有限元法
边界(拓扑)
曲线坐标
边值问题
离散化
边界元法
边界节点法
奇异边界法
数学
连续介质力学
线性化
数学分析
几何学
物理
经典力学
非线性系统
量子力学
热力学
作者
Ali Javili,Paul Steinmann
标识
DOI:10.1016/j.cma.2009.11.003
摘要
This paper, in line with part I [53], is concerned with the finite element implementation of boundary potential energies and the study of their impact on the deformations of solids thereby the main thrust is the fully three-dimensional formulation and implementation incorporating anisotropic effects. Boundary effects can play a dominant role in the material behavior, the most prominent example being surface tension. However, the common modelling in continuum mechanics takes exclusively the bulk into account, nevertheless, neglecting possible contributions from the boundary. Within this contribution the boundary potentials are allowed, in general, to depend not only on the boundary deformation but also on the boundary deformation gradient and the spatial boundary normal. For the formulation of the finite element method, the concept of convected curvilinear coordinates attached to the boundary is employed and the corresponding derivations completely based on a tensorial representation are carried out. Afterwards, the discretization of the generalized weak formulation, including boundary potentials, is performed and eventually numerical examples are presented to demonstrate the boundary effects due to different proposed material models. In contrast to the previous literature on this topic, the current manuscript covers jointly the following issues related to boundary energies: (1) the formulation and implementation represents a fully three-dimensional framework at large deformations, (2) the formulation of the problem and the proposed material models are based on finite strains, however, it is shown that the linearization would lead to the small strain models proposed previously in the literature and (3) the current manuscript covers the issue of anisotropy effects on the boundary energies which to the best of our knowledge has not been exemplified earlier.
科研通智能强力驱动
Strongly Powered by AbleSci AI