Length Scale Control for Binary Topology Optimization
Please login to view abstract download link
Topology Optimization of Binary Structures (TOBS) is a gradient-based method that employs binary design variables and integer linear programming to solve topology optimization problems. In topology optimization, ensuring length scale control poses a significant challenge for achieving practical and manufacturable designs. However, TOBS lacks a treatment to ensure the reach of a defined minimum length scale. In our research, we adopt a length scale control method based upon morphological operators, used in certain density-based methods, for Topology Optimization of Binary Structures. Here, we approximate morphological operators with differentiable nonlinear filters. We test the proposed approach through numerical experiments on benchmark minimum compliance problems subject to a volume constraint. The results show that adjusting the parameters of the employed nonlinear filters provides a length scale control for structure and void regions, ensuring the manufacturability of the generated designs.
