为了最大限度地提高有限元分析的效率,提出一种基于Delaunay细分构造旋转对称模型的最优对称单元的方法。该方法把对称单元及其网格的构造统一起来,通过一种带对称约束的局部Delaunay细分算法直接生成对称单元网格。其关键是在细分过程中增添移动三角形操作,由此可将对称边转化为内部边,从而能够对其进行翻转来维护其Delaunay属性,并可保持对称边界的一致性。细分结束后得到的局部网格就是所要求的最优对称单元网格。理论证明与实验结果均表明该方法是有效的。
To maximize the efficiency of finite element analysis,a Delaunay refinement based method was proposed to construct optimal symmetry cells for rotational symmetric models.In this method,the construction of symmetry cell and its mesh was synchronized to generate the symmetry cell mesh directly by a symmetry-constrained local Delaunay refinement algorithm,which was achieved by introducing moving triangles into the Delaunay refinement process.Thus,symmetry boundary edges were changed into interior edges to maintain their Delaunay properties and to retain the consistency of the symmetry boundary.The locally constructed mesh after the refinement was the desired symmetry cell mesh.Both theoretical proof and experimental results demonstrated the effectiveness of this method.