Rolling Based on Multi-Source Time–Frequency Feature Fusion with a Wavelet-Convolution, Channel-Attention-Residual Network-Bearing Fault Diagnosis Method
As a core component of rotating machinery, the condition of rolling bearings is directly related to the reliability and safety of equipment operation; therefore, the accurate and reliable monitoring of bearing operating status is critical. However, when dealing with non-stationary and noisy vibratio...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/13/4091 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839631518313480192 |
|---|---|
| author | Tongshuhao Feng Zhuoran Wang Lipeng Qiu Hongkun Li Zhen Wang |
| author_facet | Tongshuhao Feng Zhuoran Wang Lipeng Qiu Hongkun Li Zhen Wang |
| author_sort | Tongshuhao Feng |
| collection | DOAJ |
| description | As a core component of rotating machinery, the condition of rolling bearings is directly related to the reliability and safety of equipment operation; therefore, the accurate and reliable monitoring of bearing operating status is critical. However, when dealing with non-stationary and noisy vibration signals, traditional fault diagnosis methods are often constrained by limited feature characterization from single time–frequency analysis and inadequate feature extraction capabilities. To address this issue, this study proposes a lightweight fault diagnosis model (WaveCAResNet) enhanced with multi-source time–frequency features. By fusing complementary time–frequency features derived from continuous wavelet transform, short-time Fourier transform, Hilbert–Huang transform, and Wigner–Ville distribution, the capability to characterize complex fault patterns is significantly improved. Meanwhile, an efficient and lightweight deep learning model (WaveCAResNet) is constructed based on residual networks by integrating multi-scale analysis via a wavelet convolutional layer (WTConv) with the dynamic feature optimization properties of channel-attention-weighted residuals (CAWRs) and the efficient temporal modeling capabilities of weighted residual efficient multi-scale attention (WREMA). Experimental validation indicates that the proposed method achieves higher diagnostic accuracy and robustness than existing mainstream models on typical bearing datasets, and the classification performance of the newly proposed model exceeds that of state-of-the-art bearing fault diagnostic models on the same dataset, even under noisy conditions. |
| format | Article |
| id | doaj-art-6e5c0c76a1b446978ddf076e7c3bcfb7 |
| institution | Matheson Library |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-6e5c0c76a1b446978ddf076e7c3bcfb72025-07-11T14:43:21ZengMDPI AGSensors1424-82202025-06-012513409110.3390/s25134091Rolling Based on Multi-Source Time–Frequency Feature Fusion with a Wavelet-Convolution, Channel-Attention-Residual Network-Bearing Fault Diagnosis MethodTongshuhao Feng0Zhuoran Wang1Lipeng Qiu2Hongkun Li3Zhen Wang4School of Mechanical Engineering, Dalian University, Dalian 116622, ChinaSchool of Instrument Science and Engineering, Southeast University, Nanjing 210096, ChinaSchool of Mechanical Engineering, Dalian University, Dalian 116622, ChinaSchool of Mechanical Engineering, Dalian University of Technology, Dalian 116024, ChinaSchool of Mechanical Engineering, Dalian University, Dalian 116622, ChinaAs a core component of rotating machinery, the condition of rolling bearings is directly related to the reliability and safety of equipment operation; therefore, the accurate and reliable monitoring of bearing operating status is critical. However, when dealing with non-stationary and noisy vibration signals, traditional fault diagnosis methods are often constrained by limited feature characterization from single time–frequency analysis and inadequate feature extraction capabilities. To address this issue, this study proposes a lightweight fault diagnosis model (WaveCAResNet) enhanced with multi-source time–frequency features. By fusing complementary time–frequency features derived from continuous wavelet transform, short-time Fourier transform, Hilbert–Huang transform, and Wigner–Ville distribution, the capability to characterize complex fault patterns is significantly improved. Meanwhile, an efficient and lightweight deep learning model (WaveCAResNet) is constructed based on residual networks by integrating multi-scale analysis via a wavelet convolutional layer (WTConv) with the dynamic feature optimization properties of channel-attention-weighted residuals (CAWRs) and the efficient temporal modeling capabilities of weighted residual efficient multi-scale attention (WREMA). Experimental validation indicates that the proposed method achieves higher diagnostic accuracy and robustness than existing mainstream models on typical bearing datasets, and the classification performance of the newly proposed model exceeds that of state-of-the-art bearing fault diagnostic models on the same dataset, even under noisy conditions.https://www.mdpi.com/1424-8220/25/13/4091rolling bearingfault diagnosistime–frequency diagramwavelet convolution channel attention residual networkattention |
| spellingShingle | Tongshuhao Feng Zhuoran Wang Lipeng Qiu Hongkun Li Zhen Wang Rolling Based on Multi-Source Time–Frequency Feature Fusion with a Wavelet-Convolution, Channel-Attention-Residual Network-Bearing Fault Diagnosis Method Sensors rolling bearing fault diagnosis time–frequency diagram wavelet convolution channel attention residual network attention |
| title | Rolling Based on Multi-Source Time–Frequency Feature Fusion with a Wavelet-Convolution, Channel-Attention-Residual Network-Bearing Fault Diagnosis Method |
| title_full | Rolling Based on Multi-Source Time–Frequency Feature Fusion with a Wavelet-Convolution, Channel-Attention-Residual Network-Bearing Fault Diagnosis Method |
| title_fullStr | Rolling Based on Multi-Source Time–Frequency Feature Fusion with a Wavelet-Convolution, Channel-Attention-Residual Network-Bearing Fault Diagnosis Method |
| title_full_unstemmed | Rolling Based on Multi-Source Time–Frequency Feature Fusion with a Wavelet-Convolution, Channel-Attention-Residual Network-Bearing Fault Diagnosis Method |
| title_short | Rolling Based on Multi-Source Time–Frequency Feature Fusion with a Wavelet-Convolution, Channel-Attention-Residual Network-Bearing Fault Diagnosis Method |
| title_sort | rolling based on multi source time frequency feature fusion with a wavelet convolution channel attention residual network bearing fault diagnosis method |
| topic | rolling bearing fault diagnosis time–frequency diagram wavelet convolution channel attention residual network attention |
| url | https://www.mdpi.com/1424-8220/25/13/4091 |
| work_keys_str_mv | AT tongshuhaofeng rollingbasedonmultisourcetimefrequencyfeaturefusionwithawaveletconvolutionchannelattentionresidualnetworkbearingfaultdiagnosismethod AT zhuoranwang rollingbasedonmultisourcetimefrequencyfeaturefusionwithawaveletconvolutionchannelattentionresidualnetworkbearingfaultdiagnosismethod AT lipengqiu rollingbasedonmultisourcetimefrequencyfeaturefusionwithawaveletconvolutionchannelattentionresidualnetworkbearingfaultdiagnosismethod AT hongkunli rollingbasedonmultisourcetimefrequencyfeaturefusionwithawaveletconvolutionchannelattentionresidualnetworkbearingfaultdiagnosismethod AT zhenwang rollingbasedonmultisourcetimefrequencyfeaturefusionwithawaveletconvolutionchannelattentionresidualnetworkbearingfaultdiagnosismethod |