Using LSTM Predictions for RANS Simulations

Autores
Pasinato, Hugo D, Ing.
Año de publicación
2024
Idioma
español castellano
Tipo de recurso
artículo
Estado
versión publicada
Descripción
This study constitutes the second phase of a research endeavor aimed at evaluating the feasibility of employing Long Short-Term Memory (LSTM) neural networks as a replacement for Reynolds- Averaged Navier-Stokes (RANS) turbulence models. In the initial phase of this investigation (titled Modeling Turbulent Flows with LSTM Neural Networks, arXiv:2307.13784v1 [physics.flu-dyn] 25 Jul 2023), the application of an LSTM-based recurrent neural network (RNN) as an alternative to traditional RANS models was demonstrated. LSTM models were used to predict shear Reynolds stresses in both developed and developing tur- bulent channel flows, and these predictions were propagated through RANS simulations to obtain mean flow fields of turbulent flows. A comparative analysis was conducted, juxtaposing the LSTM results from computational fluid dynamics (CFD) simulations with outcomes from the κ − ϵ model and data from direct numerical simulations (DNS). These initial findings indicated promising per- formance of the LSTM approach. This second phase delves further into the challenges encountered and presents robust solutions. Additionally, new results are provided, demonstrating the efficacy of the LSTM model in predicting turbulent behavior in perturbed flows. While the overall study serves as a proof-of-concept for the application of LSTM networks in RANS turbulence modeling, this phase offers compelling evidence of its potential in handling more complex flow scenarios.
Fil: Pasinato, Hugo D. Universidad Tecnológica Nacional. Facultad Regional Paraná.
Peer Reviewed
Fuente
Repositorio Arxiv
Materia
Rans
Simulations
Nivel de accesibilidad
acceso abierto
Condiciones de uso
Attribution-NonCommercial-NoDerivs 2.5 Argentina
Repositorio
Repositorio Institucional Abierto (UTN)
Institución
Universidad Tecnológica Nacional
OAI Identificador
oai:ria.utn.edu.ar:20.500.12272/11712
Acceder
id RIAUTN_a6ec94350175753f03bdd63ccdca98d6
oai_identifier_str oai:ria.utn.edu.ar:20.500.12272/11712
network_acronym_str RIAUTN
repository_id_str a
network_name_str Repositorio Institucional Abierto (UTN)
spelling Using LSTM Predictions for RANS SimulationsPasinato, Hugo D, Ing.RansSimulationsThis study constitutes the second phase of a research endeavor aimed at evaluating the feasibility of employing Long Short-Term Memory (LSTM) neural networks as a replacement for Reynolds- Averaged Navier-Stokes (RANS) turbulence models. In the initial phase of this investigation (titled Modeling Turbulent Flows with LSTM Neural Networks, arXiv:2307.13784v1 [physics.flu-dyn] 25 Jul 2023), the application of an LSTM-based recurrent neural network (RNN) as an alternative to traditional RANS models was demonstrated. LSTM models were used to predict shear Reynolds stresses in both developed and developing tur- bulent channel flows, and these predictions were propagated through RANS simulations to obtain mean flow fields of turbulent flows. A comparative analysis was conducted, juxtaposing the LSTM results from computational fluid dynamics (CFD) simulations with outcomes from the κ − ϵ model and data from direct numerical simulations (DNS). These initial findings indicated promising per- formance of the LSTM approach. This second phase delves further into the challenges encountered and presents robust solutions. Additionally, new results are provided, demonstrating the efficacy of the LSTM model in predicting turbulent behavior in perturbed flows. While the overall study serves as a proof-of-concept for the application of LSTM networks in RANS turbulence modeling, this phase offers compelling evidence of its potential in handling more complex flow scenarios.Fil: Pasinato, Hugo D. Universidad Tecnológica Nacional. Facultad Regional Paraná.Peer ReviewedRepositorio Arxiv2024-11-29T21:30:21Z2024-11-19info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/resource_type/c_6501info:ar-repo/semantics/articulopdfapplication/pdfhttps://ria.utn.edu.ar/handle/20.500.12272/11712-doiRepositorio Arxivreponame:Repositorio Institucional Abierto (UTN)instname:Universidad Tecnológica Nacionalspainfo:eu-repo/semantics/openAccessAttribution-NonCommercial-NoDerivs 2.5 Argentinahttp://creativecommons.org/licenses/by-nc-nd/2.5/ar/PAsinato, Hugo D.Creative Commons / Atribución2026-09-24T12:44:10Zoai:ria.utn.edu.ar:20.500.12272/11712instacron:UTNInstitucionalhttp://ria.utn.edu.ar/Universidad públicaNo correspondehttp://ria.utn.edu.ar/oaigestionria@rec.utn.edu.ar; fsuarez@rec.utn.edu.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:a2026-09-24 12:44:12.258Repositorio Institucional Abierto (UTN) - Universidad Tecnológica Nacionalfalse
dc.title.none.fl_str_mv Using LSTM Predictions for RANS Simulations
title Using LSTM Predictions for RANS Simulations
spellingShingle Using LSTM Predictions for RANS Simulations
Pasinato, Hugo D, Ing.
Rans
Simulations
title_short Using LSTM Predictions for RANS Simulations
title_full Using LSTM Predictions for RANS Simulations
title_fullStr Using LSTM Predictions for RANS Simulations
title_full_unstemmed Using LSTM Predictions for RANS Simulations
title_sort Using LSTM Predictions for RANS Simulations
dc.creator.none.fl_str_mv Pasinato, Hugo D, Ing.
author Pasinato, Hugo D, Ing.
author_facet Pasinato, Hugo D, Ing.
author_role author
dc.subject.none.fl_str_mv Rans
Simulations
topic Rans
Simulations
dc.description.none.fl_txt_mv This study constitutes the second phase of a research endeavor aimed at evaluating the feasibility of employing Long Short-Term Memory (LSTM) neural networks as a replacement for Reynolds- Averaged Navier-Stokes (RANS) turbulence models. In the initial phase of this investigation (titled Modeling Turbulent Flows with LSTM Neural Networks, arXiv:2307.13784v1 [physics.flu-dyn] 25 Jul 2023), the application of an LSTM-based recurrent neural network (RNN) as an alternative to traditional RANS models was demonstrated. LSTM models were used to predict shear Reynolds stresses in both developed and developing tur- bulent channel flows, and these predictions were propagated through RANS simulations to obtain mean flow fields of turbulent flows. A comparative analysis was conducted, juxtaposing the LSTM results from computational fluid dynamics (CFD) simulations with outcomes from the κ − ϵ model and data from direct numerical simulations (DNS). These initial findings indicated promising per- formance of the LSTM approach. This second phase delves further into the challenges encountered and presents robust solutions. Additionally, new results are provided, demonstrating the efficacy of the LSTM model in predicting turbulent behavior in perturbed flows. While the overall study serves as a proof-of-concept for the application of LSTM networks in RANS turbulence modeling, this phase offers compelling evidence of its potential in handling more complex flow scenarios.
Fil: Pasinato, Hugo D. Universidad Tecnológica Nacional. Facultad Regional Paraná.
Peer Reviewed
description This study constitutes the second phase of a research endeavor aimed at evaluating the feasibility of employing Long Short-Term Memory (LSTM) neural networks as a replacement for Reynolds- Averaged Navier-Stokes (RANS) turbulence models. In the initial phase of this investigation (titled Modeling Turbulent Flows with LSTM Neural Networks, arXiv:2307.13784v1 [physics.flu-dyn] 25 Jul 2023), the application of an LSTM-based recurrent neural network (RNN) as an alternative to traditional RANS models was demonstrated. LSTM models were used to predict shear Reynolds stresses in both developed and developing tur- bulent channel flows, and these predictions were propagated through RANS simulations to obtain mean flow fields of turbulent flows. A comparative analysis was conducted, juxtaposing the LSTM results from computational fluid dynamics (CFD) simulations with outcomes from the κ − ϵ model and data from direct numerical simulations (DNS). These initial findings indicated promising per- formance of the LSTM approach. This second phase delves further into the challenges encountered and presents robust solutions. Additionally, new results are provided, demonstrating the efficacy of the LSTM model in predicting turbulent behavior in perturbed flows. While the overall study serves as a proof-of-concept for the application of LSTM networks in RANS turbulence modeling, this phase offers compelling evidence of its potential in handling more complex flow scenarios.
publishDate 2024
dc.date.none.fl_str_mv 2024-11-29T21:30:21Z
2024-11-19
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
http://purl.org/coar/resource_type/c_6501
info:ar-repo/semantics/articulo
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv https://ria.utn.edu.ar/handle/20.500.12272/11712
-doi
url https://ria.utn.edu.ar/handle/20.500.12272/11712
identifier_str_mv -doi
dc.language.none.fl_str_mv spa
language spa
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
Attribution-NonCommercial-NoDerivs 2.5 Argentina
http://creativecommons.org/licenses/by-nc-nd/2.5/ar/
PAsinato, Hugo D.
Creative Commons / Atribución
eu_rights_str_mv openAccess
rights_invalid_str_mv Attribution-NonCommercial-NoDerivs 2.5 Argentina
http://creativecommons.org/licenses/by-nc-nd/2.5/ar/
PAsinato, Hugo D.
Creative Commons / Atribución
dc.format.none.fl_str_mv pdf
application/pdf
dc.publisher.none.fl_str_mv Repositorio Arxiv
publisher.none.fl_str_mv Repositorio Arxiv
dc.source.none.fl_str_mv Repositorio Arxiv
reponame:Repositorio Institucional Abierto (UTN)
instname:Universidad Tecnológica Nacional
reponame_str Repositorio Institucional Abierto (UTN)
collection Repositorio Institucional Abierto (UTN)
instname_str Universidad Tecnológica Nacional
repository.name.fl_str_mv Repositorio Institucional Abierto (UTN) - Universidad Tecnológica Nacional
repository.mail.fl_str_mv gestionria@rec.utn.edu.ar; fsuarez@rec.utn.edu.ar
_version_ 1877230872634064896
score 13.24418

Publicaciones similares