Our group develops advanced data science methods to improve predictions and deepen understanding of turbulent flows and other multi-scale physical systems. Currently we work on three different research areas.
Data-Driven Turbulence Modeling
Turbulence remains one of the last unsolved problems in classical physics, spanning sub-meter to planetary scales in both natural and engineered systems. Current simulations rely on Reynolds-Averaged Navier-Stokes equations with closure models — an approach that has seen little progress in decades. We propose data-driven, physics-informed methods to address this challenge across all scenarios of data availability.
Generative Modeling of Dynamic Systems
Fast, time-resolved prediction of turbulent flows in complex environments — such as urban canopies or wind farms — remains a major challenge. High-fidelity approaches like DNS and LES are often prohibitively expensive and rely on uncertain inputs. We investigate how state-of-the-art generative AI models can be adapted for scientific modeling to enable reliable, scalable turbulence prediction with quantified uncertainty.
Subsurface flows
A fundamental challenge in the research of geologic subsurface flow is to upscale laboratory-measured, core-scale rock and flow properties to the field scale. We aim to combine physics with data analytics to address this fundamental challenge by developing novel machine-learning and data-assimilation based frameworks.
