Developing a fast machine-learning framework to replace rigorous equilibrium calculations in reactive systems and integrating it into CFD codes
Direction Conception Modélisation Procédés
Stage
Entre janvier et mai 2025
de 5 à 6 mois
Auvergne et Rhône-Alpes
Oui
IFP Energies nouvelles (IFPEN) est un acteur majeur de la recherche et de la formation dans les domaines de l’énergie, du transport et de l’environnement. De la recherche à l’industrie, l’innovation technologique est au cœur de son action, articulée autour de quatre priorités stratégiques : Mobilité Durable, Energies Nouvelles, Climat / Environnement / Economie circulaire et Hydrocarbures Responsables.
Dans le cadre de la mission d’intérêt général confiée par les pouvoirs publics, IFPEN concentre ses efforts sur :
- l’apport de solutions aux défis sociétaux de l’énergie et du climat, en favorisant la transition vers une mobilité durable et l’émergence d’un mix énergétique plus diversifié ;
- la création de richesse et d’emplois, en soutenant l’activité économique française et européenne et la compétitivité des filières industrielles associées.
Partie intégrante d’IFPEN, l’école d’ingénieurs IFP School prépare les générations futures à relever ces défis.
Developing a fast machine-learning framework to replace rigorous equilibrium calculations in reactive systems and integrating it into CFD codes
The performance of lithium-ion batteries depends highly on the properties of their cathode active material, which in turn are determined by the properties of the precursor of cathode active material (pCAM). The pCAM is usually produced by the reactive precipitation process in mixed stirred reactors, which should be optimized to obtain the pCAM with desired materials. For this purpose, Computational Fluid Dynamics (CFD) simulations can help us to find the optimum reactor design and operating conditions according to the desired pCAM properties. Therefore, it is important to develop predictive CFD tools for simulating the precipitation process of pCAM.
Description
The CFD simulation of reactive systems often requires the calculation of equilibrium thermodynamic properties, which in turn, can be computationally demanding. An example of such reactive systems is the precipitation process for pCAM production, in which the nucleation and growth of particles usually depend on the distance between the actual state of the system and its equilibrium state. In this regard, accelerating the evaluation of equilibrium thermodynamic properties can significantly help to reduce the time required for conducting simulations of this process.
The main goal of this internship is to develop a fast machine-learning framework that can be used in the CFD simulation of the pCAM synthesis process. For this task, the training data is obtained from the rigorous equilibrium models already developed in IFPEN.
The work will be divided into the following tasks:
- Conduct a literature review to select a suitable framework for accelerating equilibrium calculations
- Prepare the training and verification data set by solving the rigorous equilibrium model at different operating conditions and compositions
- Train the machine-learning tool and validate it against the predictions by the rigorous model
- Integrate the developed tool into OpenFOAM and/or Ansys Fluent CFD solvers
Required profile
Experience in python scripting, programming skills, interested in developing machine-learning tools, familiar with equilibrium calculations of chemical systems, familiar with CFD
Additional information
Duration of the internship : 6 months
Workplace : IFPEN Lyon, Rond-point de l'échangeur de Solaize, 69360 Solaize
Transport : public transportation / personal vehicle
Paid internship (1080€/month)