Multi-Joint Topology Optimization Method with Interface Geometry Control

Topology optimization (TO) has become a reliable and lightweight design approach which was widely adopted in multiple industrial applications. Over the past decades, TO has advanced through three major development stages to increase its practicality and application scope: single-material topology optimization (SMTO), multi-material topology optimization (MMTO), and multi-joint topology optimization (MJTO). SMTO involves only one candidate material, whereas MMTO takes multiple candidate materials into consideration, which widen the application scope of TO. In terms of MJTO, it not only considers multiple candidate materials but also considers the bonding material between dissimilar candidate materials, improving practicality over MMTO. However, prior MJTO methods overlook the geometry of dissimilar material interfaces, a factor that may impact material bonding effect. In this paper, a novel MJTO algorithm is introduced to enable explicit control over interface geometry in MJTO. Central to the method is a Newton’s law of gravitation-inspired dissimilar material interface orientation detection approach, which accurately calculates the bonding material’s orientation. Using the orientation detection approach, the bonding material’s Young’s modulus can be penalized when undesired orientations occur, thus enforcing geometric control of the interface. In this paper, the proposed algorithm is evaluated through multiple academic models to demonstrate its effectiveness. Numerical study results demonstrate that both vertical and horizontal dissimilar material interfaces can be effectively controlled using the proposed method, leading to improved producibility of TO-derived designs.