Project A05

Modeling of laser directed energy deposition considering the resulting properties of hybrid porous structures

The production of hybrid, graded components with locally adjustable properties by laser deposition with powder feed (laser directed energy deposition - DED-LB) requires precise knowledge about the processes and interactions in order to produce components with specifically adjusted, desired properties. Therefore, the goal of this project is the development of a simulation environment of the DED-LB process that combines process models on the micro and feature level with a virtual manufacturing method (VM) on the component level to holistically describe the properties of a HyPo-component during and after the manufacturing process. .

At micro level, a thermo-fluid structure simulation will be established, which, as a result, specifies the effective heat input and the geometry of the applied material layer. Due to computation limitations, the process simulation on micro level is limited to a small component area. Hence, on feature level a process model with some simplifying assumptions will be used to calculate the thermo-induced residual stresses with respect to the chosen path of the laser and its effective heat input, specified by the process model on micro level. Results from the models on both scales are transferred to the virtual simulation environment on the component level, which represents the spatially and temporally variable properties of the entire HyPo-component. The virtual simulation environment can infer the properties at any position in the component from scalar or vector-valued variables (e.g. material present), tensor-valued (e.g. residual stresses), multidimensional variables (local and temporal temperature profiles) and variables subject to probabilistic variance (porosity, material parameters with uncertainties) by means of interpolation models, thus enabling an efficient description even of large HyPo-components. In addition, properties from further simulations or analyses of samples can be integrated, so that the material and component properties are available during the manufacturing process and after fabrication. The component information can be retroactively fed back to the process simulations on micro and feature level as initial or boundary conditions. By combining the physical modelling of the local process operations and the material and component properties of the entire HyPo-component, process parameters can be derived and passed on to manufacturing, which enables generation of the desired properties of the HyPo-component.

Project leaders
Dr.-Ing. Dipl.-Math. Volker Böß
Prof. Dr.-Ing. Kristin de Payrebrune

Doctoral researchers
Mariem Ben Salem, M.Sc.
Alfred Jose Puthoor, M.Sc

Publications