Logo der Physikalisch-Technischen Bundesanstalt

Machine Learning meets High-fidelity numerical combustion simulations

Kolloquium der Abteilung 8

Combustion Science is in a transitional phase: hydrogen is a promising fuel for sustainable, carbon-free energy generation. However, this apparently simple fuel brings a range of challenges for fundamental research and industrial use. In this talk, I will present recent progress of our group on the importance of high-fidelity numerical simulations combined with Machine Learning techniques towards physical understanding of hydrogen combustion. Currently, we address combustion instabilities, boundary layer-induced flashback and flame-wall interactions - all of these unsteady phenomena pose problems in low-emission combustion systems. 

Links:
metrology.webex.com/metrology/j.php