A Mathematical Model-Based Optimization Method for Direct Metal Deposition of Multimaterials

Abstract

During the past few years, metal-based additive manufacturing technologies have evolved and may enable the direct fabrication of heterogeneous objects with full spatial material variations. A heterogeneous object has potentially many advantages, and in many cases can realize the appearance and/or functionality that homogeneous objects cannot achieve. In this work, we employ a preprocess computing combined with a multi-objective optimization algorithm based on the modeling of the direct metal deposition (DMD) of dissimilar materials to optimize the fabrication process. The optimization methodology is applied to the deposition of Inconel 718 and Ti–6Al–4V powders with prescribed powder feed rates. Eight design variables are accounted in the example, including the injection angles, injection velocities, and injection nozzle diameters for the two materials, as well as the laser power and scanning speed. The multi-objective optimization considers that the laser energy consumption and the powder waste during the fabrication process should be minimized. The optimization software modeFRONTIER® is used to drive the computation procedure with a matlab code. The results show the design and objective spaces of the Pareto optimal solutions and enable the users to select preferred setting configurations from the set of optimal solutions. A feasible design is selected which corresponds to a relatively low material cost, with laser power 370 W, scanning speed 55 mm/s, injection angles 15 deg, injection velocities 45 m/s for Inconel 718, 30 m/s for Ti–6Al–4V, and nozzle widths 0.5 mm under the given condition.