不可压缩流动的并行数值方法

不可压缩流动的数值模拟是计算流体力学的重要组成部分. 基于有限元离散方法, 本文设计了不可压缩Navier-Stokes (N-S)方程支配流的若干并行数值算法. 这些并行算法可归为两大类: 一类是基于两重网格离散方法, 首先在粗网格上求解非线性的N-S方程, 然后在细网格的子区域上并行求解线性化的残差方程, 以校正粗网格的解; 另一类是基于新型完全重叠型区域分解技巧, 每台处理器用一局部加密的全局多尺度网格计算所负责子区域的局部有限元解. 这些并行算法实现简单, 通信需求少, 具有良好的并行性能, 能获得与标准有限元方法相同收敛阶的有限元解. 理论分析和数值试验验证了并行算法的高效性     

交错网格上的压力Poisson方程法和抛射法的本质相容性

通过保留压力Poisson方程源项中的粘性项,建立了交错网格上的一种求解压力Poisson方程的新方法．该方法的推导过程显示,新建立的压力方程与抛射法中的压力方程具有相同的形式．数值计算显示这两种求解压力的方法不会影响速度和压力的结果,而且计算耗时相当．     

Navier-Stokes方程的一种并行两水平有限元方法

基于区域分解技巧,提出了一种求解定常Navier-Stokes方程的并行两水平有限元方法．该方法首先在一粗网格上求解Navier-Stokes方程,然后在细网格的子区域上并行求解粗网格解的残差方程,以校正粗网格解．该方法实现简单,通信需求少．使用有限元局部误差估计,推导了并行方法所得近似解的误差界,同时通过数值算例,验证了其高效性．     

三维多面体网格上扩散方程的保正格式

 针对三维任意(星形)多面体网格, 本文构造了扩散方程的一种单元中心型非线性有限体积格式, 证明了该格式具有保正性. 在该格式设计中, 除引入网格中心量外, 还引入网格节点量和网格面中心量作为中间未知量, 它们将用网格中心未知量线性组合表示, 使得格式仅有网格中心未知量作为基本未知量. 在节点量计算中, 利用网格面上的调和平均点, 设计了一种适用于三维多面体网格的局部显式加权方法. 该格式适用于求解非平面的网格表面和间断扩散系数的问题. 数值例子验证了它对光滑解具有二阶精度和保正性.     

带移动奇异源热方程的移动网格方法

研究了一个带若干奇异源热方程的数值求解,其源的移动由一个常微分方程描述.基于移动观察区域和区域分解思想提出了一个移动网格预估校正算法.网格方程可自然的通过并行高效求解,算法避免了跳跃信息[u]的计算而使物理方程的离散格式变得非常简单,且仍保持了空间上的二阶收敛性.数值例子验证了算法的收敛性和高效性,并模拟了非线性源函数带来的爆破现象.     

Bakhvalov-Shishkin网格上求解边界层问题的差分进化算法

该文在Bakhvalov-Shishkin网格上求解具有左边界层或右边界层的对流扩散方程,并采用差分进化算法对Bakhvalov-Shishkin网格中的参数进行优化,获得了该网格上具有最优精度的数值解.对三个算例进行了数值模拟,数值结果表明:采用差分进化算法求解具有较高的计算精度和收敛性,特别是边界层的数值解精度明显优于选择固定网格参数时的结果.     

非定常扩散方程的单元中心型有限体积格式

本文基于调和平均点建立了一种新的单元中心型有限体积格式,用以求解非定常扩散方程.在网格边上离散法向流时,选择该网格边两端点和该边上的一个调和平均点作为辅助插值点,并将这些辅助插值点上的未知量用网格单元中心点的未知量进行替换,最终得到一个只含网格单元中心未知量的有限体积格式.该格式满足线性精确性质和局部守恒性,且适用于任意多边形网格.在六种不同的多边形网格上进行四个数值实验,分别考虑扩散系数是连续的和间断的以及非线性的情况,数值结果表明:本文所构造的格式在六种网格上的L2误差均可达到二阶收敛精度,对于不同类型的扩散系数,该格式保持良好的鲁棒性,并且从编程实现的角度来说,该格式更易于向三维情况推广.     

Poisson方程有限差分逼近的数学Stencil 及其应用

提出了偏微分方程有限差分逼近的数学Stencil 概念和Stencil消元策略, 建立了求解Poisson方程的新型迭代算法. 新算法与经典的Jacobi方法同样具有并行性质, 而且比Jacobi方法收敛快. 数值试验表明, 新算法达到同等误差精度所需时间比Jacobi方法和Gauss-Seidel方法都少; 而且新迭代法代替Jacobi方法应用于多重网格的磨光操作, 计算速度明显提高;另外多项式加速仍然适用于新迭代法.     

二次Lagrangian有限元方程的几何多重网格法

为了构造快速求解二次Lagrangian有限元方程的几何多重网格法,在选择二次Lagrangian有限元空间和一系列线性Lagrangian有限元空间分别作为最细网格层和其余粗网格层以及构造一种新限制算子的基础上,提出了一种新的几何多重网格法,并对它的计算量进行了估计.数值实验结果,与通常的几何多重网格法和AMG01法相比,表明了新算法计算量少且稳健性强.     

求解三维高次拉格朗日有限元方程的代数多重网格法

本文针对带有间断系数的三维椭圆问题,讨论任意四面体剖分下的二次拉格朗日有限元方程的代数多重网格法．通过分析线性和高次有限元空间之间的关系,我们给出了一种新的网格粗化算法和构造提升算子的代数途径．进一步,我们还对新的代数多重网格法给出了收敛性分析．数值实验表明这种代数多重网格法对求解二次拉格朗日有限元方程是健壮和有效的。     

Algebraic multigrid preconditioning of the Hessian in optimization constrained by a partial differential equation

We construct an algebraic multigrid (AMG) based preconditioner for the reduced Hessian of a linear‐quadratic optimization problem constrained by an elliptic partial differential equation. While the preconditioner generalizes a geometric multigrid preconditioner introduced in earlier works, its construction relies entirely on a standard AMG infrastructure built for solving the forward elliptic equation, thus allowing for it to be implemented using a variety of AMG methods and standard packages. Our analysis establishes a clear connection between the quality of the preconditioner and the AMG method used. The proposed strategy has a broad and robust applicability to problems with unstructured grids, complex geometry, and varying coefficients. The method is implemented using the Hypre package and several numerical examples are presented.     

An efficient hierarchical preconditioner for quadratic discretizations of finite element problems

Higher order finite element discretizations, although providing higher accuracy, are considered to be computationally expensive and of limited use for large‐scale problems. In this paper, we have developed an efficient iterative solver for solving large‐scale quadratic finite element problems. The proposed approach shares some common features with geometric multigrid methods but does not need structured grids to create the coarse problem. This leads to a robust method applicable to finite element problems discretized by unstructured meshes such as those from adaptive remeshing strategies. The method is based on specific properties of hierarchical quadratic bases. It can be combined with an algebraic multigrid (AMG) preconditioner or with other algebraic multilevel block factorizations. The algorithm can be accelerated by flexible Krylov subspace methods. We present some numerical results on the convection–diffusion and linear elasticity problems to illustrate the efficiency and the robustness of the presented algorithm. In these experiments, the performance of the proposed method is compared with that of an AMG preconditioner and other iterative solvers. Our approach requires less computing time and less memory storage. Copyright © 2010 John Wiley & Sons, Ltd.     

并行间断有限元算法求解Navier-Stokes方程

间断Galerkin有限元方法非常适合在非结构网格上高精度求解Navier-Stokes方程,然而其十分耗费计算资源.为了提高计算效率,提出了高效的MIMD并行算法.采用隐式时间离散GMRES+LU SGS格式,结合多重网格方法,当地时间步长加速算法收敛.为了保证各处理器间负载平衡,采用区域分解二级图方法划分网格,实现内存合理分配,数据只在相邻处理器间传递.数值模拟了RAE2822翼型和M6黏性绕流,加速比基本呈线性变化且接近理想值.结果表明了该算法能有效减少计算时间、合理分配内存,具有较高的加速比和并行效率,适合于MIMD粗粒度科学计算.     

Automatic parallel generation of tetrahedral grids by using a domain decomposition approach

An algorithm for the automatic parallel generation of three-dimensional unstructured grids based on geometric domain decomposition is proposed. A software package based on this algorithm is described. Examples of generating meshes for some application problems on a multiprocessor computer are presented. It is shown that the parallel algorithm can significantly (by a factor of several tens) reduce the mesh generation time. Moreover, it can easily generate meshes with as many as 5 × 10⁷ elements, which can hardly be generated sequentially. Issues concerning the speedup and the improvement of the efficiency of the computations and of the quality of the resulting meshes are discussed.     

Adaptive reduction‐based AMG

With the ubiquity of large‐scale computing resources has come significant attention to practical details of fast algorithms for the numerical solution of partial differential equations. Included in this group are the class of multigrid and algebraic multigrid algorithms that are effective solvers for many of the large matrix problems arising from the discretization of elliptic operators. Algebraic multigrid (AMG) is especially effective for many problems with discontinuous coefficients, discretized on unstructured grids, or over complex geometries. While much effort has been invested in improving the practical performance of AMG, little theoretical understanding of this performance has emerged. This paper presents a two‐level convergence theory for a reduction‐based variant of AMG, called AMGr, which is particularly appropriate for linear systems that have M‐matrix‐like properties. For situations where less is known about the problem matrix, an adaptive version of AMGr that automatically determines the form of the reduction needed by the AMGr process is proposed. The adaptive cycle is shown, in both theory and practice, to yield an effective AMGr algorithm. Copyright © 2006 John Wiley & Sons, Ltd.     

Measures for comparing discrete models of single-valued surfaces

The problem of comparing surfaces unambiguously projected on a plane and represented by clouds of points with three-dimensional coordinates is considered. This problem can be reduced to the problem of comparing functions of two variables determined on different finite sets of points, i.e., on nodes of different grids. A new measure for comparing such surfaces and a new numerically efficient algorithm for calculating them are proposed for the general case in which both grids are unstructured and can have different densities. The linear (with respect to the number of points in two grids) estimate of the complexity of the algorithm for calculating the introduced measure, based on two Delaunay triangulations of the initial sets of points, is proved.     

Analysis of algebraic multigrid parameters for two-dimensional steady-state heat diffusion equations

In this work, it is provided a comparison for the algebraic multigrid (AMG) and the geometric multigrid (GMG) parameters, for Laplace and Poisson two-dimensional equations in square and triangular grids. The analyzed parameters are the number of: inner iterations in the solver, grids and unknowns. For the AMG, the effects of the grid reduction factor and the strong dependence factor in the coarse grid on the necessary CPU time are studied. For square grids the finite difference method is used, and for the triangular grids, the finite volume one. The results are obtained with the use of an adapted AMG1R6 code of Ruge and Stüben. For the AMG the following components are used: standard coarsening, standard interpolation, correction scheme (CS), lexicographic Gauss–Seidel and V-cycle. Comparative studies among the CPU time of the GMG, AMG and singlegrid are made. It was verified that: (1) the optimum inner iterations is independent of the multigrid, however it is dependent on the grid; (2) the optimum number of grids is the maximum number; (3) AMG was shown to be sensitive to both the variation of the grid reduction factor and the strong dependence factor in the coarse grid; (4) in square grids, the GMG CPU time is 20% of the AMG one.     

A Multigrid Block LU-SGS Algorithm for Euler Equations on Unstructured Grids

We propose an efficient and robust algorithm to solve the steady Euler equa- tions on unstructured grids.The new algorithm is a Newton-iteration method in which each iteration step is a linear multigrid method using block lower-upper symmetric Gauss-Seidel(LU-SGS)iteration as its smoother To regularize the Jacobian matrix of Newton-iteration,we adopted a local residual dependent regularization as the replace- ment of the standard time-stepping relaxation technique based on the local CFL number The proposed method can be extended to high order approximations and three spatial dimensions in a nature way.The solver was tested on a sequence of benchmark prob- lems on both quasi-uniform and local adaptive meshes.The numerical results illustrated the efficiency and robustness of our algorithm.     

Algorithms for solving inverse geophysical problems on parallel computing systems

For solving inverse gravimetry problems, efficient stable parallel algorithms based on iterative gradient methods are proposed. For solving systems of linear algebraic equations with block-tridiagonal matrices arising in geoelectrics problems, a parallel matrix sweep algorithm, a square root method, and a conjugate gradient method with preconditioner are proposed. The algorithms are implemented numerically on a parallel computing system of the Institute of Mathematics and Mechanics (PCS-IMM), NVIDIA graphics processors, and an Intel multi-core CPU with some new computing technologies. The parallel algorithms are incorporated into a system of remote computations entitled “Specialized Web-Portal for Solving Geophysical Problems on Multiprocessor Computers.” Some problems with “quasi-model” and real data are solved.     

Parallel iterative methods using factorized preconditioning matrices for solving elliptic equations on triangular grids

Parallel analogs of the variants of the incomplete Cholesky-conjugate gradient method and the modified incomplete Cholesky-conjugate gradient method for solving elliptic equations on uniform triangular and unstructured triangular grids on parallel computer systems with the MIMD architecture are considered. The construction of parallel methods is based on the use of various variants of ordering the grid points depending on the decomposition of the computation domain. Results of the theoretic and experimental studies of the convergence rate of these methods are presented. The solution of model problems on a moderate number processors is used to examine the efficiency of the proposed parallel methods.