Data-Driven Modeling for Multiphysics Parametrized Problems-Application to Induction Hardening Process
Data-driven modeling provides an efficient approach to compute approximate solutions for complex multiphysics parametrized problems such as induction hardening (IH) process. Basically, some physical quantities of interest (QoI) related to the IH process will be evaluated under real-time constraint,...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-04-01
|
Series: | Metals |
Subjects: | |
Online Access: | https://www.mdpi.com/2075-4701/11/5/738 |
_version_ | 1797535719503167488 |
---|---|
author | Khouloud Derouiche Sevan Garois Victor Champaney Monzer Daoud Khalil Traidi Francisco Chinesta |
author_facet | Khouloud Derouiche Sevan Garois Victor Champaney Monzer Daoud Khalil Traidi Francisco Chinesta |
author_sort | Khouloud Derouiche |
collection | DOAJ |
description | Data-driven modeling provides an efficient approach to compute approximate solutions for complex multiphysics parametrized problems such as induction hardening (IH) process. Basically, some physical quantities of interest (QoI) related to the IH process will be evaluated under real-time constraint, without any explicit knowledge of the physical behavior of the system. Hence, computationally expensive finite element models will be replaced by a parametric solution, called metamodel. Two data-driven models for temporal evolution of temperature and austenite phase transformation, during induction heating, were first developed by using the proper orthogonal decomposition based reduced-order model followed by a nonlinear regression method for temperature field and a classification combined with regression for austenite evolution. Then, data-driven and hybrid models were created to predict hardness, after quenching. It is shown that the results of artificial intelligence models are promising and provide good approximations in the low-data limit case. |
first_indexed | 2024-03-10T11:49:26Z |
format | Article |
id | doaj.art-a67f47ba49c540e4867052324a810ae9 |
institution | Directory Open Access Journal |
issn | 2075-4701 |
language | English |
last_indexed | 2024-03-10T11:49:26Z |
publishDate | 2021-04-01 |
publisher | MDPI AG |
record_format | Article |
series | Metals |
spelling | doaj.art-a67f47ba49c540e4867052324a810ae92023-11-21T17:51:50ZengMDPI AGMetals2075-47012021-04-0111573810.3390/met11050738Data-Driven Modeling for Multiphysics Parametrized Problems-Application to Induction Hardening ProcessKhouloud Derouiche0Sevan Garois1Victor Champaney2Monzer Daoud3Khalil Traidi4Francisco Chinesta5PIMM Lab, Arts et Metiers Institute of Technologie, 151 Boulevard de l’Hôpital, 75013 Paris, FrancePIMM Lab, Arts et Metiers Institute of Technologie, 151 Boulevard de l’Hôpital, 75013 Paris, FrancePIMM Lab, Arts et Metiers Institute of Technologie, 151 Boulevard de l’Hôpital, 75013 Paris, FranceFrench Technological Research Institute for Materials, Metallurgy and Processes (IRT- M2P), 4 rue Augustin Fresnel, 57070 Metz, FranceSafran Tech, Rue des Jeunes Bois, 78117 Châteaufort, FrancePIMM Lab, Arts et Metiers Institute of Technologie, 151 Boulevard de l’Hôpital, 75013 Paris, FranceData-driven modeling provides an efficient approach to compute approximate solutions for complex multiphysics parametrized problems such as induction hardening (IH) process. Basically, some physical quantities of interest (QoI) related to the IH process will be evaluated under real-time constraint, without any explicit knowledge of the physical behavior of the system. Hence, computationally expensive finite element models will be replaced by a parametric solution, called metamodel. Two data-driven models for temporal evolution of temperature and austenite phase transformation, during induction heating, were first developed by using the proper orthogonal decomposition based reduced-order model followed by a nonlinear regression method for temperature field and a classification combined with regression for austenite evolution. Then, data-driven and hybrid models were created to predict hardness, after quenching. It is shown that the results of artificial intelligence models are promising and provide good approximations in the low-data limit case.https://www.mdpi.com/2075-4701/11/5/738data-driven modelinginduction hardeningmetamodelproper orthogonal decompositionartificial intelligencehybrid model |
spellingShingle | Khouloud Derouiche Sevan Garois Victor Champaney Monzer Daoud Khalil Traidi Francisco Chinesta Data-Driven Modeling for Multiphysics Parametrized Problems-Application to Induction Hardening Process Metals data-driven modeling induction hardening metamodel proper orthogonal decomposition artificial intelligence hybrid model |
title | Data-Driven Modeling for Multiphysics Parametrized Problems-Application to Induction Hardening Process |
title_full | Data-Driven Modeling for Multiphysics Parametrized Problems-Application to Induction Hardening Process |
title_fullStr | Data-Driven Modeling for Multiphysics Parametrized Problems-Application to Induction Hardening Process |
title_full_unstemmed | Data-Driven Modeling for Multiphysics Parametrized Problems-Application to Induction Hardening Process |
title_short | Data-Driven Modeling for Multiphysics Parametrized Problems-Application to Induction Hardening Process |
title_sort | data driven modeling for multiphysics parametrized problems application to induction hardening process |
topic | data-driven modeling induction hardening metamodel proper orthogonal decomposition artificial intelligence hybrid model |
url | https://www.mdpi.com/2075-4701/11/5/738 |
work_keys_str_mv | AT khouloudderouiche datadrivenmodelingformultiphysicsparametrizedproblemsapplicationtoinductionhardeningprocess AT sevangarois datadrivenmodelingformultiphysicsparametrizedproblemsapplicationtoinductionhardeningprocess AT victorchampaney datadrivenmodelingformultiphysicsparametrizedproblemsapplicationtoinductionhardeningprocess AT monzerdaoud datadrivenmodelingformultiphysicsparametrizedproblemsapplicationtoinductionhardeningprocess AT khaliltraidi datadrivenmodelingformultiphysicsparametrizedproblemsapplicationtoinductionhardeningprocess AT franciscochinesta datadrivenmodelingformultiphysicsparametrizedproblemsapplicationtoinductionhardeningprocess |