- shm: have displacementMotionSolver as alternative mesh shrinker
(instead of medialAxis).
- updated iglooWithFridges tutorial to use displacementLaplacian
- selectable interpolation from cells to points in the motion solvers
using the 'interpolation' keyword:
interpolation volPointInterpolation; // default
or
interpolation patchCorrected (lowerWall upperWall);
- wrapped up mesh shrinkers (see above) for use as a displacementMotionSolver
(i.e. the opposite of the displacementMotionSolver mesh shrinker)
- shoot rays to nearest point on surface and two perpendicular rays
(instead of always shooting in the 3 coordinate directions)
- avoid bleeding through the surface intersection
1. multi-ray shooting. It now shoots rays in all the 3 coordinate directions
from the cell centre. Before it would shoot just a single ray from the
nearest point on the surface, going through the cell centre.
There is a cost overhead in that now it shoots 6 rays (+-x, +-y, +-z)
instead of just 1.
2. bleeding of refinement. It marks the cells inside a gap and walks out
the gap-size to neighbouring cells (which are just outside the gap). This
should make for a smoother refinement pattern.
The start of the layer addition loop does a synchronisation of the wanted
displacement. This also does a truncation of the displacement if it is <
minThickness. At the first iteration the displacement was initialised to
vector::one which might trigger the truncation logic (and then disable
extrusion altogether). Instead we now initialise the displacement to
vector::GREAT before entering the synchronisation.
Refinement:
-----------
// Optionally avoid patch merging - keeps hexahedral cells
// (to be used with automatic refinement/unrefinement)
//mergePatchFaces off;
// Optional multiple locationsInMesh with corresponding optional cellZone
// (automatically generates faceZones inbetween)
locationsInMesh
(
((-0.09 -0.039 -0.049) bottomAir) // cellZone bottomAir
((-0.09 0.009 -0.049) topAir) // cellZone topAir
);
// Optional faceType and patchType specification for these faceZones
faceZoneControls
{
bottomAir_to_topAir
{
faceType baffle;
}
}
/ Optional checking of 'bleeding' of mesh through a specifying a locations
// outside the mesh
locationsOutsideMesh ((0 0 0)(12.3 101.17 3.98));
// Improved refinement: refine all cells with all (or all but one) sides refined
// Improved refinement: refine all cells with opposing faces with different
// refinement level. These cells can happen on multiply curved surfaces.
// Default on, can be switched off with
//interfaceRefine false;
Snapping
--------
// Optional smoothing of points at refinement interfaces. This will reduce
// the non-orthogonality at refinement interfaces.
//nSmoothInternal $nSmoothPatch;
Layering
--------
// Layers can be added to patches or to any side of a faceZone.
// (Any faceZone internally gets represented as two patches)
// The angle to merge patch faces can be set independently of the
// featureAngle. This is especially useful for large feature angles
// Default is the same as the featureAngle.
//mergePatchFacesAngle 45;
// Optional mesh shrinking type 'displacementMotionSolver'. It uses any
// displacementMotionSolver, e.g. displacementSBRStress
// (default is the medial-axis algorithm, 'displacementMedialAxis')
//meshShrinker displacementMotionSolver;
Refinement:
-----------
// Optionally avoid patch merging - keeps hexahedral cells
// (to be used with automatic refinement/unrefinement)
//mergePatchFaces off;
// Optional multiple locationsInMesh with corresponding optional cellZone
// (automatically generates faceZones inbetween)
locationsInMesh
(
((-0.09 -0.039 -0.049) bottomAir) // cellZone bottomAir
((-0.09 0.009 -0.049) topAir) // cellZone topAir
);
// Optional faceType and patchType specification for these faceZones
faceZoneControls
{
bottomAir_to_topAir
{
faceType baffle;
}
}
/ Optional checking of 'bleeding' of mesh through a specifying a locations
// outside the mesh
locationsOutsideMesh ((0 0 0)(12.3 101.17 3.98));
// Improved refinement: refine all cells with all (or all but one) sides refined
// Improved refinement: refine all cells with opposing faces with different
// refinement level. These cells can happen on multiply curved surfaces.
// Default on, can be switched off with
//interfaceRefine false;
Snapping
--------
// Optional smoothing of points at refinement interfaces. This will reduce
// the non-orthogonality at refinement interfaces.
//nSmoothInternal $nSmoothPatch;
Layering
--------
// Layers can be added to patches or to any side of a faceZone.
// (Any faceZone internally gets represented as two patches)
// The angle to merge patch faces can be set independently of the
// featureAngle. This is especially useful for large feature angles
// Default is the same as the featureAngle.
//mergePatchFacesAngle 45;
// Optional mesh shrinking type 'displacementMotionSolver'. It uses any
// displacementMotionSolver, e.g. displacementSBRStress
// (default is the medial-axis algorithm, 'displacementMedialAxis')
//meshShrinker displacementMotionSolver;
The standard merge-algorithm is N^2 over the face-points and uses a
geometric proximity test for the merge. These are both choices for
implementation simplicity and are rather inefficient for large meshes.
I have now implemented an experimental linear topological merge
algorithm which is VERY fast and effective for meshes of any size.
Currently it will merge internal faces on meshes of arbitrary complexity
but does not yet handle edge or face collapse needed for wedges and
other degenerate blocks.
The new fast-merge algorithm may be selected using the optional
"fastMerge" entry:
fastMerge yes;
and if not present the standard N^2 algorithm will be used.
Henry G. Weller
CFD Direct
Consider a block describing a channel with two opposite walls.
Currently in order to grade the mesh towards the walls and have a
uniform region in the centre the channel would need to be spit into 3
blocks. With the new multi/sectional grading this can be achieved in a
single block e.g.
blocks
(
hex (0 1 2 3 4 5 6 7) (20 60 20)
simpleGrading
(
1
((0.2 0.3 4) (0.6 0.4 1) (0.2 0.3 0.25))
1
)
);
In this example the block is divided uniformly in the x and z -directions
and split into three grading sections in the y-direction described by
three triples: ((0.2 0.3 4) (0.6 0.4 1) (0.2 0.3 0.25)). Each of the
grading sections is described by a triple consisting of the fraction of
the block, the fraction of the divisions and the grading ratio (size of
first division/size of last division). Both the fraction of the block
and the fraction of the divisions are normalized automatically so they
can be specified scaled in anyway, e.g. as percentages:
blocks
(
hex (0 1 2 3 4 5 6 7) (20 60 20)
simpleGrading
(
1
((2 3 4) (6 4 1) (2 3 0.25))
1
)
);
and they need not sum to 1 or 100.
This is very new functionality and not well tested but backward
compatibility has been well tested so all existing blockMeshDicts should
parse correctly.
To compile with 64bit labels set
WM_LABEL_SIZE=64
in ~/OpenFOAM/dev/prefs.sh
source ~/.bashrc
then Allwmake in OpenFOAM-dev.
This will build into for example OpenFOAM-dev/platforms/linux64ClangDPInt64Opt
If WM_LABEL_SIZE is unset or set to 32:
WM_LABEL_SIZE=32
the build would be placed into OpenFOAM-dev/platforms/linux64ClangDPInt32Opt
Thus both 32bit and 64bit label builds can coexist without problem.
