Physics-informed neural network simulation of multiphase poroelasticity using stress-split sequential training
Physics-informed neural networks (PINNs) have received significant attention as a unified framework for forward, inverse, and surrogate modeling of problems governed by partial differential equations (PDEs). Training PINNs for forward problems, however, pose significant challenges, mainly because of...
Gespeichert in:
Veröffentlicht in: | arXiv.org 2021-10 |
---|---|
Hauptverfasser: | , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | arXiv.org |
container_volume | |
creator | Haghighat, Ehsan Amini, Danial Juanes, Ruben |
description | Physics-informed neural networks (PINNs) have received significant attention as a unified framework for forward, inverse, and surrogate modeling of problems governed by partial differential equations (PDEs). Training PINNs for forward problems, however, pose significant challenges, mainly because of the complex non-convex and multi-objective loss function. In this work, we present a PINN approach to solving the equations of coupled flow and deformation in porous media for both single-phase and multiphase flow. To this end, we construct the solution space using multi-layer neural networks. Due to the dynamics of the problem, we find that incorporating multiple differential relations into the loss function results in an unstable optimization problem, meaning that sometimes it converges to the trivial null solution, other times it moves very far from the expected solution. We report a dimensionless form of the coupled governing equations that we find most favourable to the optimizer. Additionally, we propose a sequential training approach based on the stress-split algorithms of poromechanics. Notably, we find that sequential training based on stress-split performs well for different problems, while the classical strain-split algorithm shows an unstable behaviour similar to what is reported in the context of finite element solvers. We use the approach to solve benchmark problems of poroelasticity, including Mandel's consolidation problem, Barry-Mercer's injection-production problem, and a reference two-phase drainage problem. The Python-SciANN codes reproducing the results reported in this manuscript will be made publicly available at https://github.com/sciann/sciann-applications. |
doi_str_mv | 10.48550/arxiv.2110.03049 |
format | Article |
fullrecord | <record><control><sourceid>proquest_arxiv</sourceid><recordid>TN_cdi_arxiv_primary_2110_03049</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2580187549</sourcerecordid><originalsourceid>FETCH-LOGICAL-a529-d61dbfecf6e8757d0541d571b12d0bdff4802a35cca051c0bf404fcbe7fb9a2b3</originalsourceid><addsrcrecordid>eNotkEtLAzEUhYMgWGp_gCsDrqcmmaQzXUrxBQVddD_kaVOnyZibUfvvTVtX53D5OPdwELqhZM5bIci9TL_-e85oOZCa8OUFmrC6plXLGbtCM4AdIYQtGiZEPUHhfXsAr6HywcW0twYHOybZF8k_MX1i8Puxl9nHgKPDxWc_bCVYPMQUbS8he-3zAY_gwweGnCxABUPvMwb7NdqQfUnLSfpQgGt06WQPdvavU7R5etysXqr12_Pr6mFdScGWlVlQo5zVbmHbRjSGCE6NaKiizBBlnOMtYbIWWksiqCbKccKdVrZxaimZqqfo9hx7GqMbkt_LdOiOo3SnUQpxdyaGFEtLyN0ujimUTh0TLaHlb6H-ABjJaT8</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2580187549</pqid></control><display><type>article</type><title>Physics-informed neural network simulation of multiphase poroelasticity using stress-split sequential training</title><source>arXiv.org</source><source>Free E- Journals</source><creator>Haghighat, Ehsan ; Amini, Danial ; Juanes, Ruben</creator><creatorcontrib>Haghighat, Ehsan ; Amini, Danial ; Juanes, Ruben</creatorcontrib><description>Physics-informed neural networks (PINNs) have received significant attention as a unified framework for forward, inverse, and surrogate modeling of problems governed by partial differential equations (PDEs). Training PINNs for forward problems, however, pose significant challenges, mainly because of the complex non-convex and multi-objective loss function. In this work, we present a PINN approach to solving the equations of coupled flow and deformation in porous media for both single-phase and multiphase flow. To this end, we construct the solution space using multi-layer neural networks. Due to the dynamics of the problem, we find that incorporating multiple differential relations into the loss function results in an unstable optimization problem, meaning that sometimes it converges to the trivial null solution, other times it moves very far from the expected solution. We report a dimensionless form of the coupled governing equations that we find most favourable to the optimizer. Additionally, we propose a sequential training approach based on the stress-split algorithms of poromechanics. Notably, we find that sequential training based on stress-split performs well for different problems, while the classical strain-split algorithm shows an unstable behaviour similar to what is reported in the context of finite element solvers. We use the approach to solve benchmark problems of poroelasticity, including Mandel's consolidation problem, Barry-Mercer's injection-production problem, and a reference two-phase drainage problem. The Python-SciANN codes reproducing the results reported in this manuscript will be made publicly available at https://github.com/sciann/sciann-applications.</description><identifier>EISSN: 2331-8422</identifier><identifier>DOI: 10.48550/arxiv.2110.03049</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Algorithms ; Computer Science - Computational Engineering, Finance, and Science ; Computer Science - Learning ; Multilayers ; Multiphase flow ; Neural networks ; Optimization ; Partial differential equations ; Poroelasticity ; Porous media ; Solution space ; Training</subject><ispartof>arXiv.org, 2021-10</ispartof><rights>2021. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,784,885,27916</link.rule.ids><backlink>$$Uhttps://doi.org/10.1016/j.cma.2022.115141$$DView published paper (Access to full text may be restricted)$$Hfree_for_read</backlink><backlink>$$Uhttps://doi.org/10.48550/arXiv.2110.03049$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Haghighat, Ehsan</creatorcontrib><creatorcontrib>Amini, Danial</creatorcontrib><creatorcontrib>Juanes, Ruben</creatorcontrib><title>Physics-informed neural network simulation of multiphase poroelasticity using stress-split sequential training</title><title>arXiv.org</title><description>Physics-informed neural networks (PINNs) have received significant attention as a unified framework for forward, inverse, and surrogate modeling of problems governed by partial differential equations (PDEs). Training PINNs for forward problems, however, pose significant challenges, mainly because of the complex non-convex and multi-objective loss function. In this work, we present a PINN approach to solving the equations of coupled flow and deformation in porous media for both single-phase and multiphase flow. To this end, we construct the solution space using multi-layer neural networks. Due to the dynamics of the problem, we find that incorporating multiple differential relations into the loss function results in an unstable optimization problem, meaning that sometimes it converges to the trivial null solution, other times it moves very far from the expected solution. We report a dimensionless form of the coupled governing equations that we find most favourable to the optimizer. Additionally, we propose a sequential training approach based on the stress-split algorithms of poromechanics. Notably, we find that sequential training based on stress-split performs well for different problems, while the classical strain-split algorithm shows an unstable behaviour similar to what is reported in the context of finite element solvers. We use the approach to solve benchmark problems of poroelasticity, including Mandel's consolidation problem, Barry-Mercer's injection-production problem, and a reference two-phase drainage problem. The Python-SciANN codes reproducing the results reported in this manuscript will be made publicly available at https://github.com/sciann/sciann-applications.</description><subject>Algorithms</subject><subject>Computer Science - Computational Engineering, Finance, and Science</subject><subject>Computer Science - Learning</subject><subject>Multilayers</subject><subject>Multiphase flow</subject><subject>Neural networks</subject><subject>Optimization</subject><subject>Partial differential equations</subject><subject>Poroelasticity</subject><subject>Porous media</subject><subject>Solution space</subject><subject>Training</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>GOX</sourceid><recordid>eNotkEtLAzEUhYMgWGp_gCsDrqcmmaQzXUrxBQVddD_kaVOnyZibUfvvTVtX53D5OPdwELqhZM5bIci9TL_-e85oOZCa8OUFmrC6plXLGbtCM4AdIYQtGiZEPUHhfXsAr6HywcW0twYHOybZF8k_MX1i8Puxl9nHgKPDxWc_bCVYPMQUbS8he-3zAY_gwweGnCxABUPvMwb7NdqQfUnLSfpQgGt06WQPdvavU7R5etysXqr12_Pr6mFdScGWlVlQo5zVbmHbRjSGCE6NaKiizBBlnOMtYbIWWksiqCbKccKdVrZxaimZqqfo9hx7GqMbkt_LdOiOo3SnUQpxdyaGFEtLyN0ujimUTh0TLaHlb6H-ABjJaT8</recordid><startdate>20211006</startdate><enddate>20211006</enddate><creator>Haghighat, Ehsan</creator><creator>Amini, Danial</creator><creator>Juanes, Ruben</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20211006</creationdate><title>Physics-informed neural network simulation of multiphase poroelasticity using stress-split sequential training</title><author>Haghighat, Ehsan ; Amini, Danial ; Juanes, Ruben</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a529-d61dbfecf6e8757d0541d571b12d0bdff4802a35cca051c0bf404fcbe7fb9a2b3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Algorithms</topic><topic>Computer Science - Computational Engineering, Finance, and Science</topic><topic>Computer Science - Learning</topic><topic>Multilayers</topic><topic>Multiphase flow</topic><topic>Neural networks</topic><topic>Optimization</topic><topic>Partial differential equations</topic><topic>Poroelasticity</topic><topic>Porous media</topic><topic>Solution space</topic><topic>Training</topic><toplevel>online_resources</toplevel><creatorcontrib>Haghighat, Ehsan</creatorcontrib><creatorcontrib>Amini, Danial</creatorcontrib><creatorcontrib>Juanes, Ruben</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection><collection>arXiv Computer Science</collection><collection>arXiv.org</collection><jtitle>arXiv.org</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Haghighat, Ehsan</au><au>Amini, Danial</au><au>Juanes, Ruben</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Physics-informed neural network simulation of multiphase poroelasticity using stress-split sequential training</atitle><jtitle>arXiv.org</jtitle><date>2021-10-06</date><risdate>2021</risdate><eissn>2331-8422</eissn><abstract>Physics-informed neural networks (PINNs) have received significant attention as a unified framework for forward, inverse, and surrogate modeling of problems governed by partial differential equations (PDEs). Training PINNs for forward problems, however, pose significant challenges, mainly because of the complex non-convex and multi-objective loss function. In this work, we present a PINN approach to solving the equations of coupled flow and deformation in porous media for both single-phase and multiphase flow. To this end, we construct the solution space using multi-layer neural networks. Due to the dynamics of the problem, we find that incorporating multiple differential relations into the loss function results in an unstable optimization problem, meaning that sometimes it converges to the trivial null solution, other times it moves very far from the expected solution. We report a dimensionless form of the coupled governing equations that we find most favourable to the optimizer. Additionally, we propose a sequential training approach based on the stress-split algorithms of poromechanics. Notably, we find that sequential training based on stress-split performs well for different problems, while the classical strain-split algorithm shows an unstable behaviour similar to what is reported in the context of finite element solvers. We use the approach to solve benchmark problems of poroelasticity, including Mandel's consolidation problem, Barry-Mercer's injection-production problem, and a reference two-phase drainage problem. The Python-SciANN codes reproducing the results reported in this manuscript will be made publicly available at https://github.com/sciann/sciann-applications.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><doi>10.48550/arxiv.2110.03049</doi><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | EISSN: 2331-8422 |
ispartof | arXiv.org, 2021-10 |
issn | 2331-8422 |
language | eng |
recordid | cdi_arxiv_primary_2110_03049 |
source | arXiv.org; Free E- Journals |
subjects | Algorithms Computer Science - Computational Engineering, Finance, and Science Computer Science - Learning Multilayers Multiphase flow Neural networks Optimization Partial differential equations Poroelasticity Porous media Solution space Training |
title | Physics-informed neural network simulation of multiphase poroelasticity using stress-split sequential training |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-15T07%3A45%3A01IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_arxiv&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Physics-informed%20neural%20network%20simulation%20of%20multiphase%20poroelasticity%20using%20stress-split%20sequential%20training&rft.jtitle=arXiv.org&rft.au=Haghighat,%20Ehsan&rft.date=2021-10-06&rft.eissn=2331-8422&rft_id=info:doi/10.48550/arxiv.2110.03049&rft_dat=%3Cproquest_arxiv%3E2580187549%3C/proquest_arxiv%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2580187549&rft_id=info:pmid/&rfr_iscdi=true |