A Deep Learning-Based Framework for Bearing RUL Prediction to Optimize Laser Shock Peening Remanufacturing DOI Creative Commons
Yuchen Liang, Yuqi Wang, An‐Ping Li

и другие.

Applied Sciences, Год журнала: 2024, Номер 14(22), С. 10493 - 10493

Опубликована: Ноя. 14, 2024

Accurate prediction of the remaining useful life (RUL) bearings is crucial for maintaining reliability and efficiency industrial systems. This study introduces a novel methodology integrating advanced machine learning optimization techniques to address this challenge. (1) A transformer-attention model was developed process segmented vibration signals, effectively capturing complex patterns. The showed better performance than traditional approaches, with an RMSE 0.989. (2) Deep Neural Network (DNN) designed predict extended RUL after laser shock peening (LSP) remanufacturing. fruit fly (FFO) algorithm employed optimize remanufacturing parameters; 29.33% improvement achieved in fitness compared baseline. (3) DNN predictions were validated against Finite Element Analysis (FEA) simulations, low relative error 2.5% 5.8%; good accuracy effects optimized LSP parameters on bearing extension.

Язык: Английский

Prediction of bearing remaining useful life based on a two-stage updated digital twin DOI
Deqiang He, Jiayang Zhao, Zhenzhen Jin

и другие.

Advanced Engineering Informatics, Год журнала: 2025, Номер 65, С. 103123 - 103123

Опубликована: Янв. 13, 2025

Язык: Английский

Процитировано

20
Multi-scale dynamic graph mutual information network for planet bearing health monitoring under imbalanced data DOI

Wenbin Cai,

Dezun Zhao, Tianyang Wang

и другие.

Advanced Engineering Informatics, Год журнала: 2025, Номер 64, С. 103096 - 103096

Опубликована: Янв. 5, 2025

Язык: Английский

Процитировано

8
DCAGGCN: A novel method for remaining useful life prediction of bearings DOI
Deqiang He, Jiayang Zhao, Zhenzhen Jin

и другие.

Reliability Engineering & System Safety, Год журнала: 2025, Номер unknown, С. 110978 - 110978

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

3
Dynamic Model-driven Dictionary Learning-inspired Domain Adaptation Strategy for Cross-domain Bearing Fault Diagnosis DOI
Zhengyu Du, Dongdong Liu, Lingli Cui

и другие.

Reliability Engineering & System Safety, Год журнала: 2025, Номер unknown, С. 110905 - 110905

Опубликована: Фев. 1, 2025

Язык: Английский

Процитировано

1
Improved sand cat swarm optimization algorithm assisted GraphSAGE-GRU for remaining useful life of engine DOI Creative Commons
Yongliang Yuan, Ruifang Li, Guohu Wang

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Фев. 26, 2025

Abstract With the development of deep learning, potential for its use in remaining useful life (RUL) has substantially increased recent years due to powerful data processing capabilities. However, relationships and interdependencies operation parameters non-Euclidean space are ignored utilizing current learning-based methods during degradation process engine. To address this challenge, an improved sand cat swarm optimization-assisted Graph SAmple aggregate gate recurrent unit (ISCSO-GraphSage-GRU) is proposed achieve RUL prediction work. Firstly, maximum information coefficient (MIC) utilized describing interdependent relations measured parameters. Building on foundation, constructed graph used as input GraphSage-GRU so overcoming shortcomings existing learning methods. Additionally, work optimization (ISCSO) improve predicted performance GraphSage-GRU, including tent mapping population initialization a novel adaptive approach enhance exploration exploitation optimization. The CMAPSS dataset validate effectiveness advancedness ISCSO-GraphSage-GRU, experimental results show that R 2 ISCSO-GraphSage-GRU greater than 0.99, RMSE less 6.

Язык: Английский

Процитировано

0
Dual graph driven-consistent representation learning method for semi-supervised fault diagnosis of rotating machinery DOI
Zhichao Jiang, Dongdong Liu, Huaqing Wang

и другие.

Advanced Engineering Informatics, Год журнала: 2025, Номер 65, С. 103274 - 103274

Опубликована: Март 22, 2025

Язык: Английский

Процитировано

0
A Deep Learning-Based Framework for Bearing RUL Prediction to Optimize Laser Shock Peening Remanufacturing DOI Creative Commons
Yuchen Liang, Yuqi Wang, An‐Ping Li

и другие.

Applied Sciences, Год журнала: 2024, Номер 14(22), С. 10493 - 10493

Опубликована: Ноя. 14, 2024

Accurate prediction of the remaining useful life (RUL) bearings is crucial for maintaining reliability and efficiency industrial systems. This study introduces a novel methodology integrating advanced machine learning optimization techniques to address this challenge. (1) A transformer-attention model was developed process segmented vibration signals, effectively capturing complex patterns. The showed better performance than traditional approaches, with an RMSE 0.989. (2) Deep Neural Network (DNN) designed predict extended RUL after laser shock peening (LSP) remanufacturing. fruit fly (FFO) algorithm employed optimize remanufacturing parameters; 29.33% improvement achieved in fitness compared baseline. (3) DNN predictions were validated against Finite Element Analysis (FEA) simulations, low relative error 2.5% 5.8%; good accuracy effects optimized LSP parameters on bearing extension.

Язык: Английский

Процитировано

0