Cargando…
A Meta-Model to Predict the Drag Coefficient of a Particle Translating in Viscoelastic Fluids: A Machine Learning Approach
This study presents a framework based on Machine Learning (ML) models to predict the drag coefficient of a spherical particle translating in viscoelastic fluids. For the purpose of training and testing the ML models, two datasets were generated using direct numerical simulations (DNSs) for the visco...
Autores principales: | , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2022
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8838701/ https://www.ncbi.nlm.nih.gov/pubmed/35160419 http://dx.doi.org/10.3390/polym14030430 |
_version_ | 1784650190819426304 |
---|---|
author | Faroughi, Salah A. Roriz, Ana I. Fernandes, Célio |
author_facet | Faroughi, Salah A. Roriz, Ana I. Fernandes, Célio |
author_sort | Faroughi, Salah A. |
collection | PubMed |
description | This study presents a framework based on Machine Learning (ML) models to predict the drag coefficient of a spherical particle translating in viscoelastic fluids. For the purpose of training and testing the ML models, two datasets were generated using direct numerical simulations (DNSs) for the viscoelastic unbounded flow of Oldroyd-B (OB-set containing 12,120 data points) and Giesekus (GI-set containing 4950 data points) fluids past a spherical particle. The kinematic input features were selected to be Reynolds number, [Formula: see text] , Weissenberg number, [Formula: see text] , polymeric retardation ratio, [Formula: see text] , and shear thinning mobility parameter, [Formula: see text]. The ML models, specifically Random Forest (RF), Deep Neural Network (DNN) and Extreme Gradient Boosting (XGBoost), were all trained, validated, and tested, and their best architecture was obtained using a 10-Fold cross-validation method. All the ML models presented remarkable accuracy on these datasets; however the XGBoost model resulted in the highest [Formula: see text] and the lowest root mean square error (RMSE) and mean absolute percentage error (MAPE) measures. Additionally, a blind dataset was generated using DNSs, where the input feature coverage was outside the scope of the training set or interpolated within the training sets. The ML models were tested against this blind dataset, to further assess their generalization capability. The DNN model achieved the highest [Formula: see text] and the lowest RMSE and MAPE measures when inferred on this blind dataset. Finally, we developed a meta-model using stacking technique to ensemble RF, XGBoost and DNN models and output a prediction based on the individual learner’s predictions and a DNN meta-regressor. The meta-model consistently outperformed the individual models on all datasets. |
format | Online Article Text |
id | pubmed-8838701 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2022 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-88387012022-02-13 A Meta-Model to Predict the Drag Coefficient of a Particle Translating in Viscoelastic Fluids: A Machine Learning Approach Faroughi, Salah A. Roriz, Ana I. Fernandes, Célio Polymers (Basel) Article This study presents a framework based on Machine Learning (ML) models to predict the drag coefficient of a spherical particle translating in viscoelastic fluids. For the purpose of training and testing the ML models, two datasets were generated using direct numerical simulations (DNSs) for the viscoelastic unbounded flow of Oldroyd-B (OB-set containing 12,120 data points) and Giesekus (GI-set containing 4950 data points) fluids past a spherical particle. The kinematic input features were selected to be Reynolds number, [Formula: see text] , Weissenberg number, [Formula: see text] , polymeric retardation ratio, [Formula: see text] , and shear thinning mobility parameter, [Formula: see text]. The ML models, specifically Random Forest (RF), Deep Neural Network (DNN) and Extreme Gradient Boosting (XGBoost), were all trained, validated, and tested, and their best architecture was obtained using a 10-Fold cross-validation method. All the ML models presented remarkable accuracy on these datasets; however the XGBoost model resulted in the highest [Formula: see text] and the lowest root mean square error (RMSE) and mean absolute percentage error (MAPE) measures. Additionally, a blind dataset was generated using DNSs, where the input feature coverage was outside the scope of the training set or interpolated within the training sets. The ML models were tested against this blind dataset, to further assess their generalization capability. The DNN model achieved the highest [Formula: see text] and the lowest RMSE and MAPE measures when inferred on this blind dataset. Finally, we developed a meta-model using stacking technique to ensemble RF, XGBoost and DNN models and output a prediction based on the individual learner’s predictions and a DNN meta-regressor. The meta-model consistently outperformed the individual models on all datasets. MDPI 2022-01-21 /pmc/articles/PMC8838701/ /pubmed/35160419 http://dx.doi.org/10.3390/polym14030430 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). |
spellingShingle | Article Faroughi, Salah A. Roriz, Ana I. Fernandes, Célio A Meta-Model to Predict the Drag Coefficient of a Particle Translating in Viscoelastic Fluids: A Machine Learning Approach |
title | A Meta-Model to Predict the Drag Coefficient of a Particle Translating in Viscoelastic Fluids: A Machine Learning Approach |
title_full | A Meta-Model to Predict the Drag Coefficient of a Particle Translating in Viscoelastic Fluids: A Machine Learning Approach |
title_fullStr | A Meta-Model to Predict the Drag Coefficient of a Particle Translating in Viscoelastic Fluids: A Machine Learning Approach |
title_full_unstemmed | A Meta-Model to Predict the Drag Coefficient of a Particle Translating in Viscoelastic Fluids: A Machine Learning Approach |
title_short | A Meta-Model to Predict the Drag Coefficient of a Particle Translating in Viscoelastic Fluids: A Machine Learning Approach |
title_sort | meta-model to predict the drag coefficient of a particle translating in viscoelastic fluids: a machine learning approach |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8838701/ https://www.ncbi.nlm.nih.gov/pubmed/35160419 http://dx.doi.org/10.3390/polym14030430 |
work_keys_str_mv | AT faroughisalaha ametamodeltopredictthedragcoefficientofaparticletranslatinginviscoelasticfluidsamachinelearningapproach AT rorizanai ametamodeltopredictthedragcoefficientofaparticletranslatinginviscoelasticfluidsamachinelearningapproach AT fernandescelio ametamodeltopredictthedragcoefficientofaparticletranslatinginviscoelasticfluidsamachinelearningapproach AT faroughisalaha metamodeltopredictthedragcoefficientofaparticletranslatinginviscoelasticfluidsamachinelearningapproach AT rorizanai metamodeltopredictthedragcoefficientofaparticletranslatinginviscoelasticfluidsamachinelearningapproach AT fernandescelio metamodeltopredictthedragcoefficientofaparticletranslatinginviscoelasticfluidsamachinelearningapproach |