosis accuracy. In addition, the method is not affected by fault type, transitional resistance and fault distance, has good anti-interference ability, strong fault tolerance, and has great potential in practical applications.https://doi.org/10.1371/journal.pone.0274613 |
| spellingShingle | Hao Wu Yuping Yang Sijing Deng Qiaomei Wang Hong Song GADF-VGG16 based fault diagnosis method for HVDC transmission lines. PLoS ONE |
| title | GADF-VGG16 based fault diagnosis method for HVDC transmission lines. |
| title_full | GADF-VGG16 based fault diagnosis method for HVDC transmission lines. |
| title_fullStr | GADF-VGG16 based fault diagnosis method for HVDC transmission lines. |
| title_full_unstemmed | GADF-VGG16 based fault diagnosis method for HVDC transmission lines. |
| title_short | GADF-VGG16 based fault diagnosis method for HVDC transmission lines. |
| title_sort | gadf vgg16 based fault diagnosis method for hvdc transmission lines |
| url | https://doi.org/10.1371/journal.pone.0274613 |
| work_keys_str_mv | AT haowu gadfvgg16basedfaultdiagnosismethodforhvdctransmissionlines AT yupingyang gadfvgg16basedfaultdiagnosismethodforhvdctransmissionlines AT sijingdeng gadfvgg16basedfaultdiagnosismethodforhvdctransmissionlines AT qiaomeiwang gadfvgg16basedfaultdiagnosismethodforhvdctransmissionlines AT hongsong gadfvgg16basedfaultdiagnosismethodforhvdctransmissionlines |