一类有限元结构分析的EBE计算方法

本文利用ＥＢＥ策略￣［１］的基本思想，给出一类有限元结构分析的ＥＢＥ计算方法，即ＥＢＥ共轭梯度法ＥＢＥ－ＣＧ和ＥＢＥ预处理共轭梯度法ＥＢＥ－ＰＣＧ，这类方法避免了传统有限元结构分析中总刚度阵的组集而可大大降低存储量要求。同时它们还特别适合在各种粒度下的多处理机系统上实现。初步数值试验结果表明：这类ＥＢＥ有限元结构分析方法对串行和并行计算都是很有效的。     

On preconditioning a nonsingular matrix

For a nonsingular n by n matrix A, a diagonal matrix D^* is derived which minimizes an upper bound on the spectral condition number of DA. Replacement of the linear system Ax=c with the prescaled system D^*Ax=D^*c requires about 3n² operations for dense matrices and fewer for sparse, banded matrices and is recommended for the conjugate gradient and other methods of solution. Examples are given showing the advantageous effect of prescaling on condition number, and a simple computational algorithm is presented. The extension to nondiagonal scaling matrices is discussed.     

Several solution methods for the generalized complex eigenvalue problem with bounded uncertainties

The aim of this paper is to evaluate the effects of uncertain-but-bounded parameters on the complex eigenvalues of the non-proportional damping structures. By combining the interval mathematics and the finite element analysis, the mass matrix, the damping matrix and the stiffness matrix were represented as the interval matrices. Firstly, with the help of the optimization theory, we presented an exact solution—the vertex solution theorem, for determining the exact upper bounds or maximum values and exact lower bounds or minimum values of complex eigenvalues of structures, where the extreme values are reached on the boundary of the interval mass, damping and stiffness matrices. Then, an interval perturbation method was proposed, which needs less computational efforts. A numerical example of a seven degree-of-freedom spring-damping-mass system was used to illustrate the computational aspects of the presented vertex solution theorem and the interval perturbation method in comparison with Deif’s method.     

流体力学问题的三次样条配置法

本文给出了三次样条配置法在流体力学问题数值解中的应用以及在这一领域的新进展。给出了流体力学方程中主要的样条函数关系和解算步骤。所有情形都是便于反演的三对角形矩阵。简要评述了SADI方法和样条方法在每一坐标方向的分步计算方法、截断误差和稳定性。给出了处理混合边界条件的一般公式。最后简要讨论了样条近似引起的数值弥散和耗散。      

Poiseuille-Bénard流的出口边界条件及其谱元法计算

研究二维矩形管道中底部加热的不可压缩Ｐｏｉｓｅｕｉｌｌｅ－Ｂｅｎａｒｄ流的谱元法数值计算问题．讨论各种不同的出口边界条件的处理及其对谱元法数值模拟的影响．通过干扰区、干扰幅度和计算时间的比较,确定比较理想的出口边界条件．     

一种基于证据理论的结构可靠性分析方法

提出了一种基于证据理论的结构可靠性高效求解方法. 通过构造优化问题求解极限状态方程的非概率可靠性指标及设计验算点, 并构造一辅助区域. 通过辅助区域显著减少需要进行极值分析的焦元个数, 并基于区间分析方法减少焦元上极限状态方程的计算次数, 从而有效降低计算成本. 数值算例及工程应用验证了该方法的有效性.     

An inversion algorithm for general tridiagonal matrix

An algorithm for the inverse of a general tridiagonal matrix is presented. For a tridiagonal matrix having the Doolittle factorization, an inversion algorithm is established. The algorithm is then generalized to deal with a general tridiagonal matrix without any restriction. Comparison with other methods is provided, indicating low computational complexity of the proposed algorithm, and its applicability to general tridiagonal matrices.     

An Analysis of a Hybrid Optimization Method for Variational Data Assimilation

In four-dimensional variational data assimilation (4D-Var) an optimal estimate of the initial state of a dynamical system is obtained by solving a large-scale unconstrained minimization problem. The gradient of the cost functional may be efficiently computed using the adjoint modeling, at the expense equivalent to a few forward model integrations; for most practical applications, the evaluation of the Hessian matrix is not feasible due to the large dimension of the discrete state vector. Hybrid methods aim to provide an improved optimization algorithm by dynamically interlacing inexpensive L-BFGS iterations with fast convergent Hessian-free Newton (HFN) iterations. In this paper, a comparative analysis of the performance of a hybrid method vs. L-BFGS and HFN optimization methods is presented in the 4D-Var context. Numerical results presented for a two-dimensional shallow-water model show that the performance of the hybrid method is sensitive to the selection of the method parameters such as the length of the L-BFGS and HFN cycles and the number of inner conjugate gradient iterations during the HFN cycle. Superior performance may be obtained in the hybrid approach with a proper selection of the method parameters. The applicability of the new hybrid method in the framework of operational 4D-Var in terms of computational cost and performance is also discussed.     

Adaptive selection of parameters for precise computation of matrix exponential based on padé approximation

讨论了基于Pad\'{e}逼近的矩阵指数精细积分方法中加权系数N 和展开项数q的自适应选择问题. 参数(N,q)的选择直接影响到矩阵指数计算的精度和效 率. 采用矩阵函数逼近理论，研究了参数N和q的增加对精度的影响程度，据此，提出了 参数(N,q)优化组合的递推自适应选择方法. 该方法可以根据矩阵本身的性态选择合适的参 数(N,q),而参数选择的计算量与矩阵指数的计算量相比几乎可以忽略，这对于增强矩阵指 数精细积分方法的适应性和提高计算效率是很有益处的. 算例验证了该方法的正确性和有效性.     

The Explicit Solution of the Matrix Equation AX-XB=C

ＴＨＥＥＸＰＬＩＣＩＴＳＯＬＵＴＩＯＮＯＦＴＨＥＭＡＴＲＩＸＥＱＵＡＴＩＯＮＡＸ－ＸＢ＝Ｃ──ＴｏｔｈｅｍｅｍｏｒｙｏｆＰｒｏｆ．ＧｕｏＺｈｏｎｇｈｅｎｇＣｈｅｎＹｕｍｉｎｇ（陈玉明）ＸｉａｏＨｅｎｇ（肖衡）（ＲｅｃｅｉｖｅｄＦｅｂ．．２８．１９９５...     

An acceleration for the eigensystem realization algorithm with partial singular values decomposition

The real-time identification of dynamic parameters is important for the control system of spacecraft. The eigensystem realization algorithm (ERA) is currently the typical method for such application. In order to identify the dynamic parameter of spacecraft rapidly and accurately, an accelerated ERA with a partial singular values decomposition (PSVD) algorithm is presented. In the PSVD, the Hankel matrix is reduced to dual diagonal form first, and then transformed into a tridiagonal matrix. The eigenvalues are computed by the bisection method in terms of the Sturm property, and the corresponding eigenvectors are obtained by the inverse iteration method. Finally, the eigenvalues and the eigenvectors are transformed into the singular values and the singular value vectors of the original matrix. An example for space station is presented to demonstrate the efficacy and accuracy of the proposed algorithm.     

Stochastic model reduction for robust dynamical characterization of structures with random parameters

In this paper, we characterize random eigenspaces with a non-intrusive method based on the decoupling of random eigenvalues from their corresponding random eigenvectors. This method allows us to estimate the first statistical moments of the random eigenvalues of the system with a reduced number of deterministic finite element computations. The originality of this work is to adapt the method used to estimate each random eigenvalue depending on a global accuracy requirement. This allows us to ensure a minimal computational cost. The stochastic model of the structure is thus reduced by exploiting specific properties of random eigenvectors associated with the random eigenfrequencies being sought. An indicator with no additional computation cost is proposed to identify when the method needs to be enhanced. Finally, a simple three-beam frame and an industrial structure illustrate the proposed approach.     

THE EXPLICIT SOLUTION OF THE MATRIX EQUATION AX-XB=C——To the memory of Prof. Guo Zhongheng

Almost all of the existing results on the explicit solutions of the matrix equation AX−XB=C are obtained under the condition that A and B have no eigenvalues in common. For both symmetric or skewsymmetric matrices A and B, we shall give out the explicit general solutions of this equation by using the notions of eigenprojections. The results we obtained are applicable not only to any cases of eigenvalues regardless of their multiplicities, but also to the discussion of the general case of this equation. To the memory of Prof. Guo Zhongheng Project Supported by the National Natural Science Foundation of China     

大型结构特征值问题的Lanczos子结构并行算法

本文利用子结构和Lanczos方法,提出了大型结构固有频率与模态的并行解法。该方法在Lanczos方法的求解过程中,仅利用子结构刚度阵和质量阵并行进行凝聚,进而求得新的迭代矢量,最终求得三对角阵对应的特征值和特征向量。该算法在西安交通大学ELXSI-6400并行计算机上程序实现,计算结果表明能有效地节省计算时间和计算机的内存,为一种有效的大型工程结构动力问题的求解方法。     

一类生物流体力学模型的稳定性

在研究颅内压集中参数模型的基础上，改进自适应的龙格-库塔法对一类生物流体力学模型进行数值模拟。通过合理控制计算量，得到了微分方程近似解的局部截断误差的估计。使用矩阵特征值分析微分方程的稳定性，在实际生理范围内变化模型参数，得到特征值随参数变化的关系，指出模型中存在实部为正的特征值。文章还讨论了控制矩阵特征值的变化对数值方法稳定步长的影响，并得到步长的取值范围。通过理论分析。得到控制方程的解随时间的发展和模型中生理参数的选取相关。分析了特征值变化的稳定性和参数的关系，并对模型进行改进，讨论其稳定性的情况。     

基于全矢量显式算法注射填充软件的研究与开发

基于前人模拟注射填充的显式算法,提出了一种新的全矢量化运算显式算法。对于各类注射填充的模拟,此算法避免了压力场的全局求解,矩阵操作仅在单元一级进行,通过反馈修正,在各时间增量步使已填充的区域满足不可压缩条件。由于避免了所有的全局耦合求解,计算代价与节点的自由度近似成正比,可实现填充模拟过程的高效运算。由于矢量化算法是完全脱耦的,容易实现并行计算。基于这种新算法,应用有限元法进行了注射填充仿真软件的初步开发。其填充模拟结果与商业软件MOLDFLOW的三维模拟结果比较,证实了该软件的有效性。     

FAST ALGORITHMS FOR DETERMINING POSITIVE DEFINITENESS OF LARGE SCALE MATRICES IN SOLID STRUCTURE ANALYSIS

对于结构稳定性分析中超大规模矩阵正定性判定,必须采用并行计算方法,传统方法如计算特征值、主子式行列式及LDLT等直接方法难以实现.本文提出了一些可适用于并行的迭代判定算法.借鉴力学系统中能量下降的思想,发展了一种判定矩阵正定性的新思路,即将矩阵的正定性判定问题转化为一个优化问题,并基于优化算法来判定矩阵的正定性.提出了基于最速下降法和共轭梯度法来进行矩阵正定性判定的算法.然后考虑到力学系统刚度矩阵的稀疏性和结构刚失稳状态的弱非正定性,提出可以先截超平面后解方程求驻值点的方法来判定弱非正定矩阵的正定性.为了保证对强非正定矩阵判定的准确性,本文提出可以高效混杂使用截平面法和共轭梯度法.数值实验结果表明,本文提出的算法具有准确性和高效性.对于强非正定矩阵而言,共轭梯度算法更加高效;而对于弱非正定矩阵,则是截平面法和混杂算法更加高效.这些算法都容易在机群上实现并行计算,能够快速判定大规模矩阵的正定性.     

On the optimal preconditioning of nonsingular matrices

This work extends a previous result in which a diagonal matrix D^* was found minimizing an upper bound on the spectral condition number of the nonsingular matrix DA. Here, minimizing matrices G are found from the set of bounded lower triangular matrices. Each of the resulting linear systems GAx=Gc is expected to be more suitable than Ax=c for numerical solution by the conjugate gradient and other methods.     

Hopf分岔的代数判据及其在车辆动力学中的应用

利用Hurwitz行列式,给出平衡点失稳而发生Hopf分岔的代数判定准则和计算方法,这一方法将Hopf分岔点的求解转化为一个非线性方程的求解问题,从而克服了以前方法在计算Hopf分岔点时,对于参数的每一次变化通过求特征根并判定特征根的实部是否为零的庞大工作量。应用这一方法,我们进行了非线性车辆系统蛇行运动稳定性的研究,得到了轮对系统发生蛇行运动的临界速度的解析表达式。     

基于自然单元法的功能梯度中厚板自由振动分析

自然单元法是一种基于自然邻近插值的无网格数值方法．相对于移动最小二乘近似而言,自然邻近插值不涉及到复杂的矩阵求逆运算,也不需要任何人为的参数．基于一阶剪切变形板理论,利用自然单元法对功能梯度中厚板的自由振动进行了数值分析．功能梯度板材料属性沿厚度方向呈梯度连续变化．由于自然邻近插值函数具有Kronecker delta函数性质,可以直接施加本质边界条件．通过本文给出的方法,对不同梯度指数和不同边界条件的功能梯度中厚板的振动频率进行了计算．通过与文献结果的对比验证了自然单元法求解的有效性．