基于面片句柄图形对象的无网格法可视化研究

Meshless numerical results of cloud chart characterization based on patch handle graphics object

无网格法的离散粒子模型特性,对于实现从纳观、微观、介观和宏观的多尺度分析计算的无缝连接具有特别重大的意义。同时由于不需要网格划分,使得其在涉及网格畸变、网格重新划分的问题上较有限元法具有明显优势。但是无网格法的离散粒子模型特性也恰恰导致其数值计算结果的全域可视化表征遇到困难。为此提出采用Delaunay三角化方法将无网格法求解域中的离散点连接成背景三角形网格,将无网格法数值结果的云图生成问题转化为三角形面片句柄图形对象的创建问题来处理。基于Matlab的算例表明,利用低级绘图指令设置面片句柄图形对象的面与边颜色的线性插值模式,能够较采用高级绘图指令更快速地生成色彩过渡平滑的高质量无网格法数值结果云图。

Unlike mesh-based finite element method （FEM）, the shape function of meshless method （MM） is defined by discrete points in computational domain. Because of mesh independence, compared with FEM, the pre-processing procedure of MM becomes simpler and MM is much more powerful for solving large deformation, high gradient and instantaneous dynamic problems, which always involve severe mesh distortion and indispensable remeshing, Due to the characteristic ofdiscrete particle model, MM is significant for the multi-scale analysis, which requires seamless transition among nano-, micro-, meso-, and macro-scale. While on the other hand, the discrete particle characteristic just results in difficulties for visualizing the meshless numerical results during the post-processing procedure. On the basis of Delaunay triangulation, which constructs background triangular mesh from discrete points in whole domain, the cloud chart characterization for MM is then converted into the creation of triangular patch handle graphics objects. Numerical sample performed in the Matlab environment shows that the low-level graphics function patch is able to set the color shading property for triangle＇s face and edges to linear interpolation mode according to the vertex colors and consequently.to generate high quality cloud chart rapidly. In contrast, the traditional high-level graphics function fill does not possess such high capability.