Research on fault diagnosis of rolling bearing based on MTA-LNN
To address the issue of rolling bearing vibration signals being obscured by noise, leading to unclear fault characteristics and low fault diagnosis accuracy, a rolling bearing fault diagnosis method based on the Multi-scale Temporal Attention Liquid Neural Network (MTA-LNN) optimized by multi-scale temporal attention mechanism is proposed. Firstly, the original vibration signals are constructed into multi-scale time series samples through different scale windows. Secondly, the temporal attention mechanism is introduced to perform adaptive weighting on fault signal features at different scales. Finally, the liquid neural network is introduced to construct a fault diagnosis model to achieve the fault diagnosis task. Experimental results show that the liquid neural network optimized by the multi-scale attention mechanism can still achieve a high fault diagnosis rate under complex working conditions, with a fault diagnosis accuracy of up to 98.99%.
Authors
- Y. B. Liu
- Weinan Zheng (ORCID: https://orcid.org/0009-0007-2215-2610)
- Yuehui Wang (ORCID: https://orcid.org/0000-0001-9379-0360)
- Ying Du
- Jian Jin
- Chunyu Liu
- Junliang Sha
- Miao Yu
Institutions
- Jilin University of Architecture and Technology (CN)
Publication Details
- Journal
- Scientific Reports
- Published
- 2026-11-15
- DOI
- https://doi.org/10.1038/s41598-026-79950-9
- Primary Topic
- Machine Fault Diagnosis Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00