Utilize este identificador para referenciar este registo:
https://hdl.handle.net/1822/75651
Registo completo
Campo DC | Valor | Idioma |
---|---|---|
dc.contributor.author | Faroughi, Salah A. | por |
dc.contributor.author | Roriz, Ana Isabel Araújo | por |
dc.contributor.author | Fernandes, Célio | por |
dc.date.accessioned | 2022-01-25T13:45:38Z | - |
dc.date.available | 2022-01-25T13:45:38Z | - |
dc.date.issued | 2022-01 | - |
dc.identifier.citation | Faroughi, S.A.; Roriz, A.I.; Fernandes, C. A Meta-Model to Predict the Drag Coefficient of a Particle Translating in Viscoelastic Fluids: A Machine Learning Approach. Polymers 2022, 14, 430. https://doi.org/10.3390/polym14030430 | - |
dc.identifier.issn | 2073-4360 | por |
dc.identifier.uri | https://hdl.handle.net/1822/75651 | - |
dc.description.abstract | 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, 0 < Re _ 50,Weissenberg number, 0 _ Wi _ 10, polymeric retardation ratio, 0 < z < 1, and shear thinning mobility parameter, 0 < a < 1. 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 R2 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 R2 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. | por |
dc.description.sponsorship | The authors would like to acknowledge the University of Minho cluster under the project NORTE-07-0162-FEDER-000086 (URL: http://search6.di.uminho.pt), the Minho Advanced Computing Center (MACC) (URL: https://macc.fccn.pt) under the project CPCA_A2_6052_2020, the Consorzio Interuniversitario dell’Italia Nord Est per il Calcolo Automatico (CINECA) under the Project HPC-EUROPA3 (INFRAIA-2016-1-730897) with the support of the EC Research Innovation Action under the H2020 Programme, and PRACE—Partnership for Advanced Computing in Europe under the project icei-prace-2020-0009, for providing HPC resources that have contributed to the research results reported within this paper. The authors thank Professor Gareth Huw McKinley from the Hatsopoulos Microfluids Laboratory, Department of Mechanical Engineering at the Massachusetts Institute of Technology for insightful comments regarding this work. | por |
dc.language.iso | eng | por |
dc.publisher | Multidisciplinary Digital Publishing Institute (MDPI) | por |
dc.relation | MIT-EXPL/TDI/0038/2019 | por |
dc.relation | info:eu-repo/grantAgreement/EC/H2020/730897/EU | - |
dc.rights | openAccess | por |
dc.rights.uri | https://creativecommons.org/licenses/by/4.0/ | - |
dc.subject | Machine learning | por |
dc.subject | Deep learning | por |
dc.subject | Stacked learning | por |
dc.subject | Viscoelastic flows | por |
dc.subject | Oldroyd-B fluid | por |
dc.subject | Giesekus fluid | por |
dc.subject | Sphere drag coefficient | por |
dc.title | A meta-model to predict the drag coefficient of a particle translating in viscoelastic fluids: a machine learning approach | por |
dc.type | article | - |
dc.peerreviewed | yes | por |
dc.relation.publisherversion | https://www.mdpi.com/2073-4360/14/3/430 | por |
oaire.citationIssue | 3 | por |
oaire.citationVolume | 14 | por |
dc.identifier.doi | 10.3390/polym14030430 | por |
dc.subject.fos | Engenharia e Tecnologia::Engenharia Mecânica | por |
dc.subject.wos | Science & Technology | por |
sdum.journal | Polymers | por |
dc.subject.ods | Indústria, inovação e infraestruturas | por |
Aparece nas coleções: | IPC - Artigos em revistas científicas internacionais com arbitragem |
Ficheiros deste registo:
Ficheiro | Descrição | Tamanho | Formato | |
---|---|---|---|---|
2022_JournalPaper_POLYMERS_DragSphereML_Faroughi.pdf | 2022_POLYM_DragSphereML | 12,15 MB | Adobe PDF | Ver/Abrir |
Este trabalho está licenciado sob uma Licença Creative Commons