DeepTurb - Deep Learning in and of Turbulence

The application of machine learning (ML) techniques in the analysis of experimental measurements and numerical simulations of turbulence opens unique possibilities to analyse complex and comprehensive data sets by new physical criteria and thus to gather a deeper understanding of the fundamental transport processes in such flows. Our project aims at new effective modeling strategies of turbulent superstructures in extended turbulent convection flows - gradually evolving large-scale patterns - by means of machine learning techniques. We want to accelerate the analysis in optical flow measurements, develop low-dimensional reduced models that can predict the coarse-scale dynamics, and extend the mathematical foundations of ML applications to obtain a more efficient prediction of these processes.

Project PIs

Prof. Christian Cierpka (Mechanical Engineering)
Prof. Patrick Mäder (Computer Science)
Prof. Karl Worthmann (Mathematics)
Prof. Jörg Schumacher (Mechanical Engineering)

PhD students and Postdoctoral Researchers

Florian Heyder (Reservoir Computing in Convection)
Johannes Viehweg (Recurrent Neural Networks)
Philipp Teutsch (Long Short-Term Memory Networks)
Mohammad S. Ghazijahani (Machine Learning Analysis of Experimental Data)
Theo Käufer (Experiments of Turbulent Convection)
Philipp Pfeffer (Quantum Machine Learning
Dr. Friedrich Philipp (Mathematical Foundations of Machine Learning)

Related Publications