CSDF:一种基于STL的无锯齿差分网格剖分算法

针对基于差分的铸造CAE系统生成体网格时图像显示呈锯齿状这一问题,提出了一种基于STL面片模型的无锯齿网格剖分算法。算法采用一种新的组件式有符号距离场(CSDF)模型来对多个STL实体进行剖分。该模型采用了一种稳健的容错方法,能够对不封闭、缺点、少面的STL实体进行正确的剖分;使用BSP树来加速计算每个网格中心点距离铸件、砂芯、铸型以及浇注系统等STL几何实体的最小距离,并规定位于内部的中心点的距离为负,外部为正,表面为零,使得每一个网格点有一个表示材质的数据,同时还有几个不同的距离值;材质数据用来表示物理属性,而距离值用来进行STL实体的布尔运算,以实现组件式的网格剖分和几何模型的重建,获得光滑的后处理图像。实例表明:算法可行,可实现组件式的几何实体无锯齿网格剖分,且剖分结果能进行数值计算。

CSDF:a New Alias-free Finite Difference Grid Generation Algorithm Based on STL

To solve the alias problem during post proce ss ing in finite difference method (FDM) based casting CAE systems, this paper prop oses a novel alias-free finite difference grid generation algorithm based on STL files . Unlike previous methods, this method utilizes a Component based Signed Distanced Field (CSDF) model to dividing different STL meshes. The CSDF firstly adopts a robust fault-tolerant method which can accurately adapt to divide broken STL meshes, then utilizes a BSP(Binary Space Partition) tree to accelerate the calc ulation of the minimum distances between the center point of each grid and diff erent STL meshes, such as cast, sand, mold and gating systems meshes. In CSDF mo del, it is defined that the distances is positive if the center point of grid is in a STL mesh, negative outside STL mesh and zero on the mesh; thus a single g rid not only corresponds to a material data but also corresponds several minimum distances. The material data is then used to represent different physical prope rties in later numerical computing. While the distance data are used to realize robust boolean operations between different STL meshes and reconstruct the origi nal STL data to get smooth images in post visualization. Case study shows that CSDF is valid for alias free finite different grid generation as well as correct numerical simulation.