Topology Optimization considering Shielding and Penetrating Features based on Fictitious Physical Model
Daiki Soma, Kota Sakai, Takayuki Yamada
TL;DR
This work tackles topology optimization with explicit geometric constraints by introducing shielding and penetrating features. It combines a level-set formulation with a fictitious steady-state temperature framework to evaluate these features, yielding objective terms that reward appropriate void-solid connectivity patterns. The method formulates a minimum mean compliance problem augmented by shielding and penetrating evaluators, solved with an augmented Lagrangian approach on 2D and 3D domains. Numerical demonstrations show the ability to form shielding walls and penetrating holes, with parameter studies guiding practical design choices for manufacturability and performance.
Abstract
This paper proposes topology optimization for considering shielding and penetrating features. Based on the fictitious physical model, which is a useful approach to control geometric features, the proposed method analyzes fictitious steady-state temperature fields and interprets target geometric features by examining the temperature change. First, the concept of topology optimization based on the level set method is introduced. Next, the basic idea of the fictitious physical model for considering geometric features is explained. Then, the differences between the shielding and penetrating features are clarified, and the fictitious physical model for evaluating these features is proposed. Furthermore, topology optimization for the minimum mean compliance problem with geometric conditions is formulated. Finally, 2D and 3D numerical examples are presented to validate the proposed method.
