沃罗诺图
共形映射
可控性
概率密度函数
边界(拓扑)
领域(数学)
几何学
功能(生物学)
算法
相似性(几何)
数学
密度梯度
拓扑(电路)
计算机科学
数学优化
数学分析
物理
应用数学
人工智能
生物
统计
组合数学
图像(数学)
进化生物学
量子力学
纯数学
作者
Bin Liu,Xiaofeng Wei,Wei Cao,Ping Lü,Xiaofeng Wang
标识
DOI:10.1016/j.tws.2024.111623
摘要
This work introduces a rapid modeling method for gradient porous structures with conformal density, aiming to address the challenges of transitional variation of porosity, and control of complex gradient variations in multiple directions. In comparison to other methods, this approach overcomes the limitations associated with coordinate systems and shape functions when designing complex gradient variations in multiple directions while achieving density variation with shape under different gradients. The method involves mapping the volumetric distance field to a density field using control functions. By adhering to the constraints of the density field, it employs a weighted random sampling method to attain gradient sites with shape-adaptive distribution. Voronoi polyhedron are then constructed based on these sites, and smooth Voronoi struts are generated using strut distance fields and improved Boolean operations. The boundary adaptation of the porous structure is subsequently achieved based on the volumetric distance field. By establishing the relationship between density and volumetric distance field values, the study illustrates the similarity between density variation in the structure and gradient variations in the control function, indicating the controllability of density variation with shape. Furthermore, the method enables the integration design of structural shape and mechanical properties through adjustments to the number of sites, radius size of struts, and gradient control functions. Finally, the method was validated through numerical simulation and experiments, demonstrating its controllability and effectiveness in generating random porous structures with conformal density gradients, providing important theoretical and practical support for research and application in related fields.
科研通智能强力驱动
Strongly Powered by AbleSci AI