摘要
对几何模型进行适当的简化是建立理想有限元分析模型的一个必不可少的步骤，而区分几何模型中的结构特征相对大小是几何简化的关键。对几何模型的边界进行等半径整体倒圆（将这种操作称为“弧化”）具有几何形体的低通滤波作用，通过比较弧化前后的几何模 型，可以确定几何边界元在构成整个形体的重要程度，即几何细微度。以二维图形为例， 说明利用弧化来量化几何边界元的细微度的原理，并提出了以细微边界元为线索识别和 分解细微特征的方法。在细微边界元识别和细微特征分解交替进行的过程中，任意用边界表示法表示的几何模型可逐步简化到任意程度。提出的方法是从设计模型中提取分析几何模型的一种有效工具。
To distinguish the small features from the large ones is the key technology of geometric simplification, which is one of the major subtasks under abstraction of model for FEA. Filleting the whole boundaries of the geometric model with constant fillet radius, which is called filletization, has the effect of low-pass filtering to geometric shape. The locality of boundary entities can be evaluated by comparing the original model with the filleted model. The principle to quantify the detail-level of boundary geometric entities based on filletization is described in detail and an approach to decompose geometric object into detail features is developed. In the cycle of detailed boundary entities recognition and detail feature decomposition, an arbitrary solid model which is represented as B-Rep model can be simplified to any level. The method proposed in this paper can be used as an efficient tool of abstracting geometric model for analysis from CAD model.
出处
《机械工程学报》
EI
CAS
CSCD
北大核心
2001年第9期44-48,52,共6页
Journal of Mechanical Engineering
