Complex flow — Beyond canonical flow configurations, we address real-world problems using simulations, experiments, high-performance computing, and data-driven methods.

Systems — We consider not only the flow itself, but also its integration into larger physical and computational systems, connecting complex flows with open-source software and optimization frameworks.

We presented our recent research on reinforcement learning (RL) for delayed thermal flow control at General Motors AI Week. The presentation highlighted how RL can address complex flow-control problems involving delayed responses and time-dependent flow conditions. It also explored potential future collaborations with GM to investigate applications in electric vehicle (EV) cooling systems, where intelligent thermal management could help improve energy efficiency and system performance.

A collaborative research study lead by Christian Lagemann has been published in Nature. The study presents HydroGym, a computational platform that combines reinforcement learning (RL) and computational fluid dynamics (CFD) for the development of flow-control strategies. A particularly striking result of the study is that learned flow-control strategies can transfer to substantially different flow environments without retraining.

[PAPER]