001026510 001__ 1026510
001026510 005__ 20250204113856.0
001026510 0247_ $$2doi$$a10.1016/j.future.2024.05.005
001026510 0247_ $$2ISSN$$a0167-739X
001026510 0247_ $$2ISSN$$a1872-7115
001026510 0247_ $$2datacite_doi$$a10.34734/FZJ-2024-03447
001026510 0247_ $$2WOS$$aWOS:001247468300001
001026510 037__ $$aFZJ-2024-03447
001026510 082__ $$a004
001026510 1001_ $$0P:(DE-HGF)0$$aHigashida, Aito$$b0
001026510 245__ $$aRobustness evaluation of large-scale machine learning-based reduced order models for reproducing flow fields
001026510 260__ $$aAmsterdam [u.a.]$$bElsevier Science$$c2024
001026510 3367_ $$2DRIVER$$aarticle
001026510 3367_ $$2DataCite$$aOutput Types/Journal article
001026510 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1718109634_32316
001026510 3367_ $$2BibTeX$$aARTICLE
001026510 3367_ $$2ORCID$$aJOURNAL_ARTICLE
001026510 3367_ $$00$$2EndNote$$aJournal Article
001026510 520__ $$aThe robustness of an artificial neural network that performs model order reduction for flow field data is studied. The network is trained with a large-scale distributed learning approach using up to 6,259 nodes of the supercomputer Fugaku. Flow around two square cylinders with a varying distance between their centers is investigated. The network is trained and tested with data from numerical simulations. First, the capability to reproduce flow fields with 2, 12, and 24 modes is investigated by comparing the reconstructed flow data to simulated data. It is shown, that reconstructions based on 2 modes cannot capture both, low- and high-frequency flow structures correctly, whereas predictions based on 12 and 24 modes yield improved flow fields, especially in the case of high-frequency waves in the vicinity of the square cylinders. Reconstructions with 24 modes provide smooth velocity fields that reproduce all relevant low- and high-frequency waves for all variations of the distance between the two square cylinders. Second, the performance of the machine learning-based reconstructions are compared to proper orthogonal decomposition, which is a commonly used reduced order model technique. The comparison only includes flow fields based on 24 modes. For all geometric variations, the mean squared errors of the reconstructions by the conventional method are higher than those of the machine learning model. This underlines the advantage of artificial neural networks over linear methods like proper orthogonal decomposition for tasks like reconstructing flow fields that are characterized by non-linear governing equations.
001026510 536__ $$0G:(DE-HGF)POF4-5111$$a5111 - Domain-Specific Simulation & Data Life Cycle Labs (SDLs) and Research Groups (POF4-511)$$cPOF4-511$$fPOF IV$$x0
001026510 536__ $$0G:(EU-Grant)951733$$aRAISE - Research on AI- and Simulation-Based Engineering at Exascale (951733)$$c951733$$fH2020-INFRAEDI-2019-1$$x1
001026510 536__ $$0G:(DE-Juel1)JLESC-20150708$$aJLESC - Joint Laboratory for Extreme Scale Computing (JLESC-20150708)$$cJLESC-20150708$$fJLESC$$x2
001026510 588__ $$aDataset connected to CrossRef, Journals: juser.fz-juelich.de
001026510 7001_ $$0P:(DE-HGF)0$$aAndo, Kazuto$$b1$$eCorresponding author
001026510 7001_ $$0P:(DE-Juel1)177985$$aRüttgers, Mario$$b2
001026510 7001_ $$0P:(DE-Juel1)165948$$aLintermann, Andreas$$b3
001026510 7001_ $$0P:(DE-HGF)0$$aTsubokura, Makoto$$b4
001026510 773__ $$0PERI:(DE-600)2020551-X$$a10.1016/j.future.2024.05.005$$gVol. 159, p. 243 - 254$$p243 - 254$$tFuture generation computer systems$$v159$$x0167-739X$$y2024
001026510 8564_ $$uhttps://juser.fz-juelich.de/record/1026510/files/1-s2.0-S0167739X24002176-main-2.pdf$$yOpenAccess
001026510 8564_ $$uhttps://juser.fz-juelich.de/record/1026510/files/1-s2.0-S0167739X24002176-main-2.gif?subformat=icon$$xicon$$yOpenAccess
001026510 8564_ $$uhttps://juser.fz-juelich.de/record/1026510/files/1-s2.0-S0167739X24002176-main-2.jpg?subformat=icon-1440$$xicon-1440$$yOpenAccess
001026510 8564_ $$uhttps://juser.fz-juelich.de/record/1026510/files/1-s2.0-S0167739X24002176-main-2.jpg?subformat=icon-180$$xicon-180$$yOpenAccess
001026510 8564_ $$uhttps://juser.fz-juelich.de/record/1026510/files/1-s2.0-S0167739X24002176-main-2.jpg?subformat=icon-640$$xicon-640$$yOpenAccess
001026510 909CO $$ooai:juser.fz-juelich.de:1026510$$pdnbdelivery$$pec_fundedresources$$pVDB$$pdriver$$popen_access$$popenaire
001026510 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)177985$$aForschungszentrum Jülich$$b2$$kFZJ
001026510 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)165948$$aForschungszentrum Jülich$$b3$$kFZJ
001026510 9131_ $$0G:(DE-HGF)POF4-511$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5111$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vEnabling Computational- & Data-Intensive Science and Engineering$$x0
001026510 9141_ $$y2024
001026510 915__ $$0LIC:(DE-HGF)CCBY4$$2HGFVOC$$aCreative Commons Attribution CC BY 4.0
001026510 915__ $$0StatID:(DE-HGF)0113$$2StatID$$aWoS$$bScience Citation Index Expanded$$d2023-08-19
001026510 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
001026510 915__ $$0StatID:(DE-HGF)0160$$2StatID$$aDBCoverage$$bEssential Science Indicators$$d2023-08-19
001026510 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bFUTURE GENER COMP SY : 2022$$d2024-12-17
001026510 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2024-12-17
001026510 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2024-12-17
001026510 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2024-12-17
001026510 915__ $$0StatID:(DE-HGF)1160$$2StatID$$aDBCoverage$$bCurrent Contents - Engineering, Computing and Technology$$d2024-12-17
001026510 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2024-12-17
001026510 915__ $$0StatID:(DE-HGF)9905$$2StatID$$aIF >= 5$$bFUTURE GENER COMP SY : 2022$$d2024-12-17
001026510 920__ $$lyes
001026510 9201_ $$0I:(DE-Juel1)JSC-20090406$$kJSC$$lJülich Supercomputing Center$$x0
001026510 980__ $$ajournal
001026510 980__ $$aVDB
001026510 980__ $$aUNRESTRICTED
001026510 980__ $$aI:(DE-Juel1)JSC-20090406
001026510 9801_ $$aFullTexts