Elimination of Hot Cracking in Laser Beam Welding

Hot cracking is one of the big problems in laser beam welding. By multi‐beam welding, first suggested in the 70‐s (cf. [1], [4]), hot cracking can be avoided. Hereby two additional laser beams are employed. These beams generate a compression which compensate for the critical tensile strain in the solid‐liquid region of the weld induced by the main laser beam. However, hot cracking can only be prevented if the positions, sizes, and powers of the additional laser beams are suitably chosen, i.e. are optimized. Non‐optimal values can even enhance hot cracking. Until now these quantities have been found either by trial and error or by prescribing them intuitively. In the present paper a constrained nonlinear programming problem is formulated to solve the problem of hot crack initiation by minimizing the accumulated transverse strain, i.e. the opening displacement, in the solid‐liquid region. This approach is based on the so‐called strip expansion technique, cf. [2]. (© 2004 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

New Applications of the Boundary Element Method in Dynamics of Solids

The boundary element method (BEM) is developed and applied in new fields of dynamic fracture mechanics, dynamics of composite, elasto–plastic and piezoelectric materials. The BEM results are compared with solutions computed by the finite element method (FEM) showing high accuracy of the BEM. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

吴消元法与四元术

在解多项式方程组的过程中,吴消元法的核心是用对多项式约化求余式的方法消元.研究中发现,清代沈钦裴四元消法的三条法则均系互乘对消,都可以写成除法变换的形式.从而找到吴消元法与四元术的内在联系.得出吴消元法是四元术的直接继承,吴消元法是四元术现代化发展的结论.     

Ground States in the Spin Boson Model

We prove that the Hamiltonian of the model describing a spin which is linearly coupled to a field of relativistic and massless bosons, also known as the spin-boson model, admits a ground state for small values of the coupling constant λ. We show that the ground-state energy is an analytic function of λ and that the corresponding ground state can also be chosen to be an analytic function of λ. No infrared regularization is imposed. Our proof is based on a modified version of the BFS operator theoretic renormalization analysis. Moreover, using a positivity argument we prove that the ground state of the spin-boson model is unique. We show that the expansion coefficients of the ground state and the ground-state energy can be calculated using regular analytic perturbation theory.     

The Operator Factorization Method in Inverse Obstacle Scattering

The standard factorization method from inverse scattering theory allows to reconstruct an obstacle pointwise from the normal far field operator F. The kernel of this method is the study of the first kind Fredholm integral equation (F^* F)^1/4 f = Φ_z with the right-hand part In this paper we extend the factorization method to cover some kinds of boundary conditions which leads to non-normal far field operators. We visualize the scatterer explicitly in terms of the singular system of the selfadjoint positive operator F_# = [(ReF)^* (ReF)]^1/2 + ImF. The following characterization criterium holds: a given point z is inside the obstacle if and only if the function Φ_z belongs to the range of F_#^1/2. Our operator approach provides the tool for treatment of a wide class of inverse elliptic problems.     

Operator Method of Solving Nonlinear Differential Equations

We discuss the possibility of applying linear structures for solving nonlinear differential equations.     

离散算子差分法的单元函数

给出了弹性力学离散算子差分法的离散格式,并给出了该方法的几个板弯曲单元和平面四边形单元,通过对它们的考察,分析了离散算子差分方法中的离散格式对单元位移函数的反映能力。在离散算子差分方法中,无论单元位移函数是否协调,其位移函数均能在离散格式中得到十分好的再现,说明了离散算子差分方法的离散格式是一种性能很优良的离散格式。     

微分算子方法在积分中的应用

函数的微分运算化积分运算容易,把积分运算化为微分运算这是一大难题.利用微分算子方法可以把某些积分运算化为微分运算,且使形如∫f(x)eαxsinβxdx,∫f(x)eαxcosβxdx的积分运算简便、快捷.     

An Approximation Method in the Variational Theory of the Spectrum of Operator Pencils

An approximation method, based on a theorem on approximating general operator-valued functions by piecewise-linear ones, is presented and analyzed. Using this method, variational characteristics of the spectrum of a class of operator functions are established.     

Operator Function Method in the Problem of Normal Waves in an Inhomogeneous Waveguide

The problem of normal waves in a closed (screened) regular waveguiding structure of arbitrary cross-section is considered by reducing it to a boundary value problem for the longitudinal electromagnetic field components in Sobolev spaces. The variational statements of the problem is used to determine the solution. The problem is reduced to studying an operator function. The properties of the operators contained in the operator function necessary to analyze its spectral properties are studied. Theorems on the spectrum discreteness and the distribution of characteristic numbers of the operator function on the complex plane are proved. The problem of completeness of the system of root vectors of the operator function is considered. The theorem on the double completeness of the system of root vectors of the operator function with finite deficiency is proved.     

接触问题中的快速凝缩消元法

用分离解法求解弹性接触问题时，在增量加载和迭代过程中，由于接触区某些节点的状态发生改变而导致方程组的系数矩阵某些行和列元素随之变化。根据此特点，本推导了一种新的自适应迭代算法-快速凝缩消元法，并给出具体的迭代步骤，避免了系数矩阵变化时必须重新形成矩阵的重复计算。     

固体中短波传播的单位分解有限元法

提出了固体中短波传播数值模拟的单位分解有限元法．有限元空间由形成单位分解的标准等参有限元形函数乘以定义为局部子空间基函数的特殊形函数构成．特殊形函数使试空间中包含了关于波动方程的已有知识,因而在单个单元内能近似地再现高度振荡性质．数值例题显示了所提出单位分解有限元在计算精度和效率上的良好性能．     

解第一类算子方程的一种新的正则化方法

对算子与右端都为近似给定的第一类算子方程提出一种新的正则化方法，依据广义Ａｒｃａｎｇｅｌｉ方法选取正则参数，建立了正则解的收敛性。这种新的正则化方法与通常的Ｔｉｋｈｏｎｏｖ正则化方法相比较，提高了正则解的渐近阶估计。     

On the Two-step Newton Method for the Solution of Nonlinear Operator Equations

Building on the method of Kantorovich majorants, we give convergence results and error estimates for the two-step Newton method for the approximate solution of a nonlinear operator equation.     

关于TOPSIS法应用中的逆序问题及消除的方法

在多指标决策问题中，TOPSIS法是一种常用的方法，然而传统的TOPSIS法在实际应用中容易产生逆序的现象。本分析了逆序产生的主要原因，并提出了一种改进的方法-RTOPSIS法。应用本的方法不仅能消除逆序的现象，而且还能正确反映指标权重对决策结果的影响。     

吴消元法在Lagrange和Hamilton方程中的应用

主要借鉴吴消元法,研究带约束动力学中多项式类型Lagrange方程和Hamilton方程,提出了一种求约束的新算法．与以前算法相比,新算法无需求Hessian矩阵的秩,无需判定方程的线性相关性,从而大为减少了计算步骤,且计算更为简单．此外,计算过程中膨胀较小,且多数情形下无膨胀．利用符号计算软件,新算法可在计算机上实现．     

差分方程解的数值余量消元法

本文提出一种新的消元方法,该法利用数值的直接迭代产生余量方程,从而构成已消去很多未知量的线性方程组.本文的方法具有求解简便、精确和快速的优点.     

关于TOPSIS法应用中的逆序问题及消除的方法

在多指标决策问题中,TOPSIS法是一种常用的方法,然而传统的TOPSIS法在实际应用中容易产生逆序的现象.本文分析了逆序产生的主要原因,并提出了一种改进的方法-RTOPSIS法.应用本文的方法不仅能消除逆序的现象,而且还能正确反映指标权重对决策结果的影响.     

Banach空间中半光滑算子方程的不精确牛顿法(英文)

本文主要解决Banach空间中抽象的半光滑算子方程的解法.提出了两种不精确牛顿法,它们的收敛性同时得到了证明.这两种方法可以看作是有限维空间中已存在的解半光滑算子方程的方法的延伸.     

Banach空间中非光滑算子方程的光滑化拟牛顿法

研究Banach空间中非光滑算子方程的光滑化拟牛顿法.构造光滑算子逼近非光滑算子,在光滑逼近算子满足方向可微相容性的条件下,证明了光滑化拟牛顿法具有局部超线性收敛性质.应用说明了算法的有效性.