Abstract

The necessity of manually converting discrete geometry to computer-aided design (CAD) is one of the challenges preventing the adoption of generative design and additive manufacturing. The output data type of most topology optimization software programs is a triangular surface mesh tessellation, typically in the stereolithography (STL) file format. However, for various product lifecycle management (PLM) activities, a smooth boundary representation that can be modeled in traditional CAD software is often preferred or required. Manually creating a CAD model from this discrete triangular surface mesh data requires hours of tedious labor for each part and introduces uncertainty and error into the workflow. There are also other areas outside of topology optimization and additive manufacturing where this model workflow is a problem. One is as-built-to-analysis workflows such as sequential process modeling, where inter-process impacts (e.g. forging impacts on machined part warping) require a multi-step simulation process, often transitioning between different software programs. Another is reverse engineering, or creating a CAD geometry from inspection scan data. All of these are long-term necessities for implementation and adoption of a model based enterprise and digital twin, and their associated efficiency and efficacy gains. Industry has responded to this demand with the creation of various tools to improve the level of automation in reconstructing CAD geometry from discrete triangular surface mesh tessellation data. In this study, several benchmark tessellations were created, either via optimization, sequential process modeling, or inspection scanning. Multiple software tools were then used to reconstruct CAD geometry from the discrete data of the benchmark tessellations. Speed and difficulty of the process were assessed, along with comparing both the geometrical accuracy and functional performance differences between the original tessellated design and the reconstructed CAD geometry. The Department of Energy?s Kansas City National Security Campus is operated and managed by Honeywell Federal Manufacturing & Technologies, LLC under contract number DE-NA0002839.