A5044127094- Machine Fault Diagnosis Techniques
- Gear and Bearing Dynamics Analysis
- Engineering Diagnostics and Reliability
- Structural Health Monitoring Techniques
- Risk and Safety Analysis
- Railway Systems and Energy Efficiency
- Mechanical Failure Analysis and Simulation
- Power System Reliability and Maintenance
- Ultrasonics and Acoustic Wave Propagation
- Fault Detection and Control Systems
- Non-Destructive Testing Techniques
Shijiazhuang Tiedao University
2022-2024
Hebei Normal University
2024
The working condition of high-speed train wheelset bearings is complex and bad, which makes its vibration signal contain strong noise. To obtain more information samples, signals are constructed into graphs fit by graph convolution network (GCN). existing GCN-based models still have some problems, such as insufficient theoretical basis the unstable overfitting during training models, significantly limits diagnosis accuracy. address these issues, first, a weighted <inline-formula...
Abstract Wheelset bearings are a core component of high-speed trains, and their fault diagnosis is the key to smooth operation. Deep learning widely used in due its powerful classification ability. To explicitly fit features vibration signals further explore relationship between signals, graph attention network (GAT) becoming focus research. Unlike traditional neural networks, GATs can on edges with stronger correlations vertices, making model more when fitting samples non-Euclidean space....
Abstract Fault diagnosis of rolling bearings is key to maintain and repair modern rotating machinery. Rolling are usually working in non-stationary conditions with time-varying loads speeds. Existing methods based on vibration signals only do not have the ability adapt rotational speed. And when load changes, their accuracy rate will be obviously reduced. A method put forward which fuses multi-modal sensor fit speed information. Firstly, features extracted from raw instantaneous signals,...
Wheelset bearings fault samples of high-speed trains often suffer from insufficient numbers and missing points. However, working conditions are one the most important factors affecting data augmentation repair, which is not sufficiently emphasized by existing methods. Also, generative adversarial networks (GANs) as effective sample generation methods still have many problems, such model collapse training difficulties. To solve these a repair method for train wheelset bearing diagnosis with...
Most fault diagnosis methods require that the source and target machines' classes should overlap number of samples be comparable. However, such assumptions are unrealistic in wheelset-bearing diagnosis. The high reliability wheelset bearings difficulty collecting signals make datasets insufficient. obtained by laboratories can rich samples, types faults, working conditions, which contain a wealth information. Therefore, graph attention-based multichannel transfer learning network (GAMTLN) is...
Diagnosing faults in wheelset bearings is critical for train safety. The main challenge that only a limited amount of fault sample data can be obtained during high-speed operations. This scarcity samples impacts the training and accuracy deep learning models bearing diagnosis. Studies show Auxiliary Classifier Generative Adversarial Network (ACGAN) demonstrates promising performance addressing this issue. However, existing ACGAN have drawbacks such as complexity, high computational expenses,...