Figure. FEA model: the domain defined by the triangulated surface has to be meshed; then materials and boundary conditions have to be added.
The final optimization result, typically delivered by ProTOp, is an
In the latter case, one can take advantage of the fact that this surface is closed so that it actually represents the boundary surface of the optimized part. Therefore, this surface can be the input surface used
To build a FEA model one needs an adequate CAD/CAE software that is capable of performing the following procedure.
This procedure returns a numerical FEA model. By using this model, detailed verification finite element analyses can be performed in order to check the optimized part against any possible conditions or situations.
Figure. FEA model: the domain defined by the triangulated surface has to be meshed; then materials and boundary conditions have to be added.
The benefits of this procedure are as follows:
High geometric accuracy is guaranteed since the FEA model matches exactly with the triangulated surface exported by ProTOp. Thus, the only geometrical errors introduced are the one accumulated during the surface smoothing procedure in ProTOp.
On the negative side, the most important drawback is the lack of a CAD model.
To reconstruct a CAD model one needs an adequate CAD/CAE software that is capable of performing advanced CAD and reverse engineering procedures. This process is typically more difficult than numerical FEA model retrieval, and a variety of approaches can be used. To illustrate the available options, two quite distinct procedures are presented briefly in the following. These two procedures are:
In this procedure the original CAD model of the optimized part is used as a point of departure. This CAD model has to be enriched by importing the triangulated surface of the optimized part. The surface has to be imported as a non-structural part.
Figure. The original CAD model with the triangulated surface imported as a non-structural part.
The optimized geometry has to be made visible to identify the regions where material should be removed. Then, the corresponding material (domain) removal from the original CAD model can be done manually by using the sketching features.
Figure. The sketching features can be used to remove the redundant material.
After the surplus of material is removed, the result is a CAD model of the optimized part with preserved original:
Figure. A CAD model of the optimized part with preserved original materials and boundary conditions.
The benefits of this procedure are as follows:
On the negative side, the most important drawback is the difficulty of manual material removal. For more complicated designs, like lattice structures, this becomes even practically impossible.
In this procedure the original CAD model is not used. Instead, the triangulated surface of the optimized part is used for CAD geometry reconstruction. If a free-style modeler is engaged, this typically means that the number of triangles of the considered surface may not be excessively large, or better said, it has to be rather low.
The triangulated surface with substantially reduced triangles count is typically used by the modeler as a mesh of control points which define some geometrical objects, for example, B-spline patches. If the triangulated surface is error free and closed, the surface patches define the boundary surface of a solid body. In this case we have a CAD model whose geometry can be adjusted by adjusting the positions of the underlying control points.
Figure. Freestyle CAD model: the restored geometry may differ significantly from the optimized one; this may induce stress concentrations.
The benefits of this procedure are as follows:
On the negative side, the most important drawbacks are the difficulty of preserving accurate geometry of the optimized part and the need to redefine materials and boundary conditions for eventual FEA.