The assembly meshing process involves the following approach:
Size functions and virtual bodies (optional) are defined.
Note: You can define virtual bodies to represent flow volumes if your model does not contain a “solid” that represents the fluid region. In this way, virtual bodies allow you to mesh fluid regions without having to use the DesignModeler application or another solid modeler to model them. The Fluid/Solid material property for virtual bodies is always set to Fluid (read-only).
The initial size of the Cartesian grid is computed based on the minimum and maximum size set for the size functions.
A uniform Cartesian grid is created within the bounding box for the geometry.
The size function values are computed and the grid is then adaptively refined based on the local size function values.
Figure 182: Mesh After Refinement shows the mesh after refining the initial grid based on size functions.
The cells intersected by the geometry are marked. Only nodes on marked cells are considered for projection. The nodes are projected to the geometry (corner, edge, and face in order of reducing priority).
Figure 183: Mesh After Projection shows the mesh after node projection.
The edges intersected by the geometry are identified. Mesh edges to be preserved/recovered are determined, and are used to construct mesh faces. Once the mesh faces are identified, cells are decomposed to recover these faces. The cells are decomposed based on a number of templates.
The mesh outside the fluid bodies is removed. The mesh inside the solid bodies may also be removed for assembly meshing, depending on the setting of the Keep Solid Mesh control.
The quality of the cells thus generated is improved.
Cells are separated into cell zones based on the respective virtual bodies. A cell included in multiple virtual bodies will be included with the body having the highest priority. The smallest body, based on the size of the bounding box enclosing it, has the highest priority to guarantee that any voids are recovered properly.
Figure 184: Cells Separated After Decomposition shows the cells separated into respective cell zones after decomposition.
The boundary mesh is recovered and separated based on the underlying geometry.
Faces whose adjacent neighboring cells are in different cell zones automatically constitute the boundary mesh.
The neighboring cells of a face on an internal baffle are in the same cell zone. In such cases, faces close to and nearly parallel to the baffle surface are recovered to represent the baffle surface.
As each cell zone is a closed region, the mesh boundary is conformal.
The boundary zone types are assigned based on the Named Selections defined on the (underlying) geometry faces.
Note: Named Selection names for internal face zones are not interpreted. In cases where two enclosed voids share a face, the face zone is assigned type WALL automatically regardless of whether a Named Selection has been defined for the face. In these cases, the mesh generation cannot cross any boundary so you must define a virtual body with material point for each flow volume void in order for the volumes to be meshed.
Figure 185: CutCell Mesh After Boundary Recovery shows the CutCell mesh after the boundary mesh is recovered.
When the mesh is exported to ANSYS Fluent, a cell zone type of either FLUID or SOLID is assigned to each body based on its material properties. Refer to Fluent Mesh Export for details.
For information about material properties and which faces are selected to be inflation boundaries when Program Controlled inflation is defined for assembly meshing, refer to Program Controlled.
If the Tetrahedrons assembly meshing algorithm was selected, tet meshing extensions are applied. The Tetrahedrons algorithm starts with the CutCell mesh and transforms it into an unstructured tet mesh.