论文部分内容阅读
自从有限元方法引入强度计算后,原来诸如一大批用差分法不能解决的问题,都一一有了新的处理办法。但随着有限元的广泛应用,我们不难发现,有限元方法有两大显著的弱点:(1)由于在求解域内进行网格划分,数据准备工作是大量、烦琐的;(2)由于对边界条件逼近和城内解的逼近而引起二种误差累计造成精度的下降。同时有限元只考虑原始变量位移有足够的精度,但当这些变量被微分后得到应力,结果精度就很差。
Since the introduction of finite element method to calculate the strength, the original such as a large number of problems can not be solved by the difference method, one by one with a new approach. However, with the wide application of finite element method, it is not difficult to find that the finite element method has two significant weaknesses: (1) the data preparation is large and cumbersome due to the meshing in the solution domain; (2) The approximation of the boundary conditions and the approximation of the solution in the city cause the accuracy of the two kinds of errors to decrease. At the same time, the FEM only considers the displacement of the original variable with enough precision, but when these variables are differentiated, the stress is obtained, resulting in poor accuracy.