On the use of deep learning and computational fluid dynamics for the estimation of uniform momentum source components of propellers
Summary: This article proposes a novel method based on Deep Learning for the resolution of uniform momentum source terms in the Reynolds-Averaged Navier-Stokes equations. These source terms can represent several industrial devices (propellers, wind turbines, and so forth) in Computational Fluid Dyna...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2023-12-01
|
Series: | iScience |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S258900422302374X |
_version_ | 1797388778046750720 |
---|---|
author | Raúl Martínez-Cuenca Jaume Luis-Gómez Sergio Iserte Sergio Chiva |
author_facet | Raúl Martínez-Cuenca Jaume Luis-Gómez Sergio Iserte Sergio Chiva |
author_sort | Raúl Martínez-Cuenca |
collection | DOAJ |
description | Summary: This article proposes a novel method based on Deep Learning for the resolution of uniform momentum source terms in the Reynolds-Averaged Navier-Stokes equations. These source terms can represent several industrial devices (propellers, wind turbines, and so forth) in Computational Fluid Dynamics simulations. Current simulation methods require huge computational power, rely on strong assumptions or need additional information about the device that is being simulated. In this first approach to the new method, a Deep Learning system is trained with hundreds of Computational Fluid Dynamics simulations with uniform momemtum sources so that it can compute the one representing a given propeller from a reduced set of flow velocity measurements near it. Results show an overall relative error below the 5% for momentum sources for uniform sources and a moderate error when describing real propellers. This work will allow to simulate more accurately industrial devices with less computational cost. |
first_indexed | 2024-03-08T22:45:44Z |
format | Article |
id | doaj.art-e6b21b8281104bbfbf2d2b82c2696c6d |
institution | Directory Open Access Journal |
issn | 2589-0042 |
language | English |
last_indexed | 2024-03-08T22:45:44Z |
publishDate | 2023-12-01 |
publisher | Elsevier |
record_format | Article |
series | iScience |
spelling | doaj.art-e6b21b8281104bbfbf2d2b82c2696c6d2023-12-17T06:40:23ZengElsevieriScience2589-00422023-12-012612108297On the use of deep learning and computational fluid dynamics for the estimation of uniform momentum source components of propellersRaúl Martínez-Cuenca0Jaume Luis-Gómez1Sergio Iserte2Sergio Chiva3Department of Mechanical Engineering and Construction, Universitat Jaume I, 12071 Castelló de la Plana, Comunitat Valenciana, Spain; Corresponding authorDepartment of Mechanical Engineering and Construction, Universitat Jaume I, 12071 Castelló de la Plana, Comunitat Valenciana, SpainDepartment of Computer Science, Barcelona Supercomputing Center - Centro Nacional de Supercomputación (BSC-CNS), 08034 Barcelona, Cataluña, SpainDepartment of Mechanical Engineering and Construction, Universitat Jaume I, 12071 Castelló de la Plana, Comunitat Valenciana, SpainSummary: This article proposes a novel method based on Deep Learning for the resolution of uniform momentum source terms in the Reynolds-Averaged Navier-Stokes equations. These source terms can represent several industrial devices (propellers, wind turbines, and so forth) in Computational Fluid Dynamics simulations. Current simulation methods require huge computational power, rely on strong assumptions or need additional information about the device that is being simulated. In this first approach to the new method, a Deep Learning system is trained with hundreds of Computational Fluid Dynamics simulations with uniform momemtum sources so that it can compute the one representing a given propeller from a reduced set of flow velocity measurements near it. Results show an overall relative error below the 5% for momentum sources for uniform sources and a moderate error when describing real propellers. This work will allow to simulate more accurately industrial devices with less computational cost.http://www.sciencedirect.com/science/article/pii/S258900422302374XArtificial intelligenceIndustrial engineering |
spellingShingle | Raúl Martínez-Cuenca Jaume Luis-Gómez Sergio Iserte Sergio Chiva On the use of deep learning and computational fluid dynamics for the estimation of uniform momentum source components of propellers iScience Artificial intelligence Industrial engineering |
title | On the use of deep learning and computational fluid dynamics for the estimation of uniform momentum source components of propellers |
title_full | On the use of deep learning and computational fluid dynamics for the estimation of uniform momentum source components of propellers |
title_fullStr | On the use of deep learning and computational fluid dynamics for the estimation of uniform momentum source components of propellers |
title_full_unstemmed | On the use of deep learning and computational fluid dynamics for the estimation of uniform momentum source components of propellers |
title_short | On the use of deep learning and computational fluid dynamics for the estimation of uniform momentum source components of propellers |
title_sort | on the use of deep learning and computational fluid dynamics for the estimation of uniform momentum source components of propellers |
topic | Artificial intelligence Industrial engineering |
url | http://www.sciencedirect.com/science/article/pii/S258900422302374X |
work_keys_str_mv | AT raulmartinezcuenca ontheuseofdeeplearningandcomputationalfluiddynamicsfortheestimationofuniformmomentumsourcecomponentsofpropellers AT jaumeluisgomez ontheuseofdeeplearningandcomputationalfluiddynamicsfortheestimationofuniformmomentumsourcecomponentsofpropellers AT sergioiserte ontheuseofdeeplearningandcomputationalfluiddynamicsfortheestimationofuniformmomentumsourcecomponentsofpropellers AT sergiochiva ontheuseofdeeplearningandcomputationalfluiddynamicsfortheestimationofuniformmomentumsourcecomponentsofpropellers |