Nonlinear static analysis of extreme structural behavior: Overcoming convergence issues via an unconditionally stable explicit dynamic approach
Nonlinear static analysis of extreme structural behavior: Overcoming convergence issues via an unconditionally stable explicit dynamic approach
复制标题
DOI：
10.1016/j.istruc.2023.01.086
复制
发表时间：
2023-03
期刊：
影响因子：
4.1
通讯作者：
Si-Cong Xie;C. Kolay;D. Feng;J. Ricles
中科院分区：
文献类型：
--
作者：
Si-Cong Xie;C. Kolay;D. Feng;J. Ricles
结构极端行为的模拟是研究工程结构损伤破坏机理的重要手段。然而，收敛性问题经常发生在模拟极端载荷下的结构响应时，使用传统的非线性解决方案的战略，如牛顿-拉夫森家庭的方法。虽然商业有限元软件提供了显式动力学方法来克服这个问题，但它们只是有条件稳定的，并且需要小的时间步长。本文提出了一种新的显式动力学方法来模拟涉及极端结构行为的静力问题。采用多点激励模式，以动态方式施加节点位移历程，然后采用无条件稳定的显式KR-α方法消除收敛问题。三个例子来说明所提出的动态方法的有效性，并进行参数研究，以调查算法参数的影响。结果表明，该方法提供了相同的精度，传统的静态方法没有任何收敛问题。
Simulating extreme structural behavior is of vital importance as it is a major means to study the damage and failure mechanism of engineering structures. However, convergence issues frequently occur when simulating structural response under extreme loading using traditional nonlinear solution strategies, eg, the Newton–Raphson family of methods. Though commercial finite element software provides explicit dynamic methods to overcome this issue, they are only conditionally stable and require a small time step size. In this paper, a new explicit dynamic approach to simulate static problems involving extreme structural behavior is presented. The multi-support excitation pattern is employed to apply a nodal displacement history of the problem in a dynamic way, then the unconditionally stable explicit KR-α method is used to eliminate the convergence issues. Three examples are presented to illustrate the effectiveness of the proposed dynamic approach, and a parametric study is conducted to investigate the influence of the algorithmic parameters. The results indicate that the proposed method provides the same accuracy as the traditional static method without any convergence issues.