Research on bearing fault diagnosis based on sparse adaptive S-transform and deep residual network
- Publication Type:
- Journal Article
- Citation:
- Dianji yu Kongzhi Xuebao/Electric Machines and Control, 2022, 26, (8), pp. 112-119
- Issue Date:
- 2022-08-01
Closed Access
| Filename | Description | Size | |||
|---|---|---|---|---|---|
| Research on bearing fault diagnosis based on sparse adaptive S-transform and deep residual network.pdf | Published version | 693.83 kB |
Copyright Clearance Process
- Recently Added
- In Progress
- Closed Access
This item is closed access and not available.
The vibration signals of complex rolling bearings are nonlinear and non-stationary, and the traditional signal processing methods are difficult to achieve effective extraction of fault features and high-precision fault classification. To address this problem, a bearing fault diagnosis method based on sparse adaptive fault diagnosis S transform and deep residual network is proposed considering of the time-frequency characteristics of bearing vibration signals. Firstly, sparse adaptive S transform was applied to the collected vibration signals to obtain the time-frequency image characteristics of bearings under different working conditions. Then, the structure of deep residual network was constructed, and network parameters such as optimizer and initial learning rate were reasonably selected,and a bearing fault diagnosis model based on deep residual network was proposed. The calculation results of a rolling bearing vibration data sets show that time-frequency analysis based on sparse adaptive S transform method has a high time-frequency resolution, and the construction of the depth of the residual network model can accurately identify under different fault conditions and severity of bearing fault state.
Please use this identifier to cite or link to this item: