Approach to an Automated Method for Load-Optimized Design of Multimaterial Joints for Additive Manufacturing

The emerging multimaterial technology is extending the potential applications of additive manufacturing in many areas. However, these possibilities also bring new challenges such as creating a sufficiently strong bond between materials and ensuring their continued recyclability. This paper presents a method for designing structures for joining different materials based on the use of experimentally determined values and artificial intelligence. The objective of the method is to find, for a given design space, an arrangement of the various materials that provides a high strength of the material composite in one loading direction and, at the same time, is designed to result in a significantly reduced strength for a further loading direction. As a result, the material composite should break at this predetermined breaking point when loaded in the second direction, so that the materials involved can be recycled. To be able to perform the calculation as quickly as possible, the geometry is reduced to arrangements of small cubes (voxels). Four steps are provided for the use of this method. In the first step, the maximum resolution that can be achieved with the respective additive process is determined. In a further step, different arrangements of voxels are examined for resulting strength using tensile tests. In the next step, the results of these tests serve as input values for an AI application that finds an arrangement of the voxels that provides the desired strengths in the various load directions, taking into account a given design space. A genetic algorithm is used to geometrically optimize the joint. Finally, these designs are used to automatically build a CAD model that enables additive manufacturing of the components. Initial investigations into the voxel sizes and manufacturability of the multi-material joints using the material extrusion (MEX) process are presented, but evaluation of the overall method is still pending.
Klahn, Christoph (Hrsg.) ; Meboldt, Mirko (Hrsg.) ; Ferchow, Julian (Hrsg.) (2023): Industrializing Additive Manufacturing. Proceedings of AMPA2023 , Cham: Springer International Publishing (Springer Tracts in Additive Manufacturing)
115 – 129