Robust bearing fault diagnosis for rotating machinery under noisy and variable operating conditions using a Wavelet-Mamba Network
Dayou Cui, Zhaoyan Xie, Zhixue Wang, Xiaowei Li, Xiangxuan Meng, Zhibin Zhao · 2026
Read Robust bearing fault diagnosis for rotating machinery under noisy and variable operating conditions using a Wavelet-Mamba Network by Dayou Cui, Zhaoyan Xie, Zhixue Wang, Xiaowei Li, Xiangxuan Meng, Zhibin Zhao (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/PMC13618967.txt.
Description
Dayou Cui and five colleagues build a lightweight neural network that diagnoses bearing faults in rotating machinery under noise and changing speeds. Bearings account for roughly half of rotating machinery failures, yet harsh conditions defeat many models. Their design, MDW-Mamba, decouples noise from fault features in stages. A three-channel input, a multi-scale convolution backbone, and a dual-scale wavelet module model long-range patterns efficiently.
Across three public variable-speed datasets, the network beat current best methods and kept average accuracy above eighty-seven percent even when noise exceeded signal. It does this with about a quarter-million parameters and answers in about a millisecond. That is small and fast enough for real-time diagnosis on small edge devices mounted on the machines. The authors position it for wind turbines, aerospace propulsion, and heavy manufacturing, where catching a failing bearing early prevents downtime and loss.
Catalog record
| CREATOR | Dayou Cui, Zhaoyan Xie, Zhixue Wang, Xiaowei Li, Xiangxuan Meng, Zhibin Zhao |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 67 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |