Time-dependent performance of large-scale dome structures subjected to earthquakes using a machine learning-based evaluation method

Zhang, H; Zhang, Y; Zhu, X; Wang, H; Song, Y

Time-dependent performance of large-scale dome structures subjected to earthquakes using a machine learning-based evaluation method

Zhang, H Zhang, Y Zhu, X

Wang, H Song, Y

Permalink

Publisher:: ELSEVIER SCI LTD
Publication Type:: Journal Article
Citation:: Engineering Structures, 2022, 273
Issue Date:: 2022-12-15

Closed Access

	Filename	Description	Size
	1-s2.0-S0141029622011415-main.pdf	Published version	2.99 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	Zhang, H
dc.contributor.author	Zhang, Y
dc.contributor.author	Zhu, X https://orcid.org/0000-0001-5083-9320
dc.contributor.author	Wang, H
dc.contributor.author	Song, Y
dc.date.accessioned	2023-03-10T02:46:15Z
dc.date.available	2023-03-10T02:46:15Z
dc.date.issued	2022-12-15
dc.identifier.citation	Engineering Structures, 2022, 273
dc.identifier.issn	0141-0296
dc.identifier.issn	1873-7323
dc.identifier.uri	http://hdl.handle.net/10453/166942
dc.description.abstract	The global time-dependent performance analysis of structural systems has aroused the interest of researchers in recent years. The conventional technique using finite element analysis is very computationally intensive in evaluating the time-dependent performance of structures with multiple variables. A new hybrid method by integrating a surrogate model with the subset simulation is proposed to evaluate the time-dependent performance of large-scale complex structures subjected to earthquakes. The surrogate model is constructed based on the multilayer perceptron network using a small dataset and the subset simulation is used to estimate the small failure probability. Large-scale dome structures under earthquakes are used to verify the proposed method. The results show that surrogate models have a high accuracy in estimating extreme structural deformations when compared to the high-fidelity results using the finite element method. The hybrid method is efficient and accurate for the global time-dependent reliability analysis. The global sensitivity analysis for the dome structures is also conducted and the importance of variables over the life cycle is evaluated using the surrogate models. The numerical examples demonstrate that the method is highly efficient and accurate in quantifying uncertainty in the time-dependent performance of large-scale complex structures.
dc.language	English
dc.publisher	ELSEVIER SCI LTD
dc.relation.ispartof	Engineering Structures
dc.relation.isbasedon	10.1016/j.engstruct.2022.115065
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	0905 Civil Engineering, 0912 Materials Engineering, 0915 Interdisciplinary Engineering
dc.subject.classification	Civil Engineering
dc.title	Time-dependent performance of large-scale dome structures subjected to earthquakes using a machine learning-based evaluation method
dc.type	Journal Article
utslib.citation.volume	273
utslib.for	0905 Civil Engineering
utslib.for	0912 Materials 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 Civil and Environmental Engineering
utslib.copyright.status	closed_access	*
dc.date.updated	2023-03-10T02:45:48Z
pubs.publication-status	Published
pubs.volume	273

Abstract:

The global time-dependent performance analysis of structural systems has aroused the interest of researchers in recent years. The conventional technique using finite element analysis is very computationally intensive in evaluating the time-dependent performance of structures with multiple variables. A new hybrid method by integrating a surrogate model with the subset simulation is proposed to evaluate the time-dependent performance of large-scale complex structures subjected to earthquakes. The surrogate model is constructed based on the multilayer perceptron network using a small dataset and the subset simulation is used to estimate the small failure probability. Large-scale dome structures under earthquakes are used to verify the proposed method. The results show that surrogate models have a high accuracy in estimating extreme structural deformations when compared to the high-fidelity results using the finite element method. The hybrid method is efficient and accurate for the global time-dependent reliability analysis. The global sensitivity analysis for the dome structures is also conducted and the importance of variables over the life cycle is evaluated using the surrogate models. The numerical examples demonstrate that the method is highly efficient and accurate in quantifying uncertainty in the time-dependent performance of large-scale complex structures.

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

http://hdl.handle.net/10453/166942