Convolutional neural network based reconstruction of flow-fields from concentration fields for liquid-droplet coalescence

Abstract The coalescence of liquid droplets and lenses is of great practical and theoretical importance in fluid dynamics and the statistical mechanics of multiphase flows. During such coalescence, there is an interesting and intricate interplay between the shapes of the interfaces, separating two p...

Full description

Saved in:
Bibliographic Details
Main Authors: Vasanth Kumar Babu, Nadia Bihari Padhan, Rahul Pandit
Format: Article
Language:English
Published: Nature Portfolio 2025-04-01
Series:Communications Physics
Online Access:https://doi.org/10.1038/s42005-025-02097-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849709661718052864
author Vasanth Kumar Babu
Nadia Bihari Padhan
Rahul Pandit
author_facet Vasanth Kumar Babu
Nadia Bihari Padhan
Rahul Pandit
author_sort Vasanth Kumar Babu
collection DOAJ
description Abstract The coalescence of liquid droplets and lenses is of great practical and theoretical importance in fluid dynamics and the statistical mechanics of multiphase flows. During such coalescence, there is an interesting and intricate interplay between the shapes of the interfaces, separating two phases, and the background flow field. In experiments, it is easier to visualize concentration fields than to obtain the flow field. We demonstrate that two-dimensional (2D) encoder-decoder CNNs, 2D U-Nets, and three-dimensional (3D) U-Nets can be used to obtain flow fields from concentration fields here. To train these networks, we use concentration and flow fields from our numerical simulations of the Cahn-Hilliard-Navier-Stokes equations. We show that, given test images of concentration fields, our trained models accurately predict the flow fields. Finally, we use data from recent experiments on droplet coalescence to show how our method can be used to obtain the flow field from measurements of the concentration field.
format Article
id doaj-art-46867b9d59cb45fbbebbf70f50c08bdb
institution DOAJ
issn 2399-3650
language English
publishDate 2025-04-01
publisher Nature Portfolio
record_format Article
series Communications Physics
spelling doaj-art-46867b9d59cb45fbbebbf70f50c08bdb2025-08-20T03:15:12ZengNature PortfolioCommunications Physics2399-36502025-04-018111310.1038/s42005-025-02097-yConvolutional neural network based reconstruction of flow-fields from concentration fields for liquid-droplet coalescenceVasanth Kumar Babu0Nadia Bihari Padhan1Rahul Pandit2Department of Physics, Centre for Condensed Matter Theory, Indian Institute of ScienceInstitute of Scientific Computing, TU DresdenDepartment of Physics, Centre for Condensed Matter Theory, Indian Institute of ScienceAbstract The coalescence of liquid droplets and lenses is of great practical and theoretical importance in fluid dynamics and the statistical mechanics of multiphase flows. During such coalescence, there is an interesting and intricate interplay between the shapes of the interfaces, separating two phases, and the background flow field. In experiments, it is easier to visualize concentration fields than to obtain the flow field. We demonstrate that two-dimensional (2D) encoder-decoder CNNs, 2D U-Nets, and three-dimensional (3D) U-Nets can be used to obtain flow fields from concentration fields here. To train these networks, we use concentration and flow fields from our numerical simulations of the Cahn-Hilliard-Navier-Stokes equations. We show that, given test images of concentration fields, our trained models accurately predict the flow fields. Finally, we use data from recent experiments on droplet coalescence to show how our method can be used to obtain the flow field from measurements of the concentration field.https://doi.org/10.1038/s42005-025-02097-y
spellingShingle Vasanth Kumar Babu
Nadia Bihari Padhan
Rahul Pandit
Convolutional neural network based reconstruction of flow-fields from concentration fields for liquid-droplet coalescence
Communications Physics
title Convolutional neural network based reconstruction of flow-fields from concentration fields for liquid-droplet coalescence
title_full Convolutional neural network based reconstruction of flow-fields from concentration fields for liquid-droplet coalescence
title_fullStr Convolutional neural network based reconstruction of flow-fields from concentration fields for liquid-droplet coalescence
title_full_unstemmed Convolutional neural network based reconstruction of flow-fields from concentration fields for liquid-droplet coalescence
title_short Convolutional neural network based reconstruction of flow-fields from concentration fields for liquid-droplet coalescence
title_sort convolutional neural network based reconstruction of flow fields from concentration fields for liquid droplet coalescence
url https://doi.org/10.1038/s42005-025-02097-y
work_keys_str_mv AT vasanthkumarbabu convolutionalneuralnetworkbasedreconstructionofflowfieldsfromconcentrationfieldsforliquiddropletcoalescence
AT nadiabiharipadhan convolutionalneuralnetworkbasedreconstructionofflowfieldsfromconcentrationfieldsforliquiddropletcoalescence
AT rahulpandit convolutionalneuralnetworkbasedreconstructionofflowfieldsfromconcentrationfieldsforliquiddropletcoalescence