Noise-Tolerant Bearing Fault Diagnosis Using Continuous Wavelet Transform-Enhanced Swin Transformer

DOI

10.2991/978-94-6239-616-6_8How to use a DOI?

Keywords

Swin Transformer; Lion Optimizer; Continuous Wavelet Transform (CWT); Bearing Fault Diagnosis; Deep Learning; Time–Frequency Analysis; Transformer-based Classification; Noise-Tolerant Diagnosis; Industrial Condition Monitoring; Data Mining

Abstract

Bearings are essential components in industrial machinery, facilitating smooth rotational motion, minimizing friction, and supporting substantial mechanical loads. In scraper conveyor systems, these bearings are exposed to extremely harsh environments involving high-impact loads, intense noise interference, fluctuating speeds, and prolonged operational hours. Such conditions significantly distort vibration signals, making accurate fault diagnosis a complex challenge. Traditional diagnostic methods heavily rely on hand-crafted features and often lack the flexibility to adapt to the intricate temporal and spectral variations present in real-world mining conditions. Consequently, they struggle with noise resilience and fail to generalize effectively across varying operational scenarios. To overcome these limitations, this proposed system introduces a deep learning-based fault diagnosis framework that employs Continuous Wavelet Transform (CWT) to generate detailed time–frequency representations of vibration signals. A Swin Transformer is then used to extract both local and global features from these representations, while the Lion optimizer ensures faster and more stable model convergence. This hybrid architecture addresses key issues such as noise interference, robust feature learning, and adaptability, making it a scalable and efficient solution for real-time bearing fault diagnosis in mining applications.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - R. Raju
AU  - S. G. Sreenidhi
AU  - M. Swathika
AU  - V. Oviya
PY  - 2026
DA  - 2026/03/31
TI  - Noise-Tolerant Bearing Fault Diagnosis Using Continuous Wavelet Transform-Enhanced Swin Transformer
BT  - Proceedings of the International Conference on Artificial Intelligence and Secure Data Analytics (ICAISDA 2025)
PB  - Atlantis Press
SP  - 97
EP  - 109
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-616-6_8
DO  - 10.2991/978-94-6239-616-6_8
ID  - Raju2026
ER  -