FairCFD

PhD Position
Hybrid machine learning models for sloshing in fuel tanks
(MSCA-DN FairCFD, DC3)

This is much more than just a PhD position! 

Within the FairCFD Doctoral Network, you will benefit from a unique three-fold experience:
–  Contribute to technological innovation in the field of Aerospace industry, in direct collaboration with Airbus, by developing advanced and efficient CFD strategies.
–  Take part in a network-wide interdisciplinary effort to define and promote numerical sustainability in scientific research.
–  Join a vibrant network of 15 doctoral candidates, across 9 European countries, with access to cutting-edge network events, high-level training to technical and transverse skills, and secondments in both academic and industrial environments.

See our website for more details on the network’s philosophy and actions, along with description of the other PhD positions (https://www.imft.fr/faircfd/project-presentation/).

Scientific program

Background

Accurate modeling of complex fluid dynamics remains a major barrier in science and engineering. Despite advances in CFD, simulating nonlinear, multi-scale flows such as sloshing in aircraft fuel tanks is still computationally prohibitive. These limitations hinder rapid design, uncertainty quantification, and real-time decision-making in industry.

Artificial Intelligence (AI) and Scientific Machine Learning (SciML) are reshaping how we approach these challenges. By embedding physical principles into data-driven architectures, hybrid AI models can achieve high predictive accuracy, low computational cost, and physical interpretability—a combination that traditional CFD or purely data-driven models cannot deliver.

This PhD will push the frontier of AI for Science by developing novel hybrid machine learning reduced-order models (ROMs) for industrial fluid dynamics applications, with a main focus on sloshing dynamics in aircraft fuel tanks in collaboration with Airbus. These tools will contribute directly to next-generation digital twins for aerospace systems and extend to other turbulent or multiphase flows.

Selected references:

Abadía-Heredia, R., Corrochano, A., López-Martín, M., Le Clainche, S., Generalization capabilities and robustness of hybrid machine learning models grounded in flow physics compared to purely deep learning models, Phys. Fluids, 27, 035149, 2025.

Hetherington, A., Corrochano, A., Abadía-Heredia, R., Lazpita, E., Muñoz, E., Díaz, E., Maiora, E., López-Martín, M., Le Clainche, S., ModelFLOWs-app: data-driven post-processing and reduced order modelling tools, Comp. Phys. Commu., 301, 109217, 2024.

Le Clainche, S., Vega, J.M., ‘Higher order dynamic mode decomposition to identify and extrapolate flow patterns’, Physics of Fluids, Vol. 29 (8), 084102, 2017.

Abadía-Heredia, R., López-Martín, M., Le Clainche, S., An Adaptive Framework for Autoregressive Forecasting in CFD Using Hybrid Modal Decomposition and Deep Learning, arXiv:2505.01531, 2025.

Your research program

The PhD will develop a new generation of AI-driven modeling tools that merge physics-based insight with modern ML architectures, bridging scientific understanding and industrial applicability.

Objectives:

• Advance Scientific Machine Learning frameworks that integrate modal decomposition (POD, DMD, Koopman) with neural architectures (CNNs, LSTMs, transformers).
• Design interpretable hybrid models capable of learning nonlinear flow dynamics while respecting conservation laws and physical symmetries.
• Apply and validate these models using Airbus’s high-fidelity sloshing databases, targeting accurate flow reconstruction, forecasting, and uncertainty quantification.
• Extend and generalize the developed methods to other turbulent or multiphase flow configurations, enhancing robustness and transferability.
• Deliver scalable AI tools for industrial use—enabling real-time analysis and reduced computational cost for complex fluid systems.

Where you will work

You will join the ModelFLOWs research group at the Universidad Politécnica de Madrid (UPM), School of Aerospace Engineering. ModelFLOWs specializes in scientific machine learning, reduced-order modeling, and computational fluid dynamics (CFD), developing interpretable and efficient models for real-world impact—from aeronautics to energy and environmental applications.

You will work at the intersection of AI, physics, and high-performance computing, contributing to the future of intelligent simulation and digital twins for the aerospace industry.

Network activities 

Integration within the FairCFD Network

Within the FairCFD network, you will contribute mainly to WP2: Efficient data-based approaches. You will regularly exchange with other DCs of the network applying similar approaches to other problems, and/or applying different numerical methods to similar problems.
Two secondments (short research stays in other partners of the network) are planned during the PhD: with 1/ Airbus (6 months) to generate industrial databases to create robust machine learning models; 2/ IFMT (2 months): to validate the tools developed with other methodologies from the project.

Interdisciplinary task: co-designing numerical frugality 

Beyond your individual research program described above, you will contribute along with all other FairCFD doctoral candidates to a network-wide multidisciplinary effort (WP5) addressing the environmental and societal dimensions of numerical simulation.
Each DC will participate in the definition of practical metrics for numerical frugality (computational cost, energy use, resource impact) and contribute data from their simulations to a collective meta-analysis. This initiative will be supported by interdisciplinary experts and accompanied by a dedicated DC in social sciences, who will lead a qualitative study on evolving practices in simulation across the network. Together, we aim to build concrete, informed recommendations for sustainable scientific computing.

Network Training Program — More Than Just a PhD

As a Doctoral Network funded by Marie Sklodowska-Curie Actions (MSCA-DN), FairCFD will offer to you a rich and engaging training experience, including

• Four one-week training events; (i) an induction week devoted to team-building, open-science practices and sustainability issues, (ii) an Essential Skills Accelerator event combining aiming to to equip DCs with essential technical and transferable skills, (iii) a Hackathon event where DCs will collaborate in teams to solve complex physics problem and compare various simulation strategies in terms of precision and sobriety, and (iv) a Career and Leadership Development Forum Aiming to equip DCs with transferable skills essential for their future careers.

• Five On-line courses combining technical training to state-of-the art simulation methods ranging from physics-based approaches to data-driven ones, exposition to industrial applications, along with Social, ethical and environmental aspects of decision-making in modelling practices.
• Involvement in the organisation of scientific events, including a mini-symposium as part of a large-audience scientific conference, a scientific symposium allowing to share the output in terms of new methods, innovation, and applications to industrial processes, and a Societal colloquium to deliver the outputs of the multidisciplinary tasks of the network.

This programme is designed to support your growth as a researcher, innovator, and engaged citizen, fully equipped to lead the next generation of responsible simulation science. See our website for more details (https://www.imft.fr/faircfd/project-presentation/).

Practical Informations

Skills/Qualifications 

• Master’s degree (or equivalent) in fluid mechanics, applied mathematics, scientific computing, or related fields.
• Strong background in fluid mechanics, numerical methods, PDEs, and/or data-driven modeling.
• Interest in interdisciplinary research and open science.

Benefits

The successful candidates will receive an attractive salary in accordance with the MSCA regulations for Doctoral Researchers. The exact (net) salary will be confirmed upon appointment and is dependent on local tax regulations and on the country correction factor (to allow for the difference in cost of living in different EU Member States). The salary includes a living allowance, a mobility allowance, and a family allowance (if applicable). The guaranteed PhD funding is for 36 months (i.e., EC funding, additional funding is possible, depending on the local Supervisor, and in accordance with the regular PhD time in the country of origin).

Eligibility criteria

According to the international mobility rules of the MSCA-DN program, the candidates must not have spent more than 12 months in the hosting country (Spain), during the 36 months preceding the starting of the PhD. Apart from this rule, worldwide applications are expected and encouraged.

Selection process

The recruitment process will adhere to the principles of equal opportunities, transparency and non-discrimination, in line with the European Charter for Researchers, and will actively promote diversity and gender balance. In particular, applications from female candidates are strongly encouraged, as women remain underrepresented in engineering disciplines.

To apply, please gather a) a cover letter explaining your motivation for the selected position and the FairCFD network, b) CV including information on previous experience with emphasis on publications and presentations if applicable, c) Academic transcripts and/or official certification demonstrating completion of, or current enrolment in, a Master’s degree (or equivalent), and d) 2 reference letters and/or contact details of 2 referees, and please send everything through the following link specifying« Application to DC3 » as the e-mail’s title.

We will start considering applications from February 1, and the process will remain open up to March 31.  After examination of the received applications, shortlisted candidates will be invited for online interviews with the recruitment committee.

If you are interested in more than one position belonging to the FairCFD network, please apply to 2 positions at most and precise the second choice in your cover letter. Extra applications not respecting this rule will be discarded.

Additional comments

A start date will be negotiated with the successful candidate. Ideally start dates would be between March 2026 and September 2026, with a potential to extend the start date to October 2026.