Intelligent Diagnosis Methods for Initial Faults in Rotor-Bearing Systems
1Shenyang Jianzhu University, Shenyang, China
2Northeastern University, Shenyang, China
3University of Science and Technology of China, Hefei, China
4Southwest Jiaotong University, Chengdu, China
5Hong Kong Polytechnic University, Hong Kong
Intelligent Diagnosis Methods for Initial Faults in Rotor-Bearing Systems
Description
Rotor-bearing systems are widely applied in various industrial domains, such as mechanical manufacturing, aerospace, and turbomachinery. Usually, rotor-bearing systems act as the key components in the mechanisms, and their characteristics directly affect the performance of the corresponding machines. Generally, faults occurring in rotor-bearing systems lead to performance degradation or even failure, and so in order to prevent the failures of these rotor-bearing systems, it is important to develop methods to detect the initial faults in a timely manner.
For traditional model-driven and data-driven diagnosis methods, the fault labels are calculated or picked from signals and are compared with the working data. However, the fault characteristics of rotor-bearing systems usually vary with working conditions, and the studied cases are not universal enough to cover all working conditions. Additionally, the features of the initial faults are buried in the stochastic noise and are therefore unable to be matched with database information. To overcome these problems, intelligent modelling methods should be proposed to make the features more adaptive, and intelligent feature recognition methods, such as deep learning- based methods, need to be developed to provide a more accurate identification of initial faults.
The aim of this Special Issue is to promote the accuracy of identification of initial faults in rotor systems and provide a communication platform for new ideas for intelligent diagnosis methods. We warmly welcome high-quality original research articles presenting new methods, and review articles stating recent developments and future directions.
Potential topics include but are not limited to the following:
- Dynamic modelling of rotor-bearing systems with initial faults
- Fault feature extraction and signal processing
- Intelligent diagnosis and signal denoising
- Data-driven diagnosis based on deep learning
- Stress analysis and prognosis for rotor-bearing systems
- Multi-source feature fusion diagnosis methods
- Fault extension and degradation analysis in rotary mechanisms
- Fault localisation and grade evaluation methods
- Dynamic responses with coupled faults in both rotors and bearings
- Condition monitoring with intelligent observers