Theme: Research
Best presentation award for George El Haber at CFC2023
1 May 2023
Congratulations to George El Haber, 3rd year PhD student, whose research work has just been awarded.
George El Haber was distinguished at the international conference CFC2023. CFC is THE conference for the Computational Fluids community. Organised by the IACM, the international association of the field, the 22nd edition was held in Cannes from 25 to 28 April 2023. It made a fantastic return to France (after 40 years) with more than 600 participants, including 207 PhD students eligible for the prize.
CFC2023 decided to reward five PhD students for the quality of their oral presentation. George El Haber was one of the winners for his outstanding presentation on a "Graph Neural Network for Two-Fluid Flows".
In his talk, he has discussed the different facets of integrating a deep learning model in the solution framework of multiphase flows. The success of such models nowadays in numerous fields has motivated him to investigate leveraging their capabilities for CFD problems, particularly for two-fluid flows. He has started his presentation by highlighting the importance of relying on a framework that caters to the specific requirements of such problems, mainly regarding the dynamically evolving discretization space and the monopoly of the flow equations on the computational burden. Then, he has provided all the details concerning our choice of the model and the various methods implemented for successfully incorporating it into the solution framework, along with some of the obtained results.
He published in the journal Physics of Fluids, Volume 35, Issue 2, February 2023: Deep learning model for two-fluid flows.
Here are few words of George about his experience at the conference:
For me, the most exciting part wasn't the presentation itself but rather the discussion that followed my presentation. Getting feedback from experienced researchers who you used to read their papers, all gathered in a single room, and being able to discuss with them and answer their concerns, really motivates you to build on your current work and work towards new objectives. In reality, the conference, as a whole, is going to be indeed a remarkable and memorable experience in my academic journey.
>> More details on the CFC2023 Conference
>> IACM website
PhD defence of Victor Claverie
10 November 2023
Victor Claverie defends his PhD in Computational Mechanics and Materials on Oct. 17, 23.
Study of the thermo-mechanical and fracture behaviour of iron oxide scales at room temperature and at high temperature.
Victor Claverie conducted his PhD work under the supervision of Pierre Montmitonnet and Karim Inal, PSF team. He defends his PhD in "Computational Mechanics and Materials" on October 17th, 2023 in front of the following jury:
– M. Henri BUSCAIL, Université Clermont Auvergne – IUT du Puy-en-Velay, Rapporteur
– Mrs Muriel BRACCINI, Grenoble INP – SIMAP, Rapporteur
– Mrs Salima BOUVIER, UTC – Roberval, Examinateur
– M. Laurent LANGLOIS, Arts et Métiers – Campus de Metz – LCFC, Examinateur
– M. Pierre MONTMITONNET, Mines Paris CEMEF, Examinateur
– M. Karim INAL, Mines Paris CEMEF, Examinateur
Abstract:
During the hot rolling process, a thin layer of oxide scale forms on the surface of steel slabs, leading to surface defects. Descaling, by applying high-pressure water to the steel surface, helps limit the thickness of the oxide and improve surface quality. This study aims to enhance the understanding of the phenomena involved in secondary descaling. Several low-carbon steel grades, characterised by differences in descaling, have been selected to analyse their discrepancies in industrial configurations. The behaviour of the oxide scale is evaluated through in situ high-temperature X-ray diffraction, enabling control over oxidation. This analysis includes both phase and stress evaluations to observe transitions in the behaviour of the oxide, particularly the relaxation of internal stresses above 700°C. Mechanical properties of the oxide are determined through indentation tests conducted at both room temperature and elevated temperatures. At lower temperatures (<600°C), indentation-induced cracks manifest in different patterns, consistent with a model of a hard film (the oxide) bending on a soft substrate (the metal), such as circular cracks and delamination. As the temperature increases, the oxide becomes ductile. Concurrently, numerical finite element simulations (FEM) of the indentation process are conducted to elucidate the mechanisms of fracture within the oxide layer. These simulations aim to extract essential fracture properties of the oxide layer and its interface. The acquired data will be used to develop a comprehensive model for hydraulic descaling.
FIB cross-section of an oxidized steel specimen after a 5N indentation and the corresponding XFEM simulation to model the fracture behavior of oxide
Keywords: iron oxide, descaling, indentation, numerical simulation, X-Ray Diffraction, hot rolling
PhD defence of Jesus-Oswaldo Garcia Carrero
11 October 2023
Jesus-Oswaldo Garcia Carrero defends his PhD in Computational Mechanics and Materials on Oct. 11, 23.
Error estimators and adaptive anisotropic remeshing in 3D coupled electromagnetic modelling – Application to electromagnetic material processing.
Jesus Oswaldo Garcia Carrero conducted his PhD work under the supervision of Fra,çois Bay, CSM team in cooperation with Transvalor. He defends his PhD in "Computational Mechanics and Materials" on October 11th, 2023 in front of the following jury:
Mrs Annie GAGNOUD, SIMAP/EPM Grenoble, Rapporteur
Mr. Marco PICASSO, EPFL Lausanne, Rapporteur
Mr. Frédéric MAGOULES, CentraleSupélec, Examinateur
Mrs Maria del Pilar SALGADO, USC Santiago de Compostela, Examinateur
Mr. Simon THIBAULT, NTN-SNR, Examinateur
Mr. François BAY, Mines Paris – PSL, CEMEF, Examinateur
Abstract:
Electromagnetic-coupled manufacturing processes involve strong multiphysics couplings between electromagnetism and other physical fields. Their design and optimisation are quite complex and relies heavily on efficient computational models. However, these models are often highly demanding in terms of resources; reducing CPU time while preserving a specified accuracy of numerical results is one of the main challenges.
The purpose of this PhD work is thus to address this challenge by developing automated anisotropic meshing procedures in conjunction with specific error estimators for the electromagnetic computations. This work has been carried out in several stages.
The first stage is the development of a robust error estimator – able to effectively identify and quantify the errors of the numerical solution in the case of complex industrial models.
The second stage deals with adaptive anisotropic remeshing and the development of a novel framework to compute the metric tensor, which needs to enable capturing the inherently anisotropic behaviour of the electromagnetic phenomena.
The third and last stage deals with modelling of complex industrial cases, based on the implementation of the developed methods in the Forge® & Thercast® software enabling multiphysical couplings with the thermodynamical phenomena.
Keywords: Computational modelling, Finite Elements, Electromagnetism, Mesh adaptation, Error estimators, Multiphysics Couplings
Upcoming conference: ICTP2023, sept. 24-29
17 September 2023
CEMEF coorganizes the International Conference on the Technology of Plasticity in Mandelieu
The International Conference on the Technology of Plasticity (ICTP) is organized for the first time in France, in the exceptional venue of the French Riviera.
ICTP, the ‘Olympic games of metal forming’ founded in 1984, is a three-yearly conference that has grown into one of the biggest international events in forming technology.
ICTP is not only a conference where the latest scientific achievements are presented, it is also a place where a bridge is built between fundamental science and industrial applications. Gathering international scientists and engineers from across industry, academia, and government gives us the opportunity to share our latest results but also to think about the solutions this ICTP community can provide to answer major challenges related to industry, society and environment issues.
Seven plenary speakers will present their works, among which, Pierre Montmitonnet, head of the PSF team, who will talk about "Space- and time-varying friction in metal forming: risks and opportunities, experimental and numerical assessment"
The 14th edition of ICTP will be held in the Mandelieu-La Napoule Congress centre from Sept. 24 to 29. The conference is co-organized by to research French Laboratories: Mines Paris – PSL – CEMEF and Arts & Métiers – LCFC. Les chercheurs impliqués sont Katia Mocelling et Pierre-Olivier Bouchard (CEMEF) et Tudor Balan et Régis Bigot (LCFC).
Fluids& Complexity conference UK
8 September 2023
CEMEF is co-organizing the 3rd Conference on Fluids & Complexity on Dec. 6-8, 23.
The 3rd edition of the International Conference on Fluids & Complexity will welcome international researchers working in different fields of fluid dynamics and soft matter physics.
The conference will be held in Nice and will cover a wide range of topics in the following fields:
– Biophysics, Active matter & Microfluidics
– Nonlinear physics, Hydrodynamic instabilities & Fluid-structure interactions
– Non-Newtonian fluids, Granular matter & Suspensions
– Multiphase flows, Capillarity & Wetting
The plenary speakers will be:
– Isabelle Cantat – Univ. Rennes, France
– Cécile Cottin-Bizonne – Univ. Lyon, France
– Philippe Coussot – Univ. Gustave Eiffel, France
– Anne De Wit – Univ. Libre de Bruxelles, Belgium
– Philippe Gondret – Univ. Paris-Saclay, France
– Karin Jacobs – Saarland Univ., Germany
– Detlef Lohse – Univ. Twente, Netherlands
– Dimos Poulikakos – ETH Zürich, Switzerland
– Olivier Pouliquen – Univ. Aix-Marseille, France
Call for papers is open. You can submit your abstract for oral or poster presentation before September 24, 23.
The conference is co-organized by Rudy Valette, CFL team, CEMEF and Céline Cohen and Ludovic Keiser, Institut de Physique de Nice.
PhD defense of Ghaniyya Medghoul
13 December 2022
Ghaniyya Medghoul will defend her PhD in Computational Mathematics, High Performance Computing and Data on Jan. 16, 23
A posteriori error estimation and adaptive control for a finite element solver framework with dynamic remeshing: application to quenching process
Ghaniyya Medghoul conducted her PhD work in the CFL team under the supervision of Elie Hachem and Aurélien Larcher. She will defend her PhD in Computational Mathematics, High Performance Computing and Data, on January 16th, 23 in front of the following jury:
– Nissrine Akkari, SafranTech, reviewer
– Joan Baiges, Polytechnical University of Catalogna, Spain, reviewer
– Alvaro Coutinho, Federal University of Rio de Janeiro, Brazil, reviewer
– Aurélien Larcher, CEMEF Mines Paris – PSL
– Elie Hachem, CEMEF Mines Paris – PSL
Abstract:
Quenching is a heat treatment process used to modify the mechanical properties of the forged, moulded or welded metal part. It consists of heating a workpiece to change its microstructure and its properties like hardness, resistance and toughness. The workpiece is then cooled in a medium (oil, water, polymer solution or air). This process is commonly used to harden and reinforce metal parts for the automotive and aerospace sectors such as rings and gears and other transmission parts. It is also used in construction sector to avoid bards' and rods' distortions and in the energy domain (for example, seamless rolled crowns).
Nowadays, with the improvement of computing power, the numerical simulation of this process become an essential tool to predict physical phenomena charactering this process such as temperature and cooling velocity, these last two are essential factors allowing to determine final characteristics of the material. Numerical simulation is an excellent tool to understand those results and to optimize them.
However, the simulation of such phenomena posed scientific difficulties because their resolution implies long computational times despite the use of important computational resources.
In this thesis, we are interested in the resolution of complex long time and large scale problems heat transfer and fluid flow problems. The goal is to offer a general adaptive stopping criteria for each iterative solver used in the aim of reducing the number of iterations and computational time. Those criteria are based on a posteriori error estimators computed on mesh's edges and based on recovery procedures. Those estimators are initially used to lead the anisotropic adaptive process to refine the mesh locally in the regions of interest. They allow to measure the quality of the approximated numerical solution by providing entirely computable upper bounds on the error between the exact solution and the approximated one.
Our numerical tests highlight the accuracy of the estimators used and the reduction in terms of iterations' number and computational cost, this reflects the efficiency of our adaptive method. The numerical framework has been validated by confrontations with experimental results provided by our industrial partners.
Keywords: a posteriori error estimation, stopping criterion, CFD modelling, anisotropic mesh adaptation, Non-smooth interpolation operators, stabilized finite elements