SIM-AM 2025

Optimizing Porosity as a Separate Design Variable Using the Thermodynamic Topology Optimization Framework

  • Pravda, Ján (Leibniz University Hannover)
  • Jantos, Dustin (Leibniz University Hannover)
  • Junker, Philipp (Leibniz University Hannover)

Please login to view abstract download link

Motivated by the advancements in the field of additive manufacturing, topology optimization has the potential to utilize the different characteristics and boundary conditions of these new manufacturing methods, compared to traditional technologies. With additive manufacturing it is possible to design and control the microstructure and material composition to a great degree of detail and the most novel techniques enable even the production of targeted gradation of material properties. Thus, appropriateoptimization tools have to be developed. Our work is based on the Thermodynamic Topology Optimization framework and expands on previous works. Introducing a new design variable, we’re able to describe and optimize the porosity distribution in the component, independently of the density distribution defining the topology, both purely on the macro-level. The effective stiffness of a finite element is defined by both design variables, which interpolate between elasticity tensors for full and porous material. Specific adjustments, such as delaying the evolution of the porosity distribution have been introduced, to improve the quality of the results. A systematic parameter study, searching for the optimal numerical parameters for fast convergence and stability, has been performed. In our work, the focus has been laid on boundary problems with multiple load cases, offering an interesting compromise between optimality and robustness and yielding porousmicrostructures for multiple load cases more optimal.