基于牛顿迭代的Navier-Stokes方程两重网格相容分解算法
two-level consistent splitting method for the Navier-Stokesequations based on Newton iteration
本文研究了一种求解不可压缩Navier-Stokes方程的全离散两重网格相容分解算法。该算法首先在粗网格空间中求解一个非线性方程,然后在细网格空间中求解一个线性方程。在细网格空间的求解过程中应用相容分解格式将速度和压力解耦求解。我们应用牛顿迭代来反映大小涡分量之间的相互作用关系,在满足inf-sup条件的基础上得到该算法的稳定性和收敛性。数值分析表明只要搭配合适的粗细网格尺度,该两重网格算法可以得到最优的逼近精度。一些数值算例表明该两重网格算法的有效性。
fully discrete two-level consistent splitting algorithm is appliedto solve the incompressible Navier-Stokes equations. This method ismotivated by solving a nonlinear equation in the coarse-levelsubspace and then computing a linear problem in the fine-levelsubspace in which the consistent splitting scheme is employed todecouple the velocity and the pressure. To reflect the interactionrelation between the large and small eddy components well, we useNewton iteration to denote such interaction relation. The stability and convergence are obtained under the inf-sup condition. Thedetailed numerical analysis shows that the two-level method can getthe optimal accuracy with the proper choice of the coarse and finemesh scales. Finally, some numerical tests are provided to indicatethe effectiveness of this method.
刘庆芳
数学力学
两重网格相容分解算法Navier-Stokes方程稳定性和收敛性
two-level consistent splitting schemeNavier-Stokesequationsstability and convergence
刘庆芳.基于牛顿迭代的Navier-Stokes方程两重网格相容分解算法[EB/OL].(2017-04-26)[2025-08-03].http://www.paper.edu.cn/releasepaper/content/201704-596.点此复制
评论