» Back to overview
Proposing Institution

Deutsches Zentrum für Luft- und Raumfahrt e.V. Göttingen
Project Manager

Christian Bauer
Bunsenstrasse 10
37073 Göttingen
Dynamics of large and very large coherent motions in turbulent pipe flow will be investigated by means of direct numerical simulations (DNS). Since a major fraction of energy needed to move fluids through pipes, channel and around vehicles, such as cars or planes, is dissipated by fluid turbulence in the vicinity of walls, a detailed understanding of wall-bounded, turbulent shear flows is of utmost importance to all engineering applications where Reynolds numbers are generally high. Within the last years very large-scale motions were found to contribute to the turbulent kinetic energy. However, a solid definition of their nature, a vivid characterisation of their evolution, and a profound understanding of their contribution to the overall energy household in the flow field is still missing. Since they become important at higher Reynolds numbers and exhibit large spatial lengths, our numerical set-ups cover Reynolds numbers based on the friction velocity up to 1500 within a flow domain of length 42 pipe radii to avoid interference of large-scale structures with the periodic computational domain. The simulation will be performed by our fortran90 based, MPI-parallel, inhouse DNS code flowsi involving FFTW3 and NetCDF libraries. We are solving the incompressible Navier-Stokes equations by means of a fourth-order finite volume discretisation in space and a second order Leapfrog-Euler scheme in time. The scheme is based on a projection method involving a direct Poisson solver. The aim of our work is to identify very large-scale motions and elucidate their dynamics by the aid of different techniques such as instantaneous flow field realizations, numerical particle tracking, turbulent statistics, one- and two-dimensional pre-multiplied velocity energy spectra, scale energy flux analysis through spatial filtering or lagrangian dynamic mode decomposition.

Impressum, Conny Wendler