ECCOMAS 2024

Integrating Multiple Loading Cases In High-Resolution Topology Optimization: A Multi-Scale Approach

  • Jensen, Peter Dørffler Ladegaard (Technical University of Denmark)
  • Sigmund, Ole (Technical University of Denmark)

Please login to view abstract download link

The seminal works on high-resolution topology optimization procedures have provided new insights into structural design [1]. Despite these advancements, their high computational costs limit practical industry applications. Concurrently, industrial-grade additive manufacturing (AM) has seen substantial improvements, highlighting the need for AM-specific high-resolution inverse design methods that integrate infill as a structural component. These challenges were addressed by Jensen et al. [2] by extending the de-homogenization topology optimization method [3] to 3D unstructured grids, where a multi-scale optimization approach was employed by considering an orthotropic periodic rectangular-hole material, which is reminiscent of the stiffness optimal Rank-3 material. The microstructure orientations and laminates are regulated to enhance stability and manufacturability. The coarse multi-scale structure is reconstructed on a high-resolution single-scale by considering a finite periodicity of the microstructure, which corresponds to a minimum length scale, by computing stream surfaces that align with the microstructure, resulting in a high-resolution structure with a minor reduction in structural performance compared to the coarse scale optimized solution. However, this method was limited to single-load case problems, which are often unrealistic. This work extends [2] by incorporating the optimal Rank-N microstructure as the material model, enabling the optimization of structures under multiple load-case scenarios. This advancement allows for the reproduction of complex structures like the giga-scale wing [1], with significantly reduced computational resources and comparable structural performance.