File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Deep global model reduction learning in porous media flow simulation

TitleDeep global model reduction learning in porous media flow simulation
Authors
KeywordsPOD
Model reduction
Porous media flow
Deep learning
Neural networks
Issue Date2020
Citation
Computational Geosciences, 2020, v. 24, n. 1, p. 261-274 How to Cite?
AbstractIn this paper, we combine deep learning concepts and some proper orthogonal decomposition (POD) model reduction methods for predicting flow in heterogeneous porous media. Nonlinear flow dynamics is studied, where the dynamics is regarded as a multi-layer network. The solution at the current time step is regarded as a multi-layer network of the solution at the initial time and input parameters. As for input, we consider various sources, which include source terms (well rates), permeability fields, and initial conditions. We consider the flow dynamics, where the solution is known at some locations and the data is integrated to the flow dynamics by modifying the reduced-order model. This approach allows modifying the reduced-order formulation of the problem. Because of the small problem size, limited observed data can be handled. We consider enriching the observed data using the computational data in deep learning networks. The basis functions of the global reduced-order model are selected such that the degrees of freedom represent the solution at observation points. This way, we can avoid learning basis functions, which can also be done using neural networks. We present numerical results, where we consider channelized permeability fields, where the network is constructed for various channel configurations. Our numerical results show that one can achieve a good approximation using forward feed maps based on multi-layer networks.
Persistent Identifierhttp://hdl.handle.net/10722/303640
ISSN
2020 Impact Factor: 2.413
2020 SCImago Journal Rankings: 0.696
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorCheung, Siu Wun-
dc.contributor.authorChung, Eric T.-
dc.contributor.authorEfendiev, Yalchin-
dc.contributor.authorGildin, Eduardo-
dc.contributor.authorWang, Yating-
dc.contributor.authorZhang, Jingyan-
dc.date.accessioned2021-09-15T08:25:43Z-
dc.date.available2021-09-15T08:25:43Z-
dc.date.issued2020-
dc.identifier.citationComputational Geosciences, 2020, v. 24, n. 1, p. 261-274-
dc.identifier.issn1420-0597-
dc.identifier.urihttp://hdl.handle.net/10722/303640-
dc.description.abstractIn this paper, we combine deep learning concepts and some proper orthogonal decomposition (POD) model reduction methods for predicting flow in heterogeneous porous media. Nonlinear flow dynamics is studied, where the dynamics is regarded as a multi-layer network. The solution at the current time step is regarded as a multi-layer network of the solution at the initial time and input parameters. As for input, we consider various sources, which include source terms (well rates), permeability fields, and initial conditions. We consider the flow dynamics, where the solution is known at some locations and the data is integrated to the flow dynamics by modifying the reduced-order model. This approach allows modifying the reduced-order formulation of the problem. Because of the small problem size, limited observed data can be handled. We consider enriching the observed data using the computational data in deep learning networks. The basis functions of the global reduced-order model are selected such that the degrees of freedom represent the solution at observation points. This way, we can avoid learning basis functions, which can also be done using neural networks. We present numerical results, where we consider channelized permeability fields, where the network is constructed for various channel configurations. Our numerical results show that one can achieve a good approximation using forward feed maps based on multi-layer networks.-
dc.languageeng-
dc.relation.ispartofComputational Geosciences-
dc.subjectPOD-
dc.subjectModel reduction-
dc.subjectPorous media flow-
dc.subjectDeep learning-
dc.subjectNeural networks-
dc.titleDeep global model reduction learning in porous media flow simulation-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1007/s10596-019-09918-4-
dc.identifier.scopuseid_2-s2.0-85077160605-
dc.identifier.volume24-
dc.identifier.issue1-
dc.identifier.spage261-
dc.identifier.epage274-
dc.identifier.eissn1573-1499-
dc.identifier.isiWOS:000519376000014-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats