Cross-Condition Fault Diagnosis of Rolling Bearings Based on Time–Frequency Ridge Extraction and Uncertainty-Guided Distribution-Regularised Convolutional Wasserstein Autoencoder
Junshen Zhang, Jixing Yang, Xinfa Shi, Qing Zhang · 2026
Read Cross-Condition Fault Diagnosis of Rolling Bearings Based on Time–Frequency Ridge Extraction and Uncertainty-Guided Distribution-Regularised Convolutional Wasserstein Autoencoder by Junshen Zhang, Jixing Yang, Xinfa Shi, Qing Zhang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13568277.txt.
Description
Junshen Zhang and co-authors build a fault finder for rolling bearings that works across changing speeds and loads, in a 2026 paper. Bearing wear is a top cause of machine failure. Vibration signals shift shape as speed and load change, so a tool trained in one state can fail in the next. Their method needs no speed sensor at all.
First, the tool reads the spin rate straight from the vibration trace by tracking its ridges over time. It then cuts the signal into single-turn chunks and evens out their scale. A special autoencoder learns the fault signs while a weighting step balances its goals on its own. Tested on two bearing datasets from Ottawa and a test rig, the method sorted faults with accuracy near 99 percent in both, even at spin rates it had never seen.
Catalog record
| CREATOR | Junshen Zhang, Jixing Yang, Xinfa Shi, Qing Zhang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 64 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |