Efficient Layer-by-Layer Simulation for Topology Optimization

Abstract

Topology optimization and additive manufacturing together enable the optimal design and direct fabrication of complex geometric parts with groundbreaking performance for diverse applications. However constraining the optimization to ensure that the generated object can be reliably manufactured via layer-by-layer 3D printing processes is challenging. The typical solution is to enforce design rules based only on geometric heuristics like overhang angles, minimum wall widths, and maximum bridge spans. Recent work has proposed instead to simulate the robustness of each partial object generated from bottom-to-top during the fabrication process as a more accurate, physics-aware printability assessment. However, this approach comes at the cost of an vast increase in the number of simulations run per design iteration, making existing implementations intractable at high resolution. We demonstrate that by developing a custom solver leveraging the close relationships between these many simulations, even voxel-level layer-by-layer simulations are feasible to incorporate into high-resolution 2D and 3D topology optimization problems on a single workstation.