Global contextual feature aggregation networks with multiscale attention mechanism for mechanical fault diagnosis under non-stationary conditions

Xu, Y; Chen, Y; Zhang, H; Feng, K; Wang, Y; Yang, C; Ni, Q

Global contextual feature aggregation networks with multiscale attention mechanism for mechanical fault diagnosis under non-stationary conditions

Xu, Y Chen, Y Zhang, H Feng, K Wang, Y Yang, C Ni, Q

Permalink

Publisher:: Elsevier
Publication Type:: Journal Article
Citation:: Mechanical Systems and Signal Processing, 2023, 203, pp. 110724
Issue Date:: 2023-11-15

Closed Access

	Filename	Description	Size
	1-s2.0-S0888327023006325-main.pdf	Published version	4.07 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Xu, Y
dc.contributor.author	Chen, Y
dc.contributor.author	Zhang, H
dc.contributor.author	Feng, K
dc.contributor.author	Wang, Y
dc.contributor.author	Yang, C
dc.contributor.author	Ni, Q
dc.date.accessioned	2024-05-04T03:01:36Z
dc.date.available	2024-05-04T03:01:36Z
dc.date.issued	2023-11-15
dc.identifier.citation	Mechanical Systems and Signal Processing, 2023, 203, pp. 110724
dc.identifier.issn	0888-3270
dc.identifier.issn	1096-1216
dc.identifier.uri	http://hdl.handle.net/10453/178647
dc.description.abstract	In recent years, the rapid development of convolutional neural networks (CNNs) has significantly advanced the progress of intelligent fault diagnosis. Most currently-available CNN-based diagnostic models are built on the premise that the monitored machine operates under stable conditions. However, in real-world scenarios, rotary machinery usually operates at varying speeds, making the fault-related pulse features susceptible to noise oversaturation. To extract discriminative features from mechanical signals under non-stationary conditions, a global contextual feature aggregation network (GCFAN) is developed in this paper. To begin with, a global contextual module (GCM) is embedded in the CNN architecture to explore multimodal features. Then, a multiscale attention module (MSAM) is introduced to guide the model to focus on global and local discriminative information. Further, a multiscale feature enhancement module (MFEM) is established to enlarge the receptive field and eliminate useless features. Finally, the GCFAN architecture is constructed based on these improvements. To achieve favourable diagnostic results under fluctuating variable speed conditions, we apply the label smoothing algorithm and the AMSGrad algorithm to assist the training of the model. Two case studies using the benchmark variable speed bearing dataset and the HF-MS variable speed gearbox dataset were carried out to test the practicality of the developed approach. Experimental results demonstrated that the developed GCFAN performs better than seven state-of-the-art approaches.
dc.language	en
dc.publisher	Elsevier
dc.relation.ispartof	Mechanical Systems and Signal Processing
dc.relation.isbasedon	10.1016/j.ymssp.2023.110724
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	0905 Civil Engineering, 0913 Mechanical Engineering, 0915 Interdisciplinary Engineering
dc.subject.classification	Acoustics
dc.subject.classification	4006 Communications engineering
dc.subject.classification	4017 Mechanical engineering
dc.title	Global contextual feature aggregation networks with multiscale attention mechanism for mechanical fault diagnosis under non-stationary conditions
dc.type	Journal Article
utslib.citation.volume	203
utslib.for	0905 Civil Engineering
utslib.for	0913 Mechanical Engineering
utslib.for	0915 Interdisciplinary Engineering
pubs.organisational-group	University of Technology Sydney
pubs.organisational-group	University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	University of Technology Sydney/Faculty of Engineering and Information Technology/School of Mechanical and Mechatronic Engineering
utslib.copyright.status	closed_access	*
dc.date.updated	2024-05-04T03:01:32Z
pubs.publication-status	Published
pubs.volume	203

Abstract:

In recent years, the rapid development of convolutional neural networks (CNNs) has significantly advanced the progress of intelligent fault diagnosis. Most currently-available CNN-based diagnostic models are built on the premise that the monitored machine operates under stable conditions. However, in real-world scenarios, rotary machinery usually operates at varying speeds, making the fault-related pulse features susceptible to noise oversaturation. To extract discriminative features from mechanical signals under non-stationary conditions, a global contextual feature aggregation network (GCFAN) is developed in this paper. To begin with, a global contextual module (GCM) is embedded in the CNN architecture to explore multimodal features. Then, a multiscale attention module (MSAM) is introduced to guide the model to focus on global and local discriminative information. Further, a multiscale feature enhancement module (MFEM) is established to enlarge the receptive field and eliminate useless features. Finally, the GCFAN architecture is constructed based on these improvements. To achieve favourable diagnostic results under fluctuating variable speed conditions, we apply the label smoothing algorithm and the AMSGrad algorithm to assist the training of the model. Two case studies using the benchmark variable speed bearing dataset and the HF-MS variable speed gearbox dataset were carried out to test the practicality of the developed approach. Experimental results demonstrated that the developed GCFAN performs better than seven state-of-the-art approaches.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/178647