Research at the ISTM

The research at the Institute of Fluid Mechanics combines theoretical, numerical and experimental methods of fluid mechanics. The Institute's research activities include the characterization, prediction and control of fluid flows. One focus is on the treatment of turbulent flows and their selective manipulation with regard to efficient energy use.

Experimental Fluid Mechanics

The ISTM has several wind tunnels in which, for instance, the influences of complex roughness on fluid flows are investigated and also mechanisms for passive and active flow control are tested. Furthermore, we investigate accelerated and rotating flows in specially developed test facilities.

Pressure sensors, hot-wire anemometers, laser Doppler anemometry and particle image velocimetry are used to measure fluid flows.

Flow Simulation

At the ISTM a large number of numerical simulations of fluid flows are performed. A special focus is on the simulation of turbulent flows by means of direct numerical simulation. This includes the investigation of turbulent flows over complex roughness and their heat transfer properties, the heat transport in pipe flows and the passive and active flow control of turbulent flows.

Another research area deals with novel strategies for friction reduction between lubricating surfaces using numerical experiments. Furthermore, numerical simulations allow us to gain detailed insights into the flow properties of the heart that are not covered by experimental measurements.


High-Performance Computing & Data Visualization

The computational effort of simulations of turbulent flows is numerically very complex, since many flow problems require several million computation points and at the same time the flows must be tracked over a long period of time. This requires the use of high-performance computing and efficient evaluation routines for large amounts of data. The ISTM has access to the high-performance computer systems at the Steinbuch Center for Computing (SCC) and performs most of the simulation on the bwUniCluster, ForHLR I and II.