TRANSDIFFUSE is a HORIZON-ERC-SYG project awarded under the ERC-2024-SyG call. As a Synergy Grant, it brings together the numerical modelling expertise of Universidad Politécnica de Madrid, the high-fidelity computational power of the Barcelona Supercomputing Center, and the experimental innovation of Purdue University.TRANSDIFFUSE aims to reduce the environmental impact of aviation by harnessing transonic flows emerging from the combustor.
The project tackles formidable aerodynamic challenges and focuses on developing compact transonic fluid machinery for hydrogen pressure gain combustion (PGC) engines—critical components for enabling long-range, high-payload, zero-emission flights. Preliminary analyses suggest the potential to improve overall thermal engine efficiency by an impressive ten percentage points. The researchers' approach is centered on the development of a disruptive, AI-driven model named FluidGPT. This intelligent system is designed to characterize, predict, and manipulate complex transonic flow instabilities—overcoming a long-standing barrier in the design of efficient turbomachinery passages for compact engines.
FluidGPT will integrate the Navier-Stokes equations under these conditions, enabling the prediction and control of previously unmanageable flow behaviors.TRANSDIFFUSE will progress through structured phases: beginning with experimental and computational investigations of unsteady transonic turbine phenomena; followed by the development of FluidGPT, informed by the physics uncovered in Phase I; and finally, applying FluidGPT to devise novel flow control strategies and capitalize on flow instabilities.
Principal Investigator, Company and Country
Eusebio Valero, Universidad Politécnica de Madrid, Spain