Robert Kremer, M.Eng.

As part of the research, the process of additive laser manufacturing is to be virtually mapped and simulated from the material to the finished component. This makes it possible to predict the mechanical-technological properties of a component depending on the chemical composition of the material and the manufacturing parameters, thus enabling optimization to be carried out virtually in advance. This saves resources in the area of material development, as optimization steps through test melts and laboratory tests can be significantly reduced. Component geometry and process parameters can also be optimized virtually in advance, which reduces previously necessary iterations in the laboratory to a minimum. For implementation, approaches and methods from materials science, thermodynamics and higher mechanics are integrated into a comprehensive multiscale model in order to simulate metallic materials from atomic composition to the component produced by selective laser melting (metal 3D printing). Laboratory tests are carried out in parallel to validate the simulations. For implementation, the simulation model for a CuSn10 substitution alloy is to be set up according to the Integrated Computational Materials Engineering (ICME) approach and coupled with a process simulation. The approaches of Density Functional Theory (DFT), Molecular Dynamic (MD) and Dislocation Dynamic (DD) as well as Crystal Plasticity (CP) and Finite Element Method (FEM) are combined by bridging across the individual size scales so that a continuous flow of information is created from the smallest - the electron level (a few angstroms) - to the largest - the finished component (many millimetres). This makes it possible to calculate the elastic and plastic material behavior as a function of the alloy composition and component production. The planned simulation model thus enables a virtual and simulative representation of the real material and the additive processing of two-material substitution alloys. Once the simulation has been successfully set up, the simulation model to be created can be extended to multi-material systems in subsequent work, used to generate data and used in combination with laboratory tests for AI-driven models.