Improving Fault Diagnosis for Train Wheelset Bearings Using a New Model
Author Information
Author(s): Zhao Jing, Li Junfeng, Yuan Zonghao, Mu Tianming, Ma Zengqiang, Liu Suyan
Primary Institution: Shijiazhuang Tiedao University
Hypothesis
Can a transformer-based generative adversarial network improve the diagnosis of wheelset bearing faults?
Conclusion
The TACGAN model significantly enhances the accuracy and effectiveness of fault diagnosis for wheelset bearings.
Supporting Evidence
- The TACGAN model achieved an accuracy of 99% in augmenting wheelset bearing data.
- Generated samples closely resemble real samples, enhancing the dataset's robustness.
- The TACGAN model effectively addresses issues of sample scarcity in fault diagnosis.
Takeaway
This study created a new model that helps find problems in train wheel parts by making better use of data, which is like having more examples to learn from.
Methodology
The study used a transformer-based generative adversarial network to generate fault samples and improve classification accuracy.
Limitations
The datasets used are part of an ongoing study and are not readily available.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website