including blockMeshDict, surfaceFeaturesDict and snappyHexMeshDict, based on the
case surface geometry.
Description
Preconfigures blockMeshDict, surfaceFeaturesDict and snappyHexMeshDict
files based on the case surface geometry files.
Starting from a standard OpenFOAM case, this utility locates surface
geometry files, e.g. OBJ, STL format, in the constant/geometry directory.
It writes out the configuration files for mesh generation with
snappyHexMesh based on assumptions which can be overridden by options on
the command line.
The utility processes the surface geometry files, attempting to anticipate
their intended purpose, trying in particular to recognise whether the
domain represents an external or internal flow. If there is a surface
which is closed, and is either single or surrounds all other surfaces,
then it is assumed that it forms the external boundary of an internal
flow. This assumption is overridden if the bounds of the background mesh
are specified using the '-bounds' option and they are more than 50% larger
than the surface bounds.
Surfaces which form boundaries of the domain may contain named regions
that are intended to become patches in the final mesh. Any surface region
whose name begins with 'inlet' or 'outlet' will become a patch of the same
name in the final mesh. On an external surface (for an internal flow),
regions can be identified as inlets and outlets using the '-inletRegions'
and '-outletRegions' options, respectively. When either option specifies a
single region, the resulting patch name will be specifically 'inlet' or
'outlet', respectively. Surfaces which are contained within the domain,
which do not surround or intersect other surfaces, are assumed by default
to be wall patches. Any closed surface which surrounds another (but not an
external surface) is used to form a cellZone within the mesh. Any surface
can be specifically identified as a cellZone with the '-cellZones' option,
with the additional '-baffles' and '-rotatingZones' options available to
assign a surface to a more specific use.
The background mesh for snappyHexMesh is a single block generated by
blockMesh, configured using a blockMeshDict file. The block bounds are
automatically calculated, but can be overridden by the '-bounds'
option. The number of cells is calculated to produce a fairly small
prototype mesh. The cell density can be overridden by the '-nCells' option
or can be scaled up by an integer factor using the '-refineBackground'
option. When the background mesh is required to form patches in the final
mesh, e.g. for an external flow, the user can specify the names and types
of the patches corresponding to the six block faces using options such as
'-xMinPatch', '-xMaxPatch', etc. The name and type of the default patch,
formed from block faces which are not configured, can also be specified
with the '-defaultPatch' option. The utility provides placeholder entries
for all block faces unless the '-clearBoundary' option is used. A special
'-cylindricalBackground' option generates a cylindrical background mesh,
oriented along the z-axis along x = y = 0.
The snappyHexMesh configuration is generated automatically, applying a set
of defaults to the main configuration parameters. By default, explicit
feature capturing is configured, for which a surfaceFeaturesDict file is
written for the user to generate the features files with the
surfaceFeatures utility. Implicit feature capturing can alternatively be
selected with the '-implicitFeatures' option. Refinement levels can be
controlled with a range of options including: '-refinementLevel' for the
baseline refinement level; '-refinementSurfaces' for levels on specific
surfaces; '-refinementRegions' for levels inside specific surfaces;
'-refinementBoxes' for quick, box-shaped refinement regions specified by
min and max bounds; '-refinementDists' for distance-based refinement; and
'-nCellsBetweenLevels' to control the transition between refinement
levels. A '-layers' option specifies additional layers of cells at wall
boundaries. The insidePoint parameter is set to '(0 0 0)' by default but
can be overridden using the '-insidePoint' option.
If the code string is delimited by '#{...#}' multiple lines and multiple code
statements can be used to generate the entry using 'os << ...;'. This is
equivalent to #codeStream but with a more compact syntax, e.g.
maxAngle 30;
nAngles 7;
Us #calc
const vector U($<vector>testCalc2!U);
const int nAngles = $nAngles;
const scalar angleStep = ($<scalar>maxAngle)/(nAngles - 1);
List<vector> Us(nAngles);
for(int i=0; i<nAngles; i++)
{
const scalar angle = degToRad(i*angleStep);
Us[i] = transform(Ry(angle), U);
}
os << Us;
Note the 'os << Us;' statement which writes the data to the dictionary entry in
the same manner as #codeStream, this provides flexibility on how the data is
created and written.
This allows #FOAM_CASE for example to be used in #calc variable lookup, e.g. in
test/dictionary/testCalc:
// assuming the testCalc2 file is local
magU #calc "mag($<vector>testCalc2!U)";
// finding the testCalc2 file using $FOAM_CASE
k #calc "1.5*magSqr(0.05*$<vector>{${FOAM_CASE}/testCalc2!U})";
This provides a smooth solution but it is not clear if this is more accurate
than running the cellMomentum p-U algorithm which generates complex transients
in the solution.
Also provide fields and curFields functions which return the list of registered
fields not including the geometryFields and current registered fields not
including the geometryFields or old-time fields to simplify mapping code for
NCC, mesh-to-mesh mapping and load-balancing.
Class
Foam::functionEntries::calcEntry
Description
Compiles and executes code string expressions,
returning the result to the dictionary
Usage
\verbatim
a 1.1;
b 3.2;
c #calc "$a*$b";
\endverbatim
Special care is required for calc entries that include a division since
"/" is also used as the scoping operator to identify keywords in
sub-dictionaries. For example, "$a/b" expects a keyword "b" within a
sub-dictionary named "a". A division can be correctly executed by using a
space between a variables and "/", e.g.
\verbatim
c #calc "$a / $b";
\endverbatim
or bracketing the variable, e.g.
\verbatim
c #calc "($a)/$b";
\endverbatim
The code string can be delimited either by '"..."' in which newlines must be
escaped or '#{...#}' which directly supports multi-line strings and is more
convenient when evaluating string expressions by avoiding the need to
escape the quotes, e.g.
\verbatim
c #calc #{ $a*sqrt($b) #};
\endverbatim
\verbatim
s "field";
fieldName #calc #{ $<string>s + "Name" #};
\endverbatim
Additional include files for the #calc code compilation can be specified
using the #calcInclude entry, e.g. if functions from transform.H are used
\verbatim
angleOfAttack 5; // degs
angle #calc "-degToRad($angleOfAttack)";
#calcInclude "transform.H"
liftDir #calc "transform(Ry($angle), vector(0, 0, 1))";
dragDir #calc "transform(Ry($angle), vector(1, 0, 0))";
\endverbatim
Note:
Internally this is just a wrapper around codeStream functionality - the
#calc string is used to construct a dictionary for codeStream.
See also
Foam::functionEntries::calcIncludeEntry
Lagrangian's dependency set is simpler than it used to be. There is no
longer a need to maintain a separate library for models that depend on
the momentum transport modelling.
Description
Specify an include file for #calc, expects a single string to follow.
For example if functions from transform.H are used in the #calc expression
\verbatim
angleOfAttack 5; // degs
angle #calc "-degToRad($angleOfAttack)";
#calcInclude "transform.H"
liftDir #calc "transform(Ry($angle), vector(0, 0, 1))";
dragDir #calc "transform(Ry($angle), vector(1, 0, 0))";
\endverbatim
The usual expansion of environment variables and other constructs
(eg, the \c ~OpenFOAM/ expansion) is retained.
See also:
Class
Foam::functionEntries::calcEntry
Description
Uses dynamic compilation to provide calculating functionality
for entering dictionary entries.
E.g.
\verbatim
a 1.0;
b 3;
c #calc "$a*$b";
\endverbatim
Note the explicit trailing 0 ('1.0') to force a to be read (and written)
as a floating point number.
Special care is required for calc entries that include a division since
"/" is also used as the scoping operator to identify keywords in
sub-dictionaries. For example, "$a/b" expects a keyword "b" within a
sub-dictionary named "a". A division can be correctly executed by using a
space between a variables and "/", e.g.
\verbatim
c #calc "$a / $b";
\endverbatim
or "()" scoping around the variable, e.g.
\verbatim
c #calc "($a)/$b";
\endverbatim
Additional include files for the #calc code compilation can be specified
using the #calcInclude entry, e.g. if functions from transform.H are used
\verbatim
angleOfAttack 5; // degs
angle #calc "-degToRad($angleOfAttack)";
#calcInclude "transform.H"
liftDir #calc "transform(Ry($angle), vector(0, 0, 1))";
dragDir #calc "transform(Ry($angle), vector(1, 0, 0))";
\endverbatim
Note:
Internally this is just a wrapper around codeStream functionality - the
#calc string is used to construct a dictionary for codeStream.
The dictionary/table-based expansions provided by stringOps::expand are
now too situation specific for this utility to concisely test with a
single complex string. These expansions are fundamental to the operation
of dictionary and dynamic code and receive good testing coverage in the
tutorial test loop anyway.
Simplifications have been made where possible, as permitted by the new
$<type>var syntax. Duplication has been reduced in similar blockMesh
files (e.g., sloshingTank cases). Settings that cannot practically be
changed have been hard-coded (e.g., angle in the mixerVessel2D
blockMeshDict). The rotor2D blockMeshDict has been centralised and
extended to work with an arbitrary number of rotor blades.
This makes the block/edge/face configuration much more similar between
the four different sections of this mesh. It is also useful as it
permits sections to be decativated by commenting them out without
this affecting all the subsequent numbering.
The purpose of these operations was unclear, and there was no
documentation or examples of their usage. The differences between these
operations behaviours for scalar and vector input seemed arbitrary.
These operations have in some cases become the subject of confusion.
They have therefore been removed.
Equivalent functionality could be easily reinstated as and when a clear
need and application becomes apparent.
