
Local Perimeter and Non-Linear Filtering for Controlling Minimum Length and Overhang in Topology Optimization
Please login to view abstract download link
One of the key challenges currently facing the transport industry is the development of new lightweight structures aimed at reducing fuel consumption and minimizing environmental impact. Topology optimization techniques offer a solution for weight reduction while also minimizing design time and cost. However, these methods often produce complex designs that can only be manufactured through additive manufacturing. Therefore, it is essential to integrate numerical optimization with 3D printing constraints to ensure manufacturability, addressing factors such as minimum length scale and overhang control to prevent the creation of complex shapes and volumes. This study aims to explore the feasibility of optimal design by incorporating minimum length and overhang constraints within the additive manufacturing process. We introduce the concept of non-linear filters to penalize overhangs in the 3D printing sense while maintaining a length scale in the opposite direction. Additionally, this filtering is combined with the concept of local perimeter to better satisfy 3D printing constraints in a more localized manner. The results indicate that small-length scale local bars are eliminated, and a vertical orientation of bars is achieved.