Machine learning of hidden variables in multiscale fluid simulation

Solving fluid dynamics equations often requires the use of closure relations that account for missing microphysics. For example, when solving equations related to fluid dynamics for systems with a large Reynolds number, sub-grid effects become important and a turbulence closure is required, and in s...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Joglekar, Archis S, Thomas, Alexander G. R
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Joglekar, Archis S
Thomas, Alexander G. R
description Solving fluid dynamics equations often requires the use of closure relations that account for missing microphysics. For example, when solving equations related to fluid dynamics for systems with a large Reynolds number, sub-grid effects become important and a turbulence closure is required, and in systems with a large Knudsen number, kinetic effects become important and a kinetic closure is required. By adding an equation governing the growth and transport of the quantity requiring the closure relation, it becomes possible to capture microphysics through the introduction of ``hidden variables'' that are non-local in space and time. The behavior of the ``hidden variables'' in response to the fluid conditions can be learned from a higher fidelity or ab-initio model that contains all the microphysics. In our study, a partial differential equation simulator that is end-to-end differentiable is used to train judiciously placed neural networks against ground-truth simulations. We show that this method enables an Euler equation based approach to reproduce non-linear, large Knudsen number plasma physics that can otherwise only be modeled using Boltzmann-like equation simulators such as Vlasov or Particle-In-Cell modeling.
doi_str_mv 10.48550/arxiv.2306.10709
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2306_10709</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2306_10709</sourcerecordid><originalsourceid>FETCH-LOGICAL-a679-387f1c5c50ec6fdabd933c456c559dd843be6efda9634bb963a3d30255def0ec3</originalsourceid><addsrcrecordid>eNotj81OwzAQhH3hgFoegBN-gQSnGzvxsYr4k4p66T1a22u6kusip63g7QmFy4w0oxnpE-K-UXXba60esXzxpV6BMnWjOmVvxfCOfs-ZZCIsmfOHPEa55xAoywsWRpdokpzl4ZxOPHlMJGM6c5ATzxGe-JiX4iZimuju3xdi9_y0G16rzfblbVhvKjSdraDvYuO114q8iQFdsAC-1cZrbUPoW3BkaC6sgda5WRECqJXWgeK8gYV4-Lu9UoyfhQ9YvsdfmvFKAz9xNkYk</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Machine learning of hidden variables in multiscale fluid simulation</title><source>arXiv.org</source><creator>Joglekar, Archis S ; Thomas, Alexander G. R</creator><creatorcontrib>Joglekar, Archis S ; Thomas, Alexander G. R</creatorcontrib><description>Solving fluid dynamics equations often requires the use of closure relations that account for missing microphysics. For example, when solving equations related to fluid dynamics for systems with a large Reynolds number, sub-grid effects become important and a turbulence closure is required, and in systems with a large Knudsen number, kinetic effects become important and a kinetic closure is required. By adding an equation governing the growth and transport of the quantity requiring the closure relation, it becomes possible to capture microphysics through the introduction of ``hidden variables'' that are non-local in space and time. The behavior of the ``hidden variables'' in response to the fluid conditions can be learned from a higher fidelity or ab-initio model that contains all the microphysics. In our study, a partial differential equation simulator that is end-to-end differentiable is used to train judiciously placed neural networks against ground-truth simulations. We show that this method enables an Euler equation based approach to reproduce non-linear, large Knudsen number plasma physics that can otherwise only be modeled using Boltzmann-like equation simulators such as Vlasov or Particle-In-Cell modeling.</description><identifier>DOI: 10.48550/arxiv.2306.10709</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Numerical Analysis ; Mathematics - Numerical Analysis ; Physics - Computational Physics ; Physics - Fluid Dynamics ; Physics - Plasma Physics</subject><creationdate>2023-06</creationdate><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2306.10709$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2306.10709$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Joglekar, Archis S</creatorcontrib><creatorcontrib>Thomas, Alexander G. R</creatorcontrib><title>Machine learning of hidden variables in multiscale fluid simulation</title><description>Solving fluid dynamics equations often requires the use of closure relations that account for missing microphysics. For example, when solving equations related to fluid dynamics for systems with a large Reynolds number, sub-grid effects become important and a turbulence closure is required, and in systems with a large Knudsen number, kinetic effects become important and a kinetic closure is required. By adding an equation governing the growth and transport of the quantity requiring the closure relation, it becomes possible to capture microphysics through the introduction of ``hidden variables'' that are non-local in space and time. The behavior of the ``hidden variables'' in response to the fluid conditions can be learned from a higher fidelity or ab-initio model that contains all the microphysics. In our study, a partial differential equation simulator that is end-to-end differentiable is used to train judiciously placed neural networks against ground-truth simulations. We show that this method enables an Euler equation based approach to reproduce non-linear, large Knudsen number plasma physics that can otherwise only be modeled using Boltzmann-like equation simulators such as Vlasov or Particle-In-Cell modeling.</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Numerical Analysis</subject><subject>Mathematics - Numerical Analysis</subject><subject>Physics - Computational Physics</subject><subject>Physics - Fluid Dynamics</subject><subject>Physics - Plasma Physics</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj81OwzAQhH3hgFoegBN-gQSnGzvxsYr4k4p66T1a22u6kusip63g7QmFy4w0oxnpE-K-UXXba60esXzxpV6BMnWjOmVvxfCOfs-ZZCIsmfOHPEa55xAoywsWRpdokpzl4ZxOPHlMJGM6c5ATzxGe-JiX4iZimuju3xdi9_y0G16rzfblbVhvKjSdraDvYuO114q8iQFdsAC-1cZrbUPoW3BkaC6sgda5WRECqJXWgeK8gYV4-Lu9UoyfhQ9YvsdfmvFKAz9xNkYk</recordid><startdate>20230619</startdate><enddate>20230619</enddate><creator>Joglekar, Archis S</creator><creator>Thomas, Alexander G. R</creator><scope>AKY</scope><scope>AKZ</scope><scope>GOX</scope></search><sort><creationdate>20230619</creationdate><title>Machine learning of hidden variables in multiscale fluid simulation</title><author>Joglekar, Archis S ; Thomas, Alexander G. R</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a679-387f1c5c50ec6fdabd933c456c559dd843be6efda9634bb963a3d30255def0ec3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Numerical Analysis</topic><topic>Mathematics - Numerical Analysis</topic><topic>Physics - Computational Physics</topic><topic>Physics - Fluid Dynamics</topic><topic>Physics - Plasma Physics</topic><toplevel>online_resources</toplevel><creatorcontrib>Joglekar, Archis S</creatorcontrib><creatorcontrib>Thomas, Alexander G. R</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv Mathematics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Joglekar, Archis S</au><au>Thomas, Alexander G. R</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Machine learning of hidden variables in multiscale fluid simulation</atitle><date>2023-06-19</date><risdate>2023</risdate><abstract>Solving fluid dynamics equations often requires the use of closure relations that account for missing microphysics. For example, when solving equations related to fluid dynamics for systems with a large Reynolds number, sub-grid effects become important and a turbulence closure is required, and in systems with a large Knudsen number, kinetic effects become important and a kinetic closure is required. By adding an equation governing the growth and transport of the quantity requiring the closure relation, it becomes possible to capture microphysics through the introduction of ``hidden variables'' that are non-local in space and time. The behavior of the ``hidden variables'' in response to the fluid conditions can be learned from a higher fidelity or ab-initio model that contains all the microphysics. In our study, a partial differential equation simulator that is end-to-end differentiable is used to train judiciously placed neural networks against ground-truth simulations. We show that this method enables an Euler equation based approach to reproduce non-linear, large Knudsen number plasma physics that can otherwise only be modeled using Boltzmann-like equation simulators such as Vlasov or Particle-In-Cell modeling.</abstract><doi>10.48550/arxiv.2306.10709</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2306.10709
ispartof
issn
language eng
recordid cdi_arxiv_primary_2306_10709
source arXiv.org
subjects Computer Science - Artificial Intelligence
Computer Science - Numerical Analysis
Mathematics - Numerical Analysis
Physics - Computational Physics
Physics - Fluid Dynamics
Physics - Plasma Physics
title Machine learning of hidden variables in multiscale fluid simulation
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-07T20%3A21%3A12IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Machine%20learning%20of%20hidden%20variables%20in%20multiscale%20fluid%20simulation&rft.au=Joglekar,%20Archis%20S&rft.date=2023-06-19&rft_id=info:doi/10.48550/arxiv.2306.10709&rft_dat=%3Carxiv_GOX%3E2306_10709%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true