Dynamic feature capturing in a fluid flow reduced-order model using attention-augmented autoencoders

This study looks into how adding adaptive attention to convolutional autoencoders can help reconstruct flow fields in fluid dynamics applications. The study compares the effectiveness of the proposed adaptive attention mechanism with the convolutional block attention module approach using two differ...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Engineering applications of artificial intelligence Jg. 149; S. 110463
Hauptverfasser: Beiki, Alireza, Kamali, Reza
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 01.06.2025
Schlagworte:
ISSN:0952-1976
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract This study looks into how adding adaptive attention to convolutional autoencoders can help reconstruct flow fields in fluid dynamics applications. The study compares the effectiveness of the proposed adaptive attention mechanism with the convolutional block attention module approach using two different sets of datasets. The analysis encompasses the evaluation of reconstruction loss, latent space characteristics, and the application of attention mechanisms to time series forecasting. Combining adaptive attention with involution layers enhances its ability to identify and highlight significant features, surpassing the capabilities of the convolutional block attention module. This result demonstrates an increase of over 20% in the accuracy of reconstruction. Latent space analysis shows the adaptive attention mechanism’s complex and flexible encoding, which makes it easier for the model to represent different types of data. The study also looks at how attention works and how it affects time series forecasting. It shows that a new method that combines multi-head attention and bidirectional long-short-term memory works well for forecasting over 5 s of futures of flow fields. This research provides valuable insights into the role of attention mechanisms in improving model accuracy, generalization, and forecasting capabilities in the field of fluid dynamics. [Display omitted] •Adaptive attention boosts flow field reconstruction.•Involution layers enhance latent space adaptability.•Multi-head attention improves time-series forecasting.•Attention mechanisms refine latent space representation.•Dataset-specific impact on clustering with attention.
AbstractList This study looks into how adding adaptive attention to convolutional autoencoders can help reconstruct flow fields in fluid dynamics applications. The study compares the effectiveness of the proposed adaptive attention mechanism with the convolutional block attention module approach using two different sets of datasets. The analysis encompasses the evaluation of reconstruction loss, latent space characteristics, and the application of attention mechanisms to time series forecasting. Combining adaptive attention with involution layers enhances its ability to identify and highlight significant features, surpassing the capabilities of the convolutional block attention module. This result demonstrates an increase of over 20% in the accuracy of reconstruction. Latent space analysis shows the adaptive attention mechanism’s complex and flexible encoding, which makes it easier for the model to represent different types of data. The study also looks at how attention works and how it affects time series forecasting. It shows that a new method that combines multi-head attention and bidirectional long-short-term memory works well for forecasting over 5 s of futures of flow fields. This research provides valuable insights into the role of attention mechanisms in improving model accuracy, generalization, and forecasting capabilities in the field of fluid dynamics. [Display omitted] •Adaptive attention boosts flow field reconstruction.•Involution layers enhance latent space adaptability.•Multi-head attention improves time-series forecasting.•Attention mechanisms refine latent space representation.•Dataset-specific impact on clustering with attention.
ArticleNumber 110463
Author Beiki, Alireza
Kamali, Reza
Author_xml – sequence: 1
  givenname: Alireza
  surname: Beiki
  fullname: Beiki, Alireza
  email: alireza.beiki@shirazu.ac.ir
– sequence: 2
  givenname: Reza
  orcidid: 0000-0002-8160-9277
  surname: Kamali
  fullname: Kamali, Reza
  email: rkamali@shirazu.ac.ir
BookMark eNqFkM1OwzAQhH0oEm3hFZBfIMFxHDuROIDKr1SJC5wtx15XrhI7chxQ355UhQuXXnb2MN9IMyu08MEDQjcFyQtS8Nt9Dn6nhkG5nBJa5UVBGC8XaEmaimZFI_glWo3jnhBS1owvkXk8eNU7jS2oNEXAWg2zOr_DzmOFbTc5M9_wjSOYSYPJQjQQcR8MdHgaj06VEvjkgs_UtOvnFwxWUwrg9eyK4xW6sKob4fpX1-jz-elj85pt31_eNg_bTFNGU0ZbS0hVC8t4ITRrKLF1W4mmZYY1qiaM8lYbahurrNaiMk0tSl5ZQYSwDTflGvFTro5hHCNYOUTXq3iQBZHHfeRe_u0jj_vI0z4zePcP1C6pY6MUlevO4_cnHOZyXw6iHLWby4NxEXSSJrhzET-31Iwr
CitedBy_id crossref_primary_10_1016_j_ijhydene_2025_151405
crossref_primary_10_1088_1361_6501_adfe08
Cites_doi 10.1146/annurev-fluid-010518-040547
10.1016/j.ijheatfluidflow.2023.109254
10.1017/jfm.2019.822
10.1063/5.0006492
10.1038/s41467-024-45323-x
10.1017/jfm.2020.948
10.1016/j.oceaneng.2023.113935
10.1016/j.cma.2020.113379
10.1007/s00162-020-00528-w
10.1063/5.0039986
10.1002/ima.22814
10.1017/S0022112010001217
10.1016/j.paerosci.2003.12.001
10.1063/5.0051155
10.2514/1.J056060
10.1063/5.0020721
10.1088/1873-7005/abb91d
10.1007/978-3-030-01234-2_1
10.1007/s41095-022-0271-y
10.1017/jfm.2019.62
10.1063/5.0177577
10.1038/s41467-024-45578-4
10.1063/5.0153186
10.1146/annurev-fluid-010719-060214
10.1109/CVPR46437.2021.01214
10.1017/jfm.2019.238
10.1016/j.jcp.2019.109056
10.1063/1.5127247
ContentType Journal Article
Copyright 2025 Elsevier Ltd
Copyright_xml – notice: 2025 Elsevier Ltd
DBID AAYXX
CITATION
DOI 10.1016/j.engappai.2025.110463
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Applied Sciences
Computer Science
ExternalDocumentID 10_1016_j_engappai_2025_110463
S0952197625004634
GroupedDBID --K
--M
.DC
.~1
0R~
1B1
1~.
1~5
4.4
457
4G.
5GY
5VS
7-5
71M
8P~
9JN
AABNK
AAEDT
AAEDW
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AATTM
AAXKI
AAXUO
AAYFN
AAYWO
ABBOA
ABJNI
ABMAC
ACDAQ
ACGFS
ACRLP
ACVFH
ACZNC
ADBBV
ADCNI
ADEZE
ADTZH
AEBSH
AECPX
AEIPS
AEKER
AENEX
AEUPX
AFJKZ
AFPUW
AFTJW
AGCQF
AGHFR
AGUBO
AGYEJ
AHHHB
AHJVU
AHZHX
AIALX
AIEXJ
AIGII
AIIUN
AIKHN
AITUG
AKBMS
AKRWK
AKYEP
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
ANKPU
AOUOD
APXCP
AXJTR
BJAXD
BKOJK
BLXMC
CS3
DU5
EBS
EFJIC
EFKBS
EO8
EO9
EP2
EP3
F5P
FDB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
GBOLZ
IHE
J1W
JJJVA
KOM
LG9
LY7
M41
MO0
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
ROL
RPZ
SDF
SDG
SDP
SES
SEW
SPC
SPCBC
SST
SSV
SSZ
T5K
TN5
~G-
29G
9DU
AAQXK
AAYXX
ABWVN
ABXDB
ACLOT
ACNNM
ACRPL
ADJOM
ADMUD
ADNMO
AGQPQ
ASPBG
AVWKF
AZFZN
CITATION
EFLBG
EJD
FEDTE
FGOYB
G-2
HLZ
HVGLF
HZ~
R2-
SBC
SET
UHS
WUQ
ZMT
~HD
ID FETCH-LOGICAL-c242t-2bf00587f4617c4920f8b579b4d49a80426bcd2f9fafcc75d987365f7077f96d3
ISICitedReferencesCount 3
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001446309000001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0952-1976
IngestDate Tue Nov 18 22:33:45 EST 2025
Sat Nov 29 07:43:54 EST 2025
Sat Aug 23 17:11:43 EDT 2025
IsPeerReviewed true
IsScholarly true
Keywords Convolutional autoencoders
Reduced-order modeling
Flow field reconstruction
Time series forecasting in fluid dynamics
Adaptive attention mechanisms
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c242t-2bf00587f4617c4920f8b579b4d49a80426bcd2f9fafcc75d987365f7077f96d3
ORCID 0000-0002-8160-9277
ParticipantIDs crossref_primary_10_1016_j_engappai_2025_110463
crossref_citationtrail_10_1016_j_engappai_2025_110463
elsevier_sciencedirect_doi_10_1016_j_engappai_2025_110463
PublicationCentury 2000
PublicationDate 2025-06-01
2025-06-00
PublicationDateYYYYMMDD 2025-06-01
PublicationDate_xml – month: 06
  year: 2025
  text: 2025-06-01
  day: 01
PublicationDecade 2020
PublicationTitle Engineering applications of artificial intelligence
PublicationYear 2025
Publisher Elsevier Ltd
Publisher_xml – name: Elsevier Ltd
References Guo, Xu, Liu, Liu, Jiang, Mu, Zhang, Martin, Cheng, Hu (b7) 2022; 8
Wang, Xie, Zhang, Xu (b29) 2023; 35
Li, D., Hu, J., Wang, C., Li, X., She, Q., Zhu, L., Zhang, T., Chen, Q., 2021. Involution: Inverting the inherence of convolution for visual recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 12321–12330.
Brunton, Noack, Koumoutsakos (b1) 2020; 52
Lumley (b13) 1967
Raj, Tafti, Muralidhar (b20) 2023; 35
Tang, Rabault, Kuhnle, Wang, Wang (b26) 2020; 32
Wang, Solera-Rico, Vila, Vinuesa (b28) 2024; 105
Maulik, Lusch, Balaprakash (b14) 2021; 33
Wu, Gong, Pan, Qiu, Feng, Pain (b31) 2021; 33
Fukami, Fukagata, Taira (b4) 2021; 909
Schmid (b22) 2010; 656
Sushama, Menon (b24) 2023; 33
Murata, Fukami, Fukagata (b17) 2020; 882
Minping, Chen, Eyink, Meneveau, Johnson, Perlman, Burns, Li, Szalay, Hamilton (b15) 2012
Regazzoni, Pagani, Salvador, Dede, Quarteroni (b21) 2024; 15
Vaswani, Shazeer, Parmar, Uszkoreit, Jones, Gomez, Kaiser, Polosukhin (b27) 2017; 30
Fukami, Fukagata, Taira (b3) 2019; 870
Fukami, Nakamura, Fukagata (b5) 2020; 32
Moni, Yao, Malekmohamadi (b16) 2024; 36
Rabault, Kuchta, Jensen, Réglade, Cerardi (b19) 2019; 865
Woo, S., Park, J., Lee, J.-Y., Kweon, I.S., 2018. Cbam: Convolutional block attention module. In: Proceedings of the European Conference on Computer Vision. ECCV, pp. 3–19.
Qu, Zhao, Cai (b18) 2023; 35
Taira, Brunton, Dawson, Rowley, Colonius, McKeon, Schmidt, Gordeyev, Theofilis, Ukeiley (b25) 2017; 55
Xu, Sha, Wang, Cao, Wei (b33) 2023; 272
Lucia, Beran, Silva (b12) 2004; 40
Duraisamy, Iaccarino, Xiao (b2) 2019; 51
Hasegawa, Fukami, Murata, Fukagata (b10) 2020; 34
Xu, Duraisamy (b32) 2020; 372
Solera-Rico, Sanmiguel Vila, Gómez-López, Wang, Almashjary, Dawson, Vinuesa (b23) 2024; 15
Hasegawa, Fukami, Murata, Fukagata (b9) 2020; 52
Geneva, Zabaras (b6) 2020; 403
Han, Wang, Zhang, Chen (b8) 2019; 31
Murata (10.1016/j.engappai.2025.110463_b17) 2020; 882
Vaswani (10.1016/j.engappai.2025.110463_b27) 2017; 30
Guo (10.1016/j.engappai.2025.110463_b7) 2022; 8
Hasegawa (10.1016/j.engappai.2025.110463_b10) 2020; 34
Fukami (10.1016/j.engappai.2025.110463_b5) 2020; 32
Xu (10.1016/j.engappai.2025.110463_b32) 2020; 372
Minping (10.1016/j.engappai.2025.110463_b15) 2012
Rabault (10.1016/j.engappai.2025.110463_b19) 2019; 865
Qu (10.1016/j.engappai.2025.110463_b18) 2023; 35
Wang (10.1016/j.engappai.2025.110463_b28) 2024; 105
Taira (10.1016/j.engappai.2025.110463_b25) 2017; 55
Sushama (10.1016/j.engappai.2025.110463_b24) 2023; 33
Tang (10.1016/j.engappai.2025.110463_b26) 2020; 32
Wu (10.1016/j.engappai.2025.110463_b31) 2021; 33
Moni (10.1016/j.engappai.2025.110463_b16) 2024; 36
Brunton (10.1016/j.engappai.2025.110463_b1) 2020; 52
Duraisamy (10.1016/j.engappai.2025.110463_b2) 2019; 51
Fukami (10.1016/j.engappai.2025.110463_b3) 2019; 870
10.1016/j.engappai.2025.110463_b11
Xu (10.1016/j.engappai.2025.110463_b33) 2023; 272
Lumley (10.1016/j.engappai.2025.110463_b13) 1967
10.1016/j.engappai.2025.110463_b30
Schmid (10.1016/j.engappai.2025.110463_b22) 2010; 656
Wang (10.1016/j.engappai.2025.110463_b29) 2023; 35
Hasegawa (10.1016/j.engappai.2025.110463_b9) 2020; 52
Solera-Rico (10.1016/j.engappai.2025.110463_b23) 2024; 15
Fukami (10.1016/j.engappai.2025.110463_b4) 2021; 909
Maulik (10.1016/j.engappai.2025.110463_b14) 2021; 33
Raj (10.1016/j.engappai.2025.110463_b20) 2023; 35
Regazzoni (10.1016/j.engappai.2025.110463_b21) 2024; 15
Geneva (10.1016/j.engappai.2025.110463_b6) 2020; 403
Lucia (10.1016/j.engappai.2025.110463_b12) 2004; 40
Han (10.1016/j.engappai.2025.110463_b8) 2019; 31
References_xml – volume: 52
  start-page: 477
  year: 2020
  end-page: 508
  ident: b1
  article-title: Machine learning for fluid mechanics
  publication-title: Annu. Rev. Fluid Mech.
– volume: 909
  start-page: A9
  year: 2021
  ident: b4
  article-title: Machine-learning-based spatio-temporal super resolution reconstruction of turbulent flows
  publication-title: J. Fluid Mech.
– volume: 35
  year: 2023
  ident: b20
  article-title: Comparison of reduced order models based on dynamic mode decomposition and deep learning for predicting chaotic flow in a random arrangement of cylinders
  publication-title: Phys. Fluids
– volume: 870
  start-page: 106
  year: 2019
  end-page: 120
  ident: b3
  article-title: Super-resolution reconstruction of turbulent flows with machine learning
  publication-title: J. Fluid Mech.
– volume: 51
  start-page: 357
  year: 2019
  end-page: 377
  ident: b2
  article-title: Turbulence modeling in the age of data
  publication-title: Annu. Rev. Fluid Mech.
– volume: 40
  start-page: 51
  year: 2004
  end-page: 117
  ident: b12
  article-title: Reduced-order modeling: new approaches for computational physics
  publication-title: Prog. Aerosp. Sci.
– volume: 15
  start-page: 1361
  year: 2024
  ident: b23
  article-title: -Variational autoencoders and transformers for reduced-order modelling of fluid flows
  publication-title: Nat. Commun.
– volume: 105
  year: 2024
  ident: b28
  article-title: Towards optimal
  publication-title: Int. J. Heat Fluid Flow
– volume: 34
  start-page: 367
  year: 2020
  end-page: 383
  ident: b10
  article-title: Machine-learning-based reduced-order modeling for unsteady flows around bluff bodies of various shapes
  publication-title: Theor. Comput. Fluid Dyn.
– volume: 656
  start-page: 5
  year: 2010
  end-page: 28
  ident: b22
  article-title: Dynamic mode decomposition of numerical and experimental data
  publication-title: J. Fluid Mech.
– reference: Woo, S., Park, J., Lee, J.-Y., Kweon, I.S., 2018. Cbam: Convolutional block attention module. In: Proceedings of the European Conference on Computer Vision. ECCV, pp. 3–19.
– volume: 865
  start-page: 281
  year: 2019
  end-page: 302
  ident: b19
  article-title: Artificial neural networks trained through deep reinforcement learning discover control strategies for active flow control
  publication-title: J. Fluid Mech.
– volume: 30
  year: 2017
  ident: b27
  article-title: Attention is all you need
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 33
  year: 2021
  ident: b14
  article-title: Reduced-order modeling of advection-dominated systems with recurrent neural networks and convolutional autoencoders
  publication-title: Phys. Fluids
– volume: 33
  year: 2021
  ident: b31
  article-title: Reduced order model using convolutional auto-encoder with self-attention
  publication-title: Phys. Fluids
– volume: 372
  year: 2020
  ident: b32
  article-title: Multi-level convolutional autoencoder networks for parametric prediction of spatio-temporal dynamics
  publication-title: Comput. Methods Appl. Mech. Engrg.
– volume: 403
  year: 2020
  ident: b6
  article-title: Modeling the dynamics of PDE systems with physics-constrained deep auto-regressive networks
  publication-title: J. Comput. Phys.
– volume: 33
  start-page: 701
  year: 2023
  end-page: 713
  ident: b24
  article-title: Attention augmented residual autoencoder for efficient polyp segmentation
  publication-title: Int. J. Imaging Syst. Technol.
– volume: 32
  year: 2020
  ident: b26
  article-title: Robust active flow control over a range of Reynolds numbers using an artificial neural network trained through deep reinforcement learning
  publication-title: Phys. Fluids
– volume: 272
  year: 2023
  ident: b33
  article-title: Comparative studies of predictive models for unsteady flow fields based on deep learning and proper orthogonal decomposition
  publication-title: Ocean Eng.
– start-page: 166
  year: 1967
  end-page: 178
  ident: b13
  article-title: The structure of inhomogeneous turbulent flows
  publication-title: Atmospheric Turbulence and Radio Wave Propagation
– volume: 15
  start-page: 1834
  year: 2024
  ident: b21
  article-title: Learning the intrinsic dynamics of spatio-temporal processes through Latent Dynamics Networks
  publication-title: Nat. Commun.
– volume: 31
  year: 2019
  ident: b8
  article-title: A novel spatial-temporal prediction method for unsteady wake flows based on hybrid deep neural network
  publication-title: Phys. Fluids
– year: 2012
  ident: b15
  article-title: Forced Isotropic Turbulence Data Set (Extended)
– volume: 52
  year: 2020
  ident: b9
  article-title: CNN-LSTM based reduced order modeling of two-dimensional unsteady flows around a circular cylinder at different Reynolds numbers
  publication-title: Fluid Dyn. Res.
– volume: 882
  start-page: A13
  year: 2020
  ident: b17
  article-title: Nonlinear mode decomposition with convolutional neural networks for fluid dynamics
  publication-title: J. Fluid Mech.
– volume: 35
  year: 2023
  ident: b18
  article-title: Nonlinear dynamic mode decomposition from time-resolving snapshots based on deep convolutional autoencoder
  publication-title: Phys. Fluids
– volume: 8
  start-page: 331
  year: 2022
  end-page: 368
  ident: b7
  article-title: Attention mechanisms in computer vision: A survey
  publication-title: Comput. Vis. Media
– reference: Li, D., Hu, J., Wang, C., Li, X., She, Q., Zhu, L., Zhang, T., Chen, Q., 2021. Involution: Inverting the inherence of convolution for visual recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 12321–12330.
– volume: 36
  year: 2024
  ident: b16
  article-title: Data-driven reduced-order modeling for nonlinear aerodynamics using an autoencoder neural network
  publication-title: Phys. Fluids
– volume: 32
  year: 2020
  ident: b5
  article-title: Convolutional neural network based hierarchical autoencoder for nonlinear mode decomposition of fluid field data
  publication-title: Phys. Fluids
– volume: 55
  start-page: 4013
  year: 2017
  end-page: 4041
  ident: b25
  article-title: Modal analysis of fluid flows: An overview
  publication-title: Aiaa J.
– volume: 35
  year: 2023
  ident: b29
  article-title: Physics-assisted reduced-order modeling for identifying dominant features of transonic buffet
  publication-title: Phys. Fluids
– volume: 51
  start-page: 357
  year: 2019
  ident: 10.1016/j.engappai.2025.110463_b2
  article-title: Turbulence modeling in the age of data
  publication-title: Annu. Rev. Fluid Mech.
  doi: 10.1146/annurev-fluid-010518-040547
– volume: 105
  year: 2024
  ident: 10.1016/j.engappai.2025.110463_b28
  article-title: Towards optimal β-variational autoencoders combined with transformers for reduced-order modelling of turbulent flows
  publication-title: Int. J. Heat Fluid Flow
  doi: 10.1016/j.ijheatfluidflow.2023.109254
– volume: 882
  start-page: A13
  year: 2020
  ident: 10.1016/j.engappai.2025.110463_b17
  article-title: Nonlinear mode decomposition with convolutional neural networks for fluid dynamics
  publication-title: J. Fluid Mech.
  doi: 10.1017/jfm.2019.822
– volume: 32
  issue: 5
  year: 2020
  ident: 10.1016/j.engappai.2025.110463_b26
  article-title: Robust active flow control over a range of Reynolds numbers using an artificial neural network trained through deep reinforcement learning
  publication-title: Phys. Fluids
  doi: 10.1063/5.0006492
– volume: 15
  start-page: 1834
  issue: 1
  year: 2024
  ident: 10.1016/j.engappai.2025.110463_b21
  article-title: Learning the intrinsic dynamics of spatio-temporal processes through Latent Dynamics Networks
  publication-title: Nat. Commun.
  doi: 10.1038/s41467-024-45323-x
– volume: 909
  start-page: A9
  year: 2021
  ident: 10.1016/j.engappai.2025.110463_b4
  article-title: Machine-learning-based spatio-temporal super resolution reconstruction of turbulent flows
  publication-title: J. Fluid Mech.
  doi: 10.1017/jfm.2020.948
– volume: 272
  year: 2023
  ident: 10.1016/j.engappai.2025.110463_b33
  article-title: Comparative studies of predictive models for unsteady flow fields based on deep learning and proper orthogonal decomposition
  publication-title: Ocean Eng.
  doi: 10.1016/j.oceaneng.2023.113935
– volume: 372
  year: 2020
  ident: 10.1016/j.engappai.2025.110463_b32
  article-title: Multi-level convolutional autoencoder networks for parametric prediction of spatio-temporal dynamics
  publication-title: Comput. Methods Appl. Mech. Engrg.
  doi: 10.1016/j.cma.2020.113379
– volume: 34
  start-page: 367
  year: 2020
  ident: 10.1016/j.engappai.2025.110463_b10
  article-title: Machine-learning-based reduced-order modeling for unsteady flows around bluff bodies of various shapes
  publication-title: Theor. Comput. Fluid Dyn.
  doi: 10.1007/s00162-020-00528-w
– start-page: 166
  year: 1967
  ident: 10.1016/j.engappai.2025.110463_b13
  article-title: The structure of inhomogeneous turbulent flows
– volume: 33
  issue: 3
  year: 2021
  ident: 10.1016/j.engappai.2025.110463_b14
  article-title: Reduced-order modeling of advection-dominated systems with recurrent neural networks and convolutional autoencoders
  publication-title: Phys. Fluids
  doi: 10.1063/5.0039986
– volume: 30
  year: 2017
  ident: 10.1016/j.engappai.2025.110463_b27
  article-title: Attention is all you need
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 35
  issue: 6
  year: 2023
  ident: 10.1016/j.engappai.2025.110463_b18
  article-title: Nonlinear dynamic mode decomposition from time-resolving snapshots based on deep convolutional autoencoder
  publication-title: Phys. Fluids
– volume: 33
  start-page: 701
  issue: 2
  year: 2023
  ident: 10.1016/j.engappai.2025.110463_b24
  article-title: Attention augmented residual autoencoder for efficient polyp segmentation
  publication-title: Int. J. Imaging Syst. Technol.
  doi: 10.1002/ima.22814
– volume: 656
  start-page: 5
  year: 2010
  ident: 10.1016/j.engappai.2025.110463_b22
  article-title: Dynamic mode decomposition of numerical and experimental data
  publication-title: J. Fluid Mech.
  doi: 10.1017/S0022112010001217
– volume: 40
  start-page: 51
  issue: 1–2
  year: 2004
  ident: 10.1016/j.engappai.2025.110463_b12
  article-title: Reduced-order modeling: new approaches for computational physics
  publication-title: Prog. Aerosp. Sci.
  doi: 10.1016/j.paerosci.2003.12.001
– volume: 33
  issue: 7
  year: 2021
  ident: 10.1016/j.engappai.2025.110463_b31
  article-title: Reduced order model using convolutional auto-encoder with self-attention
  publication-title: Phys. Fluids
  doi: 10.1063/5.0051155
– volume: 55
  start-page: 4013
  issue: 12
  year: 2017
  ident: 10.1016/j.engappai.2025.110463_b25
  article-title: Modal analysis of fluid flows: An overview
  publication-title: Aiaa J.
  doi: 10.2514/1.J056060
– volume: 32
  issue: 9
  year: 2020
  ident: 10.1016/j.engappai.2025.110463_b5
  article-title: Convolutional neural network based hierarchical autoencoder for nonlinear mode decomposition of fluid field data
  publication-title: Phys. Fluids
  doi: 10.1063/5.0020721
– volume: 52
  issue: 6
  year: 2020
  ident: 10.1016/j.engappai.2025.110463_b9
  article-title: CNN-LSTM based reduced order modeling of two-dimensional unsteady flows around a circular cylinder at different Reynolds numbers
  publication-title: Fluid Dyn. Res.
  doi: 10.1088/1873-7005/abb91d
– ident: 10.1016/j.engappai.2025.110463_b30
  doi: 10.1007/978-3-030-01234-2_1
– volume: 8
  start-page: 331
  issue: 3
  year: 2022
  ident: 10.1016/j.engappai.2025.110463_b7
  article-title: Attention mechanisms in computer vision: A survey
  publication-title: Comput. Vis. Media
  doi: 10.1007/s41095-022-0271-y
– volume: 865
  start-page: 281
  year: 2019
  ident: 10.1016/j.engappai.2025.110463_b19
  article-title: Artificial neural networks trained through deep reinforcement learning discover control strategies for active flow control
  publication-title: J. Fluid Mech.
  doi: 10.1017/jfm.2019.62
– volume: 35
  issue: 6
  year: 2023
  ident: 10.1016/j.engappai.2025.110463_b29
  article-title: Physics-assisted reduced-order modeling for identifying dominant features of transonic buffet
  publication-title: Phys. Fluids
– volume: 36
  issue: 1
  year: 2024
  ident: 10.1016/j.engappai.2025.110463_b16
  article-title: Data-driven reduced-order modeling for nonlinear aerodynamics using an autoencoder neural network
  publication-title: Phys. Fluids
  doi: 10.1063/5.0177577
– volume: 15
  start-page: 1361
  issue: 1
  year: 2024
  ident: 10.1016/j.engappai.2025.110463_b23
  article-title: β-Variational autoencoders and transformers for reduced-order modelling of fluid flows
  publication-title: Nat. Commun.
  doi: 10.1038/s41467-024-45578-4
– volume: 35
  issue: 7
  year: 2023
  ident: 10.1016/j.engappai.2025.110463_b20
  article-title: Comparison of reduced order models based on dynamic mode decomposition and deep learning for predicting chaotic flow in a random arrangement of cylinders
  publication-title: Phys. Fluids
  doi: 10.1063/5.0153186
– volume: 52
  start-page: 477
  year: 2020
  ident: 10.1016/j.engappai.2025.110463_b1
  article-title: Machine learning for fluid mechanics
  publication-title: Annu. Rev. Fluid Mech.
  doi: 10.1146/annurev-fluid-010719-060214
– ident: 10.1016/j.engappai.2025.110463_b11
  doi: 10.1109/CVPR46437.2021.01214
– volume: 870
  start-page: 106
  year: 2019
  ident: 10.1016/j.engappai.2025.110463_b3
  article-title: Super-resolution reconstruction of turbulent flows with machine learning
  publication-title: J. Fluid Mech.
  doi: 10.1017/jfm.2019.238
– year: 2012
  ident: 10.1016/j.engappai.2025.110463_b15
– volume: 403
  year: 2020
  ident: 10.1016/j.engappai.2025.110463_b6
  article-title: Modeling the dynamics of PDE systems with physics-constrained deep auto-regressive networks
  publication-title: J. Comput. Phys.
  doi: 10.1016/j.jcp.2019.109056
– volume: 31
  issue: 12
  year: 2019
  ident: 10.1016/j.engappai.2025.110463_b8
  article-title: A novel spatial-temporal prediction method for unsteady wake flows based on hybrid deep neural network
  publication-title: Phys. Fluids
  doi: 10.1063/1.5127247
SSID ssj0003846
Score 2.445332
Snippet This study looks into how adding adaptive attention to convolutional autoencoders can help reconstruct flow fields in fluid dynamics applications. The study...
SourceID crossref
elsevier
SourceType Enrichment Source
Index Database
Publisher
StartPage 110463
SubjectTerms Adaptive attention mechanisms
Convolutional autoencoders
Flow field reconstruction
Reduced-order modeling
Time series forecasting in fluid dynamics
Title Dynamic feature capturing in a fluid flow reduced-order model using attention-augmented autoencoders
URI https://dx.doi.org/10.1016/j.engappai.2025.110463
Volume 149
WOSCitedRecordID wos001446309000001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  issn: 0952-1976
  databaseCode: AIEXJ
  dateStart: 19950201
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.sciencedirect.com
  omitProxy: false
  ssIdentifier: ssj0003846
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1La9tAEF7cpIdemj5pkrbsoTchV5YsrfZo2pQ-IJSSgm9itY-g4MjGkZKQX5_Zp5VimubQy2Iv7Er2fJqdGc18g9CHXCS69Ql4qpKAgzKlDPSgSGIyoarQAKkzZppNkOPjcj6nP0ej3tfCXC5I25bX13T1X0UNcyBsXTr7AHGHTWECPoPQYQSxw_hPgv9se8xHShrOzoizVWdrEZs2YpFa9I2AcXkVrTVtqxSxod-0PXGi3sQONOmmSYOMWX9qaDtFxPpuqVkvhcuZD_H8DaNhNHwdbjIM1iYVyTQGGXB_hhiAbGzX7NkCNO9NOCF-sHNmy7Z_-VkXmEjzTQJViDCm8YTa9i5B2VqCUqcuJ_oNc7ZVk9ugwtlYtqdw76wZ60uMNwvuUmf_caSFREOfw3ZW-X0qvU9l93mEdlOSU1CGu7NvR_Pv4QjPSlvh5X_BoLR8-x1tt2oGlsrJM_TUuRh4ZqHxHI1k-wLtOXcDO2V-AVO-o4efe4mEAw924MEBPLhpMcMGPFiDB98BDzbgwQY8eAt48BA8r9DvL0cnn77Grg9HzMGA6-K0Vrr7JFFTMHf5lKaJKuuc0Hoq4NkutRdec5EqqpjinOSCliQrckUSQhQtRPYa7bTLVr5BWMqiULwQVKUMTOGszOA7WPhMKEJYneyj3P-PFXck9bpXyqL6uyT30cewbmVpWu5dQb2YKmdsWiOyAgTes_bgwVc7RE82j8hbtNOte_kOPeaXXXOxfu_gdwv5zahJ
linkProvider Elsevier
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Dynamic+feature+capturing+in+a+fluid+flow+reduced-order+model+using+attention-augmented+autoencoders&rft.jtitle=Engineering+applications+of+artificial+intelligence&rft.au=Beiki%2C+Alireza&rft.au=Kamali%2C+Reza&rft.date=2025-06-01&rft.issn=0952-1976&rft.volume=149&rft.spage=110463&rft_id=info:doi/10.1016%2Fj.engappai.2025.110463&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_engappai_2025_110463
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0952-1976&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0952-1976&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0952-1976&client=summon