Bearing fault diagnosis for variable operating conditions based on KAN convolution and dual branch fusion attention
Abstract This paper proposes a bearing fault diagnosis method based on Kolmogorov–Arnold Convolutional Network: Adaptive Context-aware Graph Channel Attention with Squeeze-and-Excitation Networks (KANConv-ACGCA-SENet). Firstly, a new structure of KANs is applied to Convolutional Neural Networks (CNN...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-07-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-04620-1 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849238485509079040 |
|---|---|
| author | Qibing Wang Chuanjie Yin Kun She Qinfeng Tong Guoxiong Lu Hongbing Zhang Jiawei Lu |
| author_facet | Qibing Wang Chuanjie Yin Kun She Qinfeng Tong Guoxiong Lu Hongbing Zhang Jiawei Lu |
| author_sort | Qibing Wang |
| collection | DOAJ |
| description | Abstract This paper proposes a bearing fault diagnosis method based on Kolmogorov–Arnold Convolutional Network: Adaptive Context-aware Graph Channel Attention with Squeeze-and-Excitation Networks (KANConv-ACGCA-SENet). Firstly, a new structure of KANs is applied to Convolutional Neural Networks (CNN) for replacing traditional linear convolutional kernels. Secondly, a dual-branch fusion attention module, comprising the ACGCA modules, is proposed for use in learning fault features. This is achieved by capturing feature differences and utilising non-local(NL) operations, thereby enhancing the feature representation ability under different working conditions. Subsequently, context-aware features and non-local aggregation features are combined with the objective of obtaining global features. Finally, the SENet module is introduced with the aim of further enhancing the key information in the global features and improving the robustness of the model. The experimental results demonstrate that the method proposed in this paper achieves an average accuracy of 99.63% in a single load scenario and 99.05% in a variable working condition scenario. It exhibits high diagnostic accuracy and a superior capacity for generalization, proves that the KANConv represents a formidable alternative to the existing CNN-based variants for bearing fault diagnosis. |
| format | Article |
| id | doaj-art-c19b3e8b411248ba933da21d941b4578 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-c19b3e8b411248ba933da21d941b45782025-08-20T04:01:35ZengNature PortfolioScientific Reports2045-23222025-07-0115111610.1038/s41598-025-04620-1Bearing fault diagnosis for variable operating conditions based on KAN convolution and dual branch fusion attentionQibing Wang0Chuanjie Yin1Kun She2Qinfeng Tong3Guoxiong Lu4Hongbing Zhang5Jiawei Lu6College of Mechanical and Electrical Engineering, China Jiliang UniversityCollege of Mechanical and Electrical Engineering, China Jiliang UniversityGuangdong Institute of Special Equipment Inspection and ResearchNingbo Hosting Elevator Co., Ltd.Hitachi Elevator (China) Co., Ltd.XIOLIFT Co., Ltd.College of Mechanical and Electrical Engineering, China Jiliang UniversityAbstract This paper proposes a bearing fault diagnosis method based on Kolmogorov–Arnold Convolutional Network: Adaptive Context-aware Graph Channel Attention with Squeeze-and-Excitation Networks (KANConv-ACGCA-SENet). Firstly, a new structure of KANs is applied to Convolutional Neural Networks (CNN) for replacing traditional linear convolutional kernels. Secondly, a dual-branch fusion attention module, comprising the ACGCA modules, is proposed for use in learning fault features. This is achieved by capturing feature differences and utilising non-local(NL) operations, thereby enhancing the feature representation ability under different working conditions. Subsequently, context-aware features and non-local aggregation features are combined with the objective of obtaining global features. Finally, the SENet module is introduced with the aim of further enhancing the key information in the global features and improving the robustness of the model. The experimental results demonstrate that the method proposed in this paper achieves an average accuracy of 99.63% in a single load scenario and 99.05% in a variable working condition scenario. It exhibits high diagnostic accuracy and a superior capacity for generalization, proves that the KANConv represents a formidable alternative to the existing CNN-based variants for bearing fault diagnosis.https://doi.org/10.1038/s41598-025-04620-1ACGCAAttention mechanismBearing fault diagnosisConvolutionKolmogorov–Arnold networks (KANs) |
| spellingShingle | Qibing Wang Chuanjie Yin Kun She Qinfeng Tong Guoxiong Lu Hongbing Zhang Jiawei Lu Bearing fault diagnosis for variable operating conditions based on KAN convolution and dual branch fusion attention Scientific Reports ACGCA Attention mechanism Bearing fault diagnosis Convolution Kolmogorov–Arnold networks (KANs) |
| title | Bearing fault diagnosis for variable operating conditions based on KAN convolution and dual branch fusion attention |
| title_full | Bearing fault diagnosis for variable operating conditions based on KAN convolution and dual branch fusion attention |
| title_fullStr | Bearing fault diagnosis for variable operating conditions based on KAN convolution and dual branch fusion attention |
| title_full_unstemmed | Bearing fault diagnosis for variable operating conditions based on KAN convolution and dual branch fusion attention |
| title_short | Bearing fault diagnosis for variable operating conditions based on KAN convolution and dual branch fusion attention |
| title_sort | bearing fault diagnosis for variable operating conditions based on kan convolution and dual branch fusion attention |
| topic | ACGCA Attention mechanism Bearing fault diagnosis Convolution Kolmogorov–Arnold networks (KANs) |
| url | https://doi.org/10.1038/s41598-025-04620-1 |
| work_keys_str_mv | AT qibingwang bearingfaultdiagnosisforvariableoperatingconditionsbasedonkanconvolutionanddualbranchfusionattention AT chuanjieyin bearingfaultdiagnosisforvariableoperatingconditionsbasedonkanconvolutionanddualbranchfusionattention AT kunshe bearingfaultdiagnosisforvariableoperatingconditionsbasedonkanconvolutionanddualbranchfusionattention AT qinfengtong bearingfaultdiagnosisforvariableoperatingconditionsbasedonkanconvolutionanddualbranchfusionattention AT guoxionglu bearingfaultdiagnosisforvariableoperatingconditionsbasedonkanconvolutionanddualbranchfusionattention AT hongbingzhang bearingfaultdiagnosisforvariableoperatingconditionsbasedonkanconvolutionanddualbranchfusionattention AT jiaweilu bearingfaultdiagnosisforvariableoperatingconditionsbasedonkanconvolutionanddualbranchfusionattention |