An Optimization Approach for Components Built by Fused Deposition Modeling with Parametric Internal Structures

Abstract

Additive manufacturing processes are employed to create physical models from three-dimensional (3D) computer-aided design (CAD) math data. A solid model or water-tight surface model is used as the input, which is sliced into layers, and travel paths are created for each layer. The object is built by layer by layer stacking, with supporting structures for overhanging geometry and undercuts being created where necessary (process dependent). Fused deposition modeling (FDM) is an additive fabrication process that builds a part from extruded filaments of a melted thermoplastic. Several studies have focused on the depositing parameters; however, none of them have characterized internal support structures in different geometrical arrangements. The incorporation of reconfigurable parametric internal matrix structures based on primitive elements will balance the mechanical properties, the material usage and the build time. Parametric internal structures are designed, and compressive test components built and tested both experimentally and using simulation tools to depict the compressive characteristics. Extensive physical testing is done as the components built by the FDM process have anisotropic properties. The material usage, build time, and loading characteristics are captured for a variety of parametric structures (solid, shell, orthogonal, hexagonal, pyramid) build orientations, and internal densities (loose, compact). From this data, a model is developed that serves as a predictive tool to: (i) estimate the mechanical properties and (ii) calculate the build time and materials utilized based on various internal structural configurations for the component's application. A model that generates an optimal solution (minimum material, minimum build time, etc.) needs to be developed. Using the collected data as a foundation, an optimization model that considers the build time, material usage, surface finish, interior geometry, strength characteristics, and related parameters is presented and can be used to assist designers making informed decision with respect to strength, material usage and time, etc. is developed using the Genetic Algorithm approach.