A Lightweight Fault Diagnosis with Domain Adaptation for Defected Bearings DOI Creative Commons

Xiaohong Jiao,

Yongsheng Zhou,

Xuan Liu

et al.

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(8), P. 4579 - 4579

Published: April 21, 2025

This paper presents a lightweight fault diagnosis framework for bearing defects, integrating time-frequency analysis, deep learning, and model compression techniques to address challenges in resource-constrained environments. The proposed method combines the S-transform high-resolution representation with MobileNet as an efficient backbone network, enabling robust feature extraction from complex vibration signals. To enhance deployment on edge devices, knowledge distillation is employed compress model, significantly reducing computational complexity while maintaining diagnostic accuracy. Additionally, domain adaptation considered mitigate shift issues, ensuring performance across varying operating conditions. Experimental results demonstrate framework’s effectiveness, achieving high accuracy reduced overhead, making it practical solution real-time industrial applications. approach bridges gap between advanced learning requirements, offering scalable diagnosis.

Language: Английский

A Lightweight Fault Diagnosis with Domain Adaptation for Defected Bearings DOI Creative Commons

Xiaohong Jiao,

Yongsheng Zhou,

Xuan Liu

et al.

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(8), P. 4579 - 4579

Published: April 21, 2025

This paper presents a lightweight fault diagnosis framework for bearing defects, integrating time-frequency analysis, deep learning, and model compression techniques to address challenges in resource-constrained environments. The proposed method combines the S-transform high-resolution representation with MobileNet as an efficient backbone network, enabling robust feature extraction from complex vibration signals. To enhance deployment on edge devices, knowledge distillation is employed compress model, significantly reducing computational complexity while maintaining diagnostic accuracy. Additionally, domain adaptation considered mitigate shift issues, ensuring performance across varying operating conditions. Experimental results demonstrate framework’s effectiveness, achieving high accuracy reduced overhead, making it practical solution real-time industrial applications. approach bridges gap between advanced learning requirements, offering scalable diagnosis.

Language: Английский

Citations

0