Topology optimization of a flap track support for additive manufacturing

Topology optimization with Solid Isotropic Material with Penalization (SIMP) method is a mature structural optimization formulation that has been commercially used for years. Great innovations were made through the years to... [ view full abstract ]

Topology optimization with Solid Isotropic Material with Penalization (SIMP) method is a mature structural optimization formulation that has been commercially used for years. Great innovations were made through the years to incorporate manufacturing constraints within the formulation so it can be used for the design of machined or casted parts. However, the recent emergence of additive manufacturing technologies with metallic materials allows the aerospace industry to explore further topological optimization of small parts with much less manufacturing constraints. Through the case study of a flap track support in the trailing edge of a business aircraft, the possibilities and limitations of designing with topology optimization for additive manufacturing are explored.

First, the sensitivity of the optimization results to setup parameters and design space is studied. A lot of optimization parameters leads to a lot of different results. It can be subjective to compare one to each other by only relying on a visual analysis and the experience of the user. Here, these results are instead compared based on the displacement of specific nodes of the model and load distribution through the model. Other numerical limitations of the optimization such as the buckling factor permits to leave out some results. In addition, the modeling of the design space is a key aspect of topology optimization and its impact on the results and the numerical stability through the use of minimum member size constraint will be exposed.

Results interpretation is complex and often based on the experience of the designer. Indeed, the optimization result is a mesh with relatively coarse features depending on the size of the mesh. In a lot of cases the coarseness of those features can be misleading when they are interpreted as is. The experienced stress engineer who performs the topology optimization study must rationalize the results before interpretation is made by a designer. These modifications should be made directly on the mesh file (STL) to simplify this step. However, almost no mesh manipulating tools exists in the optimization software (Altair® Hypermesh) or the CAD software (CATIA®).

The free and open source video game engine Blender®, a mesh-based modeling software with really powerful free-form tools, is used here to this aim. Several sculpting manipulations can be performed on the optimization results to smooth the surfaces, fill checkerboard areas or erase numerical artifacts. It is used here as a preliminary step before the CAD final design.

To validate the usefulness of the smoothing process, a direct interpretation of the optimization results is compared to an interpretation of the results after being processed by stress engineer with sculpting tools in Blender®. Both net-shape parts are modeled using a CAD software and their stiffness and displacement are compared to validate each model.