HFDF-EffNetV2: A Lightweight, Noise-Robust Model for Fault Diagnosis in Rolling Bearings
In rolling bearing intelligent fault diagnosis (FD), lightweight models are constrained by issues such as noise interference and the scarcity of fault data, making it challenging to achieve real-time, high-accuracy diagnosis on resource-limited devices. To address these challenges, this study propos...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-04-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/9/4902 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850155750621446144 |
|---|---|
| author | Donglei Zhang Jiafang Pan Tianping Huang Junlin Niu Faguo Huang |
| author_facet | Donglei Zhang Jiafang Pan Tianping Huang Junlin Niu Faguo Huang |
| author_sort | Donglei Zhang |
| collection | DOAJ |
| description | In rolling bearing intelligent fault diagnosis (FD), lightweight models are constrained by issues such as noise interference and the scarcity of fault data, making it challenging to achieve real-time, high-accuracy diagnosis on resource-limited devices. To address these challenges, this study proposes a lightweight model that combines the hierarchical fine-grained decision fusion (HFDF) strategy with an improved EfficientNetV2 architecture (HFDF-EffNetV2). The model optimizes depth and width multiplicity factors to enhance parameter utilization efficiency. It uses pyramidal convolution (PyConv) combined with Fused-MBConv (Fused-MBPyConv) to obtain multi-scale time-frequency information. Additionally, an enhanced MBConv, termed BSMB-Conv-MLCA, integrates subspace blueprint separable convolution (BSConv-S) with mixed local channel attention (MLCA) extract deep-grained fault features. The HFDF strategy outputs confidence in stages and updates weights to prevent the model from falling into local overfitting when handling confusable samples. Experimental results on Case 1 and Case 2 show that HFDF-EffNetV2 achieved 100% accuracy with diagnostic times of 18.67 millisecond (ms) and 17.56 ms, respectively, and 1.85 million (M) parameters. Under noisy conditions, average accuracies reached 98.19% and 85.68%, respectively. Additionally, the model performed well with small samples, yielding accuracies of 98.69% and 97.51%. These results highlight its superior robustness to noise and lightweight performance compared with other advanced models. |
| format | Article |
| id | doaj-art-0b7c8a2803ed444ba19e674b94266def |
| institution | OA Journals |
| issn | 2076-3417 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-0b7c8a2803ed444ba19e674b94266def2025-08-20T02:24:47ZengMDPI AGApplied Sciences2076-34172025-04-01159490210.3390/app15094902HFDF-EffNetV2: A Lightweight, Noise-Robust Model for Fault Diagnosis in Rolling BearingsDonglei Zhang0Jiafang Pan1Tianping Huang2Junlin Niu3Faguo Huang4Key Laboratory of Advanced Manufacturing and Automation Technology (Guilin University of Technology), Education Department of Guangxi Zhuang Autonomous Region, Guilin 541006, ChinaKey Laboratory of Advanced Manufacturing and Automation Technology (Guilin University of Technology), Education Department of Guangxi Zhuang Autonomous Region, Guilin 541006, ChinaKey Laboratory of Advanced Manufacturing and Automation Technology (Guilin University of Technology), Education Department of Guangxi Zhuang Autonomous Region, Guilin 541006, ChinaKey Laboratory of Advanced Manufacturing and Automation Technology (Guilin University of Technology), Education Department of Guangxi Zhuang Autonomous Region, Guilin 541006, ChinaKey Laboratory of Advanced Manufacturing and Automation Technology (Guilin University of Technology), Education Department of Guangxi Zhuang Autonomous Region, Guilin 541006, ChinaIn rolling bearing intelligent fault diagnosis (FD), lightweight models are constrained by issues such as noise interference and the scarcity of fault data, making it challenging to achieve real-time, high-accuracy diagnosis on resource-limited devices. To address these challenges, this study proposes a lightweight model that combines the hierarchical fine-grained decision fusion (HFDF) strategy with an improved EfficientNetV2 architecture (HFDF-EffNetV2). The model optimizes depth and width multiplicity factors to enhance parameter utilization efficiency. It uses pyramidal convolution (PyConv) combined with Fused-MBConv (Fused-MBPyConv) to obtain multi-scale time-frequency information. Additionally, an enhanced MBConv, termed BSMB-Conv-MLCA, integrates subspace blueprint separable convolution (BSConv-S) with mixed local channel attention (MLCA) extract deep-grained fault features. The HFDF strategy outputs confidence in stages and updates weights to prevent the model from falling into local overfitting when handling confusable samples. Experimental results on Case 1 and Case 2 show that HFDF-EffNetV2 achieved 100% accuracy with diagnostic times of 18.67 millisecond (ms) and 17.56 ms, respectively, and 1.85 million (M) parameters. Under noisy conditions, average accuracies reached 98.19% and 85.68%, respectively. Additionally, the model performed well with small samples, yielding accuracies of 98.69% and 97.51%. These results highlight its superior robustness to noise and lightweight performance compared with other advanced models.https://www.mdpi.com/2076-3417/15/9/4902fault diagnosislightweightEfficientNetV2robustness |
| spellingShingle | Donglei Zhang Jiafang Pan Tianping Huang Junlin Niu Faguo Huang HFDF-EffNetV2: A Lightweight, Noise-Robust Model for Fault Diagnosis in Rolling Bearings Applied Sciences fault diagnosis lightweight EfficientNetV2 robustness |
| title | HFDF-EffNetV2: A Lightweight, Noise-Robust Model for Fault Diagnosis in Rolling Bearings |
| title_full | HFDF-EffNetV2: A Lightweight, Noise-Robust Model for Fault Diagnosis in Rolling Bearings |
| title_fullStr | HFDF-EffNetV2: A Lightweight, Noise-Robust Model for Fault Diagnosis in Rolling Bearings |
| title_full_unstemmed | HFDF-EffNetV2: A Lightweight, Noise-Robust Model for Fault Diagnosis in Rolling Bearings |
| title_short | HFDF-EffNetV2: A Lightweight, Noise-Robust Model for Fault Diagnosis in Rolling Bearings |
| title_sort | hfdf effnetv2 a lightweight noise robust model for fault diagnosis in rolling bearings |
| topic | fault diagnosis lightweight EfficientNetV2 robustness |
| url | https://www.mdpi.com/2076-3417/15/9/4902 |
| work_keys_str_mv | AT dongleizhang hfdfeffnetv2alightweightnoiserobustmodelforfaultdiagnosisinrollingbearings AT jiafangpan hfdfeffnetv2alightweightnoiserobustmodelforfaultdiagnosisinrollingbearings AT tianpinghuang hfdfeffnetv2alightweightnoiserobustmodelforfaultdiagnosisinrollingbearings AT junlinniu hfdfeffnetv2alightweightnoiserobustmodelforfaultdiagnosisinrollingbearings AT faguohuang hfdfeffnetv2alightweightnoiserobustmodelforfaultdiagnosisinrollingbearings |