PetIBM: toolbox and applications of the immersed-boundary method on distributed-memory architectures
PetIBM: toolbox and applications of the immersed-boundary method on distributed-memory architectures
复制标题
DOI：
10.21105/joss.00558
复制
发表时间：
2018-05
期刊：
影响因子：
--
通讯作者：
Pi-Yueh Chuang;O. Mesnard;A. Krishnan;L. Barba
中科院分区：
文献类型：
--
作者：
Pi-Yueh Chuang;O. Mesnard;A. Krishnan;L. Barba
在IBM框架中，一组拉格朗日标记定义了浸没边界(其中强制了边界条件)，并在扩展域(包括身体域)上求解流体方程。在计算多个移动的沉浸体周围的流动时，欧拉网格保持不变，这样就不需要在每个时间步长重新划分网格。PetIBM使用二阶有限差分格式、各种可选的时间积分器和完全离散的投影方法来离散流体方程(Perot(1993))。它实现了两种浸入边界算法：浸入边界投影法(Taira和Colorius(2007))及其解耦版本(Li等人)。(2016))。
In the IBM framework, a collection of Lagrangian markers defines the immersed boundary (where boundary conditions are enforced) and the fluid equations are solved over the extended domain (including the body domain). The Eulerian mesh remains unmodified when computing the flow around multiple moving immersed bodies, which removes the need for remeshing at every time step. PetIBM discretizes the fluid equations using a second-order finite-difference scheme, various optional time-integrators, and a fully discrete projection method (Perot (1993)). It implements two immersed-boundary algorithms: the immersed-boundary projection method (Taira and Colonius (2007)) and its decoupled version (Li et al. (2016)).