找回密码
 注册
查看: 6854|回复: 23

[转帖]精彩网格图片

[复制链接]
发表于 2005-11-22 18:17:36 | 显示全部楼层 |阅读模式

马上注册,结交更多好友,享用更多功能,让你轻松玩转社区。

您需要 登录 才可以下载或查看,没有账号?注册

x
[这个贴子最后由houston在 2005/11/22 06:18pm 第 1 次编辑]

hope it helpful for you.
发表于 2005-11-23 11:41:39 | 显示全部楼层

[转帖]精彩网格图片

好漂亮的结构化网格,什么软件做的,介绍一下.
 楼主| 发表于 2005-11-23 14:35:23 | 显示全部楼层

[转帖]精彩网格图片

Gridgen弄的
转贴过来的
觉得这个网格生成得真是好.
发表于 2005-11-23 18:52:40 | 显示全部楼层

[转帖]精彩网格图片

真的好漂亮
发表于 2005-11-23 19:45:59 | 显示全部楼层

[转帖]精彩网格图片

请问,这个应用在什么问题呢?
 楼主| 发表于 2005-11-24 11:06:10 | 显示全部楼层

[转帖]精彩网格图片

应用在机翼问题
看过这样的图后,觉得对自己的研究非常有帮助
可能事先,你不会想到用结构网格也可以对这么复杂的边界几何形状生成正交性这么好的网格.
发表于 2005-11-25 11:04:15 | 显示全部楼层

[转帖]精彩网格图片

谢谢哦,我做的多是算法方面,对于模型设计不多,所以一般都涉及不到这么复杂的几何形状,对于网格的知识特别缺乏不过还是不得不说,这张图看起来真漂亮呀。
发表于 2005-11-25 22:33:40 | 显示全部楼层

[转帖]精彩网格图片

网格质量确实不错,不过这样的三元翼型的网格总数也不少吧?
发表于 2005-11-30 23:07:53 | 显示全部楼层

[转帖]精彩网格图片

看不懂
 楼主| 发表于 2005-12-4 08:26:45 | 显示全部楼层

[转帖]精彩网格图片

网格数不会怎么多,毕竟是二维的问题,由这张图给我们一个启发,那就是不一定稍复杂一点的几何形状就用非结构网格,结构网格很成熟的.
发表于 2005-12-7 16:56:29 | 显示全部楼层

[转帖]精彩网格图片

不错啊
 楼主| 发表于 2005-12-9 21:14:21 | 显示全部楼层

[转帖]精彩网格图片

这些老外很多弄起东西来还真细心
不知国内有没有人整得这么仔细
贴点图来分享一下...
发表于 2006-6-25 22:35:52 | 显示全部楼层

[转帖]精彩网格图片

怎么看得出网格生成的好坏,或者这个网格进行计算合不合理?
发表于 2006-6-26 16:38:38 | 显示全部楼层

[转帖]精彩网格图片

zz网格质量:
Jacobian(必须为正值,即网格体积必须为正),网格的正交性,最小边与最大边的比值,扭角(skew angle, skewness)
网格的正交性: 每一个网格的体积和其的长,宽,高的乘积的比A=V/(a*b*c). 所以正交性越好,A就越接近1,因为网格更接近长方体。
Cell skewness is calculated by determining the intersection angles of each individual face and its four adjacent neighbors. The largest of these is considered to be the cell skewness (0.0 implies orthogonality)

Mesh Quality (Tgrid Help)
Mesh quality is determined by four basic measures: clustering, smoothness,  skewness, and aspect ratio.  The relationship between these criteria and the numerical solution obtained for the computational grid is discussed in detail in Boundary Mesh Quality, and summarized briefly here.
The requirement for clustering is simply that the mesh be fine enough to resolve the primary features of the flow being analyzed. You can control the resolution with the boundary mesh that you start from and also with the parameters that control the generation of the interior mesh.  
In a high-quality mesh, the change in size  from one face or cell to the next should be gradual (smooth); large differences in size between adjacent faces or cells will result in a poor computational grid because the differential equations being solved assume that the cells shrink or grow smoothly.  (Figure 1 illustrates different changes in size for triangular cells.)
Skewness determines how close to ideal (i.e., equilateral or equiangular) a face or cell is (see Figure 2).
Highly skewed  faces and cells are unacceptable because the equations being solved assume that the cells are relatively equilateral/equiangular.
Aspect ratio can also be used to determine how close to ideal a face or cell is. The aspect ratio of a face or cell is simply the ratio of the longest edge length to the shortest edge length. For an equilateral face or cell (e.g., an equilateral triangle or a square), the aspect ratio will be 1. For less regularly-shaped faces or cells, the aspect ratio will be greater than 1, since the edges  differ in length.
For triangular and tetrahedral faces and cells and for pyramids, you can usually focus on improving the  skewness, and the smoothness and aspect ratio will consequently be improved as well. For prisms (including 2D quadrilateral cells),  however, it is important to check the aspect ratio and/or the change in size in addition to the skewness, because  it is possible to have a large jump in cell size between two cells with low skewness (see Figure 3) or a high-aspect-ratio low-skew cell (such as the quadrilateral on the  right in Figure 3).
发表于 2006-10-24 23:28:46 | 显示全部楼层

[转帖]精彩网格图片

您需要登录后才可以回帖 登录 | 注册

本版积分规则

快速回复 返回顶部 返回列表