A comprehensive review of deep learning-based fault diagnosis approaches for rolling bearings: Advancements and challenges DOI Creative Commons

Jingxin ZHAO,

Wenming Wang, Ji Huang

et al.

AIP Advances, Journal Year: 2025, Volume and Issue: 15(2)

Published: Feb. 1, 2025

Rolling bearing fault diagnosis is an important technology for health monitoring and pre-maintenance of mechanical equipment, which great significance improving equipment operation reliability reducing maintenance costs. This article reviews the research progress methods rolling bearings, with a focus on analyzing applications, advantages, disadvantages traditional data-driven methods, deep learning graph embedding Transformer in this field. In addition, further analysis was conducted main issues current research, including complex network structures, insufficient information attention, difficulties data processing, challenges long-term dependency modeling. response to these challenges, future should designing more lightweight efficient models, computational efficiency, robustness strengthening attention mining features.

Language: Английский

A comprehensive review of deep learning-based fault diagnosis approaches for rolling bearings: Advancements and challenges DOI Creative Commons

Jingxin ZHAO,

Wenming Wang, Ji Huang

et al.

AIP Advances, Journal Year: 2025, Volume and Issue: 15(2)

Published: Feb. 1, 2025

Rolling bearing fault diagnosis is an important technology for health monitoring and pre-maintenance of mechanical equipment, which great significance improving equipment operation reliability reducing maintenance costs. This article reviews the research progress methods rolling bearings, with a focus on analyzing applications, advantages, disadvantages traditional data-driven methods, deep learning graph embedding Transformer in this field. In addition, further analysis was conducted main issues current research, including complex network structures, insufficient information attention, difficulties data processing, challenges long-term dependency modeling. response to these challenges, future should designing more lightweight efficient models, computational efficiency, robustness strengthening attention mining features.

Language: Английский

Citations

0