Research Topic

Numerical Simulation of turbulent flow in context of FSI



In many industrial fields the calculation and prediction of turbulent flows is a very important task. In the last sixty years simulations were used as an aid to physical experiments. They offer the possibility to predict many different kinds of flows faster and cheaper than this is possible with expensive experiments. However the current computing power is not sufficient for a complete resolution of turbulent structures, like they are used in so-called Direct Numerical Simulation - DNS, in case of large Reynolds Numbers or complicated geometries. Therefore further methods were developed. Although they use some simplifications, they allow sufficiently exact predictions of the behavior of turbulent flows. A very common model for the simulation of turbulent flows is the so-called Reynolds Average Navier-Stokes (RANS) model. The computational costs are low in comparison to DNS. However the delivered results are often not satisfactory. Another possibility to calculate turbulent flows is the Large Eddy Simulation (LES). But the computational costs increase very quickly with an increasing Reynolds number. This leads to high computing time. That's why in the last ten years in the field of simulation of turbulent flows so-called hybrid turbulent models became increasingly popular. They combine advantages of the above described turbulent models. For example, the so called Very Large Eddy Simulation (VLES) switches between RANS and DNS depending on the numerical resolution. Between these two limits a LES will be recovered. The new VLES is able to give good predictions on a wide range of turbulent flow simulations on a coarse mesh in comparison to the LES.


So called Fluid-Structure Interaction (FSI) problems plays an essential role in many engineering fields. The FSI multiphysics interface combines fluid flow with solid mechanics to capture the interaction between the fluid and the solid structure. Here the geometry is deformed according to the modification of structure and then the new fluid iteration is performed.

Nowadays in the research of the FSI - problems mostly algorithms of coupling and moving of grid are investigated. The turbulent FSI is currently not an established research object. The science works on this topic are mostly specialized on applications and the relationships between the hybrid models in the context of fluid-structure coupling problems are hardly investigated. But especially in the context of fluid-structure interaction problems (FSI) the problem of computational cost becomes much more significant. Therefore the use of hybrid turbulence modeling is particularly useful.

This work focuses on the investigation of turbulent flow problems with moving mesh and in fluid-structure coupled systems.

Key Research Area

Turbulence, Numerical Simulation of Coupled Multi-physical Problems


Anastasia Kondratyuk


Dolivostraße 15

D-64293 Darmstadt



+49 6151 16 - 24374


+49 6151 16 - 24404




kondratyuk (at) gsc.tu...

 Print |  Impressum |  Sitemap |  Search |  Contact |  Privacy Policy
zum Seitenanfangzum Seitenanfang