Civil Engineering Reference
In-Depth Information
The first (and, to our knowledge, to date the only)
publication that confirms the existence of a population of
HPVs intensively populating the turbulent boundary layer is
fairly recent [WU 09]. Wu and Moin [WU 09] used DNS to
simulate a turbulent boundary layer developing spatially
within the range of low Reynolds numbers
80
≤≤
Re
940
(
). Two important peculiarities of the DNS
performed by Wu and Moin [WU 09] must be emphasized.
First, their simulation relates to a boundary layer developing
in space and therefore does not use periodic boundary
conditions (frequently used in DNS for a channel flow, and
also in the simulations conducted by Spallart [SPA 88],
although differently). Second, [WU 09] inject parcels of
isotropic disturbances from the potential zone of
Re
τ
≤
300
Re
θ
=
80
onward, rather than artificially or at the wall. They attribute
the fact that few or no HPVs are observed in the above
studies, particularly in channel flows, to the periodic
boundary conditions: the streamwise periodicity reinjects
HPVs into the domain of computation; these HPVs interact
with the existing structures, causing distortion of the latter.
This hypothesis has yet to be validated. It is surprising, in
itself, that any modification at all of the boundary conditions
and the way in which the flow is disturbed causes so radical
a change in the morphology identified in the turbulent
boundary layer. The question posed in the title of this section
remains to be answered, at present.
4.3. Frequency distribution of energetic events in the
inner sublayer
In this section, we present the frequency distribution of
the ejections and shear layers identified by single-point
Eulerian techniques, outlined in the previous chapter. The
measurements used were obtained in the Laboratoire des
Ecoulements Géophysiques et Industriels (LEGI), Grenoble
hydrodynamic channel described in [TAR 94] at
.
Re
τ
=
500
Search WWH ::
Custom Search