Skip to main content
Go to the home page of the European Commission (opens in new window)
English en
CORDIS - EU research results
CORDIS

Entropy-driven model reduction for fluid flows

Project description

Improving turbulence models

Turbulent flows affect wind farms, industrial processes and even the climate, but it is impossible to compute every swirling eddy. Instead, engineers use ‘closure models’ to approximate the impact of the unseen turbulence. The ERC-funded SYMBIOSIS project aims to build closure models around the concept of entropy. Drawing on both entropy’s physical meaning, linked to stability, and its information-theoretic role in measuring uncertainty, SYMBIOSIS will create probabilistic, entropy-stable simulations of fluid flow. The framework combines model reduction, stochastic equations and data assimilation to update predictions with real-world observations. Applied to wind-turbine wakes and CO2 transport, the approach could redefine how scientists simulate and manage complex turbulent systems.

Objective

Turbulent fluid flows play a critical role in science and engineering in energy, process design, climate and life sciences. Accurate and efficient computational fluid dynamics (CFD) simulation tools are therefore essential. As resolving all turbulent scales of motion is computationally infeasible, every CFD code requires a closure model to represent the effect of unresolved scales. A recent paradigm shift is to construct closure models by combining physics- and data-driven methods. However, these lack the cornerstones of mathematical-physical modelling of fluid flows: stability, physical consistency and accuracy guarantees.

In SYMBIOSIS I develop a radically novel approach to construct closure models. I propose entropy as concept to bridge physics- and data-driven models: the mathematical-physical concept of entropy is used to enforce stability and physical consistency, while the information-theoretic concept of entropy provides a new view on measuring accuracy. Together they provide an innovative framework for accurate probabilistic predictions of turbulence that I apply to simulate wind-turbine wakes and CO2 transport.

Both entropy concepts are combined in three steps to construct new closure models. 1) I propose a new entropy-stable model reduction strategy, providing the environment for effective development of closure models. 2) I propose a new entropy-stable stochastic formulation of the fluid flow equations, providing a powerful way to model uncertainties arising from initial conditions, fluid properties, and neglected small scales. 3) I propose a new entropy-stable data-assimilation strategy to enable continuous updating of the closure model and flow field with observation data. The result is a new foundational mathematical framework for simulating turbulent flows with ground-breaking implications for physics- and data-driven modelling.

My track record of leadership, creativity and expertise positions me to successfully lead this highly ambitious project.

Fields of science (EuroSciVoc)

Keywords

Programme(s)

Topic(s)

Funding Scheme

HORIZON-ERC - HORIZON ERC Grants

See all projects funded under this funding scheme

Call for proposal

(opens in new window) ERC-2025-COG

See all projects funded under this call

Host institution

STICHTING NEDERLANDSE WETENSCHAPPELIJK ONDERZOEK INSTITUTEN
Net EU contribution
€ 2 000 000,00
Total cost
€ 2 000 000,00

Beneficiaries (1)