Data‐Driven Subgrid‐Scale Modeling for Convection‐Dominated Concentration Boundary Layers

Weiner, Andre; Hillenbrand, Dennis; Marschall, Holger; Bothe, Dieter

doi:10.1002/ceat.201900044

Cited by 12 publications

(6 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Rather than approximating concentration profiles from which the fluxes are computed, an ML model can directly map the available information onto the target quantities. For example, an ML model may be used to predict the diffusive flux at the interface directly based on the cell-centered concentration and the cell geometry [29].…”

Section: Extension To Reactions Involving Multiple Speciesmentioning

confidence: 99%

Computing Mass Transfer at Deformable Bubbles for High Schmidt Numbers

Weiner

Gründing

Bothe

2020

Chemie Ingenieur Technik

Self Cite

View full text Add to dashboard Cite

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Dedicated to Prof. Dr.-Ing. Matthias Kraume on the occasion of his 65th birthday The occurrence of extremely thin liquid-sided concentration boundary layers at bubble or droplet interfaces for realistic, i.e., high Schmidt numbers is a severe obstacle for the numerical simulation of mass transfer processes in gas-liquid systems. This contribution provides a survey of different approaches to overcome this problem, with the main emphasis put on the approach introduced and further developed by the authors. This approach employs a nonlinear flux computation and is based on the modeling of subgrid-scale concentration profiles. Based on the latest developments, recommendations for future research are also provided.

show abstract

Section: Extension To Reactions Involving Multiple Speciesmentioning

confidence: 99%

Computing Mass Transfer at Deformable Bubbles for High Schmidt Numbers

Weiner

Gründing

Bothe

2020

Chemie Ingenieur Technik

Self Cite

View full text Add to dashboard Cite

show abstract

“…With this premise, one of its functionalities offers a function object that can execute Python code using an integrated Python interpreter. In addition, Maulik et al [11] and Weiner et al [12,13] have aimed to provide capabilities to use data analysis tools from Python within OpenFOAM; the former constructed OpenFOAM applications that have bindings to data libraries in Python and the latter transfers data and neural network models from Python to OpenFOAM via the PyTorch library. More recently, Anderluh and Jasak [14] have proposed to re-write the core OpenFOAM functionality entirely in Python.…”

Section: Introductionmentioning

confidence: 99%

A general approach for running Python codes in OpenFOAM using an embedded pybind11 Python interpreter

Rodríguez¹,

Cardiff²

2022

Preprint

View full text Add to dashboard Cite

As the overlap between traditional computational mechanics and machine learning grows, there is an increasing demand for straight-forward approaches to interface Python-based procedures with C++-based OpenFOAM. This article introduces one such general methodology, allowing the execution of Python code directly within an OpenFOAM solver without the need for Python code translation. The proposed approach is based on the lightweight library pybind11, where OpenFOAM data is transferred to an embedded Python interpreter for manipulation, and results are returned as needed. Following a review of related approaches, the article describes the approach, with a particular focus on data transfer between Python and OpenFOAM, executing Python scripts and functions, and practical details about the implementation in OpenFOAM. Three complementary test cases are presented to highlight the functionality and demonstrate the effect of different data transfer approaches: a Python-based velocity profile boundary condition; a Python-based solver for prototyping; and a machine learning mechanical constitutive law class for solids4foam which performs field calculations.

show abstract

“…Two families of SGS models for computing the convection‐dominated mass transfer at rising bubbles are available in literature. The first group 13,14,17 uses an analytical profile function or machine learning (ML) model to correct convective and diffusive species fluxes in cells containing or aligned with the gas–liquid interface. This type of SGS model has been implemented in volume‐of‐fluid 13,14 and interface‐tracking 15,16 two‐phase flow solvers.…”

Section: Introductionmentioning

confidence: 99%