Weak fault feature extraction and remaining useful life prediction for diagnostics and prognostics of rolling element bearings
- Publication Type:
- Thesis
- Issue Date:
- 2023
Open Access
Copyright Clearance Process
- Recently Added
- In Progress
- Open Access
This item is open access.
Rolling element bearings are extensively utilized in modern industry. In practice, the degradation and failure of rolling element bearings are inevitable, and can result in enormous costs due to productivity losses, maintenance, or fatal accidents. Therefore, this thesis aims to develop a comprehensive condition monitoring framework for rolling element bearing fault diagnosis and prognosis.
In real industrial applications, the collected rolling element bearing signals are usually multi-components and surrounded by intensive noise. Therefore, the collected fault-related information is usually weak, resulting in difficulty in fault diagnosis. To address it, two improved rolling element bearing diagnostic methodologies are proposed to enhance the weak fault information. In addition, the interferences from adjacent components or the external environment and spurious fluctuations can also bring negative effects on constructing health indicator (HI) to depict the degradation characteristics, thus decreasing the prediction accuracy. To this end, two prognostic methodologies based on two novel HIs are established for accurate remaining useful life prediction.
The proposed methodologies are validated using different datasets. Successful diagnostic and prognostic results demonstrate that the proposed methodologies can generate compelling and innovative tools for rolling element bearing condition monitoring under complicated operating conditions, providing reliable maintenance decisions for modern industry.
Please use this identifier to cite or link to this item: