Feature Enhancement via Linear Transformation and Its Application in Fault Diagnosis DOI
Biao He, Quan Qian, Yi Qin

et al.

IEEE Internet of Things Journal, Journal Year: 2024, Volume and Issue: 11(12), P. 21895 - 21903

Published: March 18, 2024

The performance of neural networks is directly affected by the features obtained from backbones fault diagnosis networks. To obtain clear and improve networks, this paper constructs a new block based on linear transformation. Firstly, feature vector divided into decisive component an invalid component. Then, it worth noting that orthogonality these two components beneficial to model learning. According this, are extracted using spaces constructed relationships between four fundamental sub-spaces matrix. In sub-spaces, row space null employed extract useless component, respectively. Both implemented layers designed as encoder-decoder structure ensure existence space. spaces, constraint term proposed modify their weights. Lastly, cosine similarity input entirely. When incorporating some classic classifying they can achieve improved accuracy. Moreover, when comparing conventional spatial attention mechanisms, module demonstrates superior overall performance, including accuracy, antinoise ability, generalization ability.

Language: Английский

Multi-sensor data fusion-enabled lightweight convolutional double regularization contrast transformer for aerospace bearing small samples fault diagnosis DOI
Yutong Dong, Hongkai Jiang, Mingzhe Mu

et al.

Advanced Engineering Informatics, Journal Year: 2024, Volume and Issue: 62, P. 102573 - 102573

Published: May 2, 2024

Language: Английский

Citations

30

FEV-Swin: Multi-source heterogeneous information fusion under a variant swin transformer framework for intelligent cross-domain fault diagnosis DOI
Keyi Zhou, Keyi Zhou, Bin Jiang

et al.

Knowledge-Based Systems, Journal Year: 2025, Volume and Issue: unknown, P. 112982 - 112982

Published: Jan. 1, 2025

Language: Английский

Citations

5

Time-frequency supervised contrastive learning via pseudo-labeling: An unsupervised domain adaptation network for rolling bearing fault diagnosis under time-varying speeds DOI
Bin Pang,

Qiuhai Liu,

Zhenduo Sun

et al.

Advanced Engineering Informatics, Journal Year: 2023, Volume and Issue: 59, P. 102304 - 102304

Published: Dec. 11, 2023

Language: Английский

Citations

44

An Intelligent Deep Learning Framework for Traffic Flow Imputation and Short-term Prediction Based on Dynamic Features DOI
Xianhui Zong, Yong Qi, He Yan

et al.

Knowledge-Based Systems, Journal Year: 2024, Volume and Issue: 300, P. 112178 - 112178

Published: June 25, 2024

Language: Английский

Citations

9

A digital twin system for centrifugal pump fault diagnosis driven by transfer learning based on graph convolutional neural networks DOI
Zifeng Xu, Zhe Wang,

Chaojia Gao

et al.

Computers in Industry, Journal Year: 2024, Volume and Issue: 163, P. 104155 - 104155

Published: Aug. 30, 2024

Language: Английский

Citations

7

MGTN-DSI: A multi-sensor graph transfer network considering dual structural information for fault diagnosis under varying working conditions DOI

Jianjie Liu,

Xianfeng Yuan, Xilin Yang

et al.

Advanced Engineering Informatics, Journal Year: 2025, Volume and Issue: 65, P. 103119 - 103119

Published: Jan. 15, 2025

Language: Английский

Citations

1

Bi-structural spatial–temporal network for few-shot fault diagnosis of rotating machinery DOI
Zixu Chen, Jinchen Ji, Qing Ni

et al.

Mechanical Systems and Signal Processing, Journal Year: 2025, Volume and Issue: 227, P. 112378 - 112378

Published: Jan. 28, 2025

Language: Английский

Citations

1

DyGAT-FTNet: A Dynamic Graph Attention Network for Multi-Sensor Fault Diagnosis and Time–Frequency Data Fusion DOI Creative Commons
Hongjun Duan, Guorong Chen, Yuan Yu

et al.

Sensors, Journal Year: 2025, Volume and Issue: 25(3), P. 810 - 810

Published: Jan. 29, 2025

Fault diagnosis in modern industrial and information systems is critical for ensuring equipment reliability operational safety, but traditional methods have difficulty effectively capturing spatiotemporal dependencies fault-sensitive features multi-sensor data, especially rarely considering dynamic between data. To address these challenges, this study proposes DyGAT-FTNet, a novel graph neural network model tailored to fault detection. The dynamically constructs association graphs through learnable construction mechanism, enabling automatic adjacency matrix generation based on time–frequency derived from the short-time Fourier transform (STFT). Additionally, attention (DyGAT) enhances extraction of by assigning node weights. pooling layer further aggregates optimizes feature representation.Experimental evaluations two benchmark detection datasets, XJTUSuprgear dataset SEU dataset, show that DyGAT-FTNet significantly outperformed existing classification accuracy, with accuracies 1.0000 0.9995, respectively, highlighting its potential practical applications.

Language: Английский

Citations

1

SCG-GFFE: A Self-Constructed graph fault feature extractor based on graph Auto-encoder algorithm for unlabeled single-variable vibration signals of harmonic reducer DOI
Shilong Sun, Hao Ding,

Zida Zhao

et al.

Advanced Engineering Informatics, Journal Year: 2024, Volume and Issue: 62, P. 102579 - 102579

Published: May 3, 2024

Language: Английский

Citations

5

Fault detection and isolation for multi-type sensors in nuclear power plants via a knowledge-guided spatial-temporal model DOI
Weiqing Lin, Xiren Miao, Jing Chen

et al.

Knowledge-Based Systems, Journal Year: 2024, Volume and Issue: 300, P. 112182 - 112182

Published: June 26, 2024

Language: Английский

Citations

5