Generative AI and Transfer Learning for Turbulence Modelling in Large-Eddy Simulation

We invite applications for a fully funded PhD position focused on next-generation subgrid-scale (SGS) modelling for Large-Eddy Simulation (LES), combining generative artificial intelligence with physics-based turbulence modelling. Accurate SGS closures remain a major bottleneck for predictive LES of complex turbulent flows, particularly in aeronautics, energy systems, and environmental applications. While classical models have seen limited progress, recent advances in generative AI, especially Generative Adversarial Networks (GANs), offer unprecedented opportunities to reconstruct multiscale turbulent dynamics and learn physically consistent representations from data. The PhD project aims to develop a hybrid LES closure that couples GAN-based super-resolution reconstruction with established physical models to predict SGS stresses and scalar fluxes. Building on successful results in homogeneous isotropic turbulence, the research will extend these methods to wall-bounded flows (channels and boundary layers) and free shear flows (mixing layers and jets). A central focus will be generalisation, ensuring robustness across Reynolds numbers, geometries, and filter conditions. The project will also exploit transfer learning to create reusable, multi-flow foundational models that can be efficiently adapted to specific applications using limited data. The student will work with high-quality DNS datasets already available within the group and collaborators, and will have access to national and European high-performance computing facilities. The work includes a priori and a posteriori LES validation, stability and robustness analysis, and integration into practical LES frameworks. Candidate profile: We seek highly motivated candidates with a strong background in fluid mechanics, turbulence, or computational physics/engineering. Experience with numerical simulation, data-driven methods, or machine learning is desirable but not mandatory; training will be provided. This PhD offers an excellent opportunity to work at the interface of turbulence physics and modern AI, addressing a problem of high scientific and industrial relevance.

Turbulence and Reactive Flow Simulation Laboratory

Cornell University - Dynamic mixed turbulence modeling using a super-resolution generative adversarial approach

The funding is available to Home applicants only (UK and EU settled/pre-settled).

The funding is available to Home applicants only (UK and EU settled/pre-settled).

This is a competitive funding opportunity. The application process involves two interviews – an initial technical interview performed by the project supervisors (to be completed before 30th January), and a second competitive interview with a panel in either the weeks commencing 16th or 23rd February 2026.

Further information and other funding options.