Development of a deep and active learning approach in the real fluid modeling (RFM) framework - Application to NH3-H2 injection and mixing LES simulations

Join a cutting-edge European research network!
As part of the MSCA-DN DT-HATS project, this PhD study addresses a high-impact challenge: developing a next-generation simulation framework for hydrogen/ammonia (H₂/NH₃) heavy-duty engines, two key energy carriers in the path toward carbon neutrality.
You will join the internationally renowned research and innovation Center IFP Energies Nouvelles (IFPEN) and collaborate closely with leading European academic and industrial partners. The doctoral candidate will develop a comprehensive and effective multicomponent real fluid modeling (RFM) framework designed for simulating, for the first time, the dual-fuel injection and mixture preparation of NH3 and H2 in engines. He will introduce a deep and active learning (DAL) methodology capable of providing the thermodynamic properties required by the RFM framework during runtime when coupled with a detailed chemistry (up to 30 chemical species). The doctoral candidate will develop the required models in C++ inside the CONVERGE CFD software sources, enriched with advanced IFPEN in-house models. The resulting RFM-DAL methodology will be validated experimentally using experimental databases acquired by partners in the MSCA-DN DT-HATS project. Join a prestigious international training network to work at the cutting edge of energy transition technologies and develop sought-after skills (AI, CFD and thermodynamics), an ideal springboard for a career in industrial R&D or academic research. The work will yield significant recognition through scientific publications, participation in international conferences, and collaborations within academic and industrial communities.

Keywords: Artificial intelligence, Deep learning, Numerical Fluid Dynamics (CFD), Ammonia, Hydrogen, real gas thermodynamics, big data.

Recruitment requirement within the European project rules: the candidate must have not lived in France for more than 12 months over the last 36 months.

• Academic supervisor    Dr HABCHI Chaouki, Chaouki.Habchi@ifpen.fr,ORCID : 0000-0002-6234-3434
• Doctoral School    École Doctorale « SMEMAG » ED579 (Université Paris Saclay)
• IFPEN supervisor    Dr DELHOM Bruno, ORCID: 0009-0008-2774-9237
• PhD location    IFP Energies Nouvelles, Rueil-Malmaison, France and 4 months secondment at WinGD company, Switzerland.
• Duration and start date    3 years, starting preferentially in the fourth quarter 2025. 
• Employer    IFPEN
• Academic requirements    University Master degree involving Machine Learning, CFD and physics/thermodynamics numerical modelling. The Master's degree must have been obtained very recently (> 2022).
• Language requirements    English level B2 (CEFR); willingness to learn French
• Other requirements    Programming skills (Python, C++)
• To apply, please send your cover letter and CV to the IFPEN supervisors indicated here below.