显式有限元中的一种并行接触算法

开发一种显式非线性有限元分析中的并行接触算法.基于区域分割技术将桶排序全局搜索方法并行化,各处理器通过桶编号向量检测相互交叠及潜在的点一面接触对.根据数据通信的特点将接触对分为三类,对各类接触对分别设计通信策略.数值算例表明,并行算法具有较高的加速比、并行效率及良好的可扩展性.     

基于几何区域分解的三维输运问题并行迭代算法

对三维直角坐标下的输运隐式差分方程,研究了基于几何区域分解的并行迭代算法,给出了串、并行迭代误差估计.并对相关数值结果进行了分析、比较.     

一种求解流体力学方程组的自适应显式时间积分算法及其应用

针对交替方向显式离散格式,提出一个基于结构网格局部加密技术(SAMR)的求解流体力学方程组的自适应时间积分算法;基于该算法,在JASMIN框架上研制多介质流体力学并行自适应数值模拟程序;在512个处理器上模拟惯性约束聚变中的二维内爆模型.数值模拟结果和并行性能分析显示了算法的正确性和并行实现的高效率.     

并行BAORJ算法及其收敛性

本文提出了一个适用于并行的BAORJ算法,并讨论了当矩阵A是对称正定和H-阵时的收敛性,最后给出了数值实验。     

惯性约束聚变内爆中基于多块结构网格的高效辐射扩散并行算法

针对三维球形靶丸内爆高效模拟需求和传统笛卡尔正交网格上辐射加源困难的问题, 发展一种多块结构非正交网格生成方法, 并基于此种计算网格提出高效的三维扩散格式并行算法, 将其应用于辐射流体方程组的求解和三维内爆不对称性的数值模拟, 数值结果显示了算法的有效性。并行性能测试显示该算法可扩展到5400个核上, 并行效率达到69%。     

二维柱几何中子输运方程的并行区域分解方法

分析不同的区域分解方法及优先级插入算法对二维柱几何下中子输运方程S_n间断有限元方程并行效率的影响,给出基于最小面体比的正方形区域分解方法及沿径向的优先级插入算法,并通过将正方形区域分解方法与径向优先级插入算法进行组合,形成新的算法.新算法更适应于二维柱几何下输运方程S_n间断有限元方法的并行计算.数值试验表明,在通信延迟较高的大型国产并行机上,新算法用数百个CPU还可以取得较好的并行效果,比已有方法具有更良好的可扩展性.     

支持向量学习并行区域增长结合活动轮廓模型的图像分割算法

为克服经典区域增长算法门限设置困难和图像分割精度不高的问题,提出了基于支持向量机学习的区域增长与活动轮廓模型结合的高精度图像分割算法。首先交互式选择属于目标区域的子块和背景区域的子块形成支持向量机的训练样本;并利用这些已知的训练样本训练支持向量分类器。在目标与背景的并行竞争增长过程中,利用训练好的支持向量分类器(SVC)进行分类判决,得到目标对象的初始轮廓。为提高分割对象的精度,采用活动轮廓模型获得准确的边缘。仿真实验获得了较好的分割效果,表明该提出的算法是合理可行的。     

隐式TVD有限差分算法区域分裂并行实现

借助静态非重叠离散区域分裂法，研究了Ｈａｒｔｅｎ－Ｙｅｅ隐式ＴＶＤ差分算法在ＭＩＭＤ多机系统上的并行实现问题，通过对超－超喷流流场的定常数值模拟，检验了相关并行处理的收敛性和稳定性，并得到了较好的并行参数。     

EBE技术在结构分析中的应用(Ⅲ)-EBE-PCG法

在建立单元向量、伪单元向量等概念的基础上,提出了当不形成总刚度矩阵时,预处理共轭梯度法(PCG)的一种高度并行的EBE计算方法,其基本思想是把PCG法各步的计算都单元化。文[3]中的数值试验结果表明了它的有效性。     

不可压缩流的并行两水平稳定有限元算法

提出一种数值求解定常不可压缩Stokes方程的并行两水平Grad-div稳定有限元算法。首先在粗网格中求解Grad-div稳定化的全局解, 再在相互重叠的细网格子区域上并行纠正。通过对稳定化参数、粗细网格尺寸恰当的选取, 该方法可得到最优收敛率, 数值结果验证了算法的高效性。     

Phase retrieval using the transport of intensity equation solved by the FMG–CG method

The relationship between the object-plane phase and the intensity distribution in the Fresnel region can be described by the transport of intensity equation (TIE). The phase distribution can thus be uniquely determined by solving TIE. In this study, a full multigrid preconditioned conjugate gradient (FMG–CG) method is proposed to numerically solve the TIE for phase retrieval. The full multigrid method is a scalable algorithm, and can be parallelized readily and efficiently. By using this method as a preconditioner of the preconditioned conjugate gradient (PCG) method, fast convergence is obtained. The simulation experiments show that complicated phase distributions with fast convergence speed can be retrieved by this composite method.     

解结构动力问题的半隐式Runge-Kutta型并行直接积分法

本文利用三级三阶半隐式Runge-Kutta法解结构动力问题,并用多项式预处理共轭梯度法解有关方程组。提出了半隐式Runge-Kutta型并行直接积分法RK33P。在YH-1机上,与相应的串行算法RK33S相比较,当有关方程组的阶数为10³~10⁴时,加速比可达24~27。     

一种基于约束共轭梯度的闪光照相图像重建算法

 针对闪光照相系统成像信噪比低的特点,提出了正则化预优约束共轭梯度算法（RPCCG）。RPCCG算法在闪光照相重建方程中引入Tikhonov正则化准则,利用预优约束共轭梯度法迭代求图像重建的最优解。数值试验表明,采用最小二乘+平滑准则的RPCCG算法是一种具有较高的抗噪能力的有效闪光照相图像重建算法,具有良好的收敛性和稳定性以及较高的重建精度。     

Application of Robust Incomplete Cholesky Factorizations Preconditioned CG Algorithms for FEM Analysis of Millimeter Wave Filters

The Incomplete Cholesky factorizations preconditioning scheme is applied to the conjugate gradient (CG) method for solving a large system of linear equations resulting from finite element method (FEM) analysis of millimeter wave filters. As is well known, the convergence of CG method deteriorates with increasing EM wave number and in millimeter wave band the eigen-values of A are more and more scattered between both the right and the left half-plane. The efficient implementation of this preconditioned CG (PCG) algorithm is described in details for Complex coefficient matrix. With incomplete factorization preconditioning scheme in the conjugate gradient algorithm, this PCG approach can reach convergence in 20 times CPU time shorter than CG for several typical millimeter wave structures.     

An h-adaptive finite element solver for the calculations of the electronic structures

In this paper, a framework of using h-adaptive finite element method for the Kohn–Sham equation on the tetrahedron mesh is presented. The Kohn–Sham equation is discretized by the finite element method, and the h-adaptive technique is adopted to optimize the accuracy and the efficiency of the algorithm. The locally optimal block preconditioned conjugate gradient method is employed for solving the generalized eigenvalue problem, and an algebraic multigrid preconditioner is used to accelerate the solver. A variety of numerical experiments demonstrate the effectiveness of our algorithm for both the all-electron and the pseudo-potential calculations.     

Application of SSOR Preconditioning Technique to Method of Lines for Millimeter Wave Scattering

In this paper, symmetric successive overrelaxation (SSOR) preconditioned CG technique are introduced into method of lines (MOL) to further enhance the computational efficiency of this semi-analytic method. Millimeter wave scattering by an infinite plane metallic grating is used as the examples to describe its implementation, whose analysis usually needs fast algorithms because of electrically large dimension. For arbitrary incident wave, Helmholz equation and boundary condition are used to calculate the impedance matrix and then to obtain reduced current-voltage linear matrix equation in spatial domain. An effective symmetric successive overrelaxation preconditioned conjugate gradient iterative method, SSOR-PCG, is chosen to solve this matrix equation. With SSOR as the preconditioner as well as its efficient implementing in CG algorithm, PCG method can converge to accurate solution in much fewer iteration steps.     

块多分裂方法与预条件子空间迭代方法

提出一种块多分裂并行PE迭代算法(MPPE),可以克服M^-1r^(s)并行化处理的困难。这种算法格式简单明了,收敛速度快。并证明了当矩阵A是M-阵和H-阵时,该算法是收敛的。同时把这种分裂作为预处理矩阵,对子空间方法类进行了预处理,并给出的计算实例显示该算法很有效,对子空间方法类的余量光滑和加速都起到了比较好的作用。     

On the discrete dynamic nature of the conjugate gradient method

The primary objective of this technical note is to establish an equivalence between the (preconditioned) conjugate gradient (PCG) algorithm and a special central difference based DR procedure, thereby revealing a discrete dynamic nature of the CG iterative procedure. This may therefore provide an alterative viewpoint to gain a further understanding of the CG method and its variants.     

A Two-Level Preconditioned Conjugate-Gradient Method in Distorted and Structured Grids

In this paper, we propose a new two-level preconditioned C-G method which uses the quadratic smoothing and the linear correction in distorted but topologically structured grid. The CPU time of this method is less than that of the multigrid preconditioned C-G method (MGCG) using the quadratic element, but their accuracy is almost the same. Numerical experiments and eigenvalue analysis are given and the results show that the proposed two-level preconditioned method is efficient.