A Generative Topology Optimization Framework for Additive Manufacturing Based on Constructive Solid Geometry and Evolutionary Optimization

کد مقاله : 1434-ISME2026

نویسندگان

محدثه لطفی، فربد ناظمی، سعید خدایگان * دانشگاه صنعتی شریف

چکیده مقاله

The integration of topology optimization within the generative design paradigm offers significant potential for leveraging the geometric freedom provided by Additive Manufacturing (AM). This research presents a novel topology optimization framework for generative design in AM, integrating the concept of Constructive Solid Geometry (CSG) with evolutionary optimization techniques. While AM offers unparalleled freedom in fabricating complex geometries, conventional topology optimization methods often struggle with computational efficiency, direct compatibility with computer-aided design systems. To address these limitations, this study proposes a geometric representation scheme where the topology is defined by the Boolean union of parametric primitives derived from a Delaunay triangulation of variable nodes. A Genetic Algorithm (GA) is employed to optimize the positions and dimensions of these primitives, effectively exploring the design space for optimal structural compliance under volume constraints. The proposed method ensures structural connectivity and explicit boundary definition, eliminating the need for complex density filtering or interpretation required in grid-based methods. The efficacy of the algorithm is demonstrated through a comprehensive numerical study of a cantilever beam problem. The results indicate that the CSG-based evolutionary approach successfully generates diverse, manufacturable, and structurally efficient topologies with significantly reduced design variables compared to traditional voxel-based methods.

کلیدواژه ها

Topology optimization, Generative design, Constructive solid geometry, Genetic algorithm, Design for additive manufacturing

وضعیت: پذیرفته شده برای ارائه شفاهی