基于Epsilon算法的不确定性结构二阶区间优化方法

该文提出一种求解不确定性结构模态的二阶区间优化算法,首先应用拉格朗日乘子法将带有约束条件的模态优化问题转化为非约束优化,再用区间扩展的二阶泰勒展开式近似表述不确定性结构的模态区间函数。由于其二阶常数项(海森矩阵)的计算十分繁琐,这里采用DFP方法(Davidon and Fletcher-Powell method)近似迭代计算该常数项,同时计算满足约束条件和优化目标的结构参数和参数不确定性区间。在结构重分析中采用Epsilon算法,从而在保证计算精度的同时节省了计算时间。通过算例计算进一步证明该方法对于板壳加筋不确定结构的优化是有效的。     

膜结构极小曲面找形分析的一种线性化近似方法及其有限元线法求解

膜结构的极小曲面找形分析是一个非线性问题,求解时需要进行大量非线性迭代,并需要一个合理的初始解作为收敛的保证,计算繁琐、量大且难度高。该文利用积分中值定理和归一化手段对曲面面积的表达式进行一种特殊的线性化,将原非线性问题转化为线性问题,使问题得到本质性的简化。该线性问题的解答作为原问题高质量的近似解,既可用于结构的初步设计阶段了解膜面的大概形状,亦可作为精细的找形分析中非线性迭代求解的初始解。该线性化方法的误差主要来源于映射参数分布的不均匀性,对于常见的可用平行四边形剖分的膜,其逼近精度相当高。有限元线法(FEMOL)是一种基于常微分方程(ODE)求解的半解析方法,其高度的解析性和解的光滑性特别适合于膜结构的分析。该文采用高次线法单元分析求解转化后的线性问题,只需一次求解,无需任何迭代。数值算例表明:该方法是一种简单、高效、高逼近度的膜结构找形分析方法。     

基于弱牛顿方程的一种无记忆方法

给出了一种求解无约束优化问题的方法,为了求下一次迭代点时它不需要进行矩阵计算,在不精确线搜索下,该方法对一般目标为全局收敛,且对一类特殊的凸函数,其定步长型算法蛮有收敛的,并给出了计算的步骤和计算的数值例子。     

基于混合萤火虫算法的桥梁颤振分析方法

在进行颤振临界状态分析时需要求解高次非线性方程组,为了克服传统解法的缺陷,采用混合萤火虫算法对方程组进行求解。使用双参数优化模型,将桥梁颤振临界状态的求解问题转化为优化问题。为弥补萤火虫算法的不足,在萤火虫算法基础上,将量子遗传算法的量子计算、交叉和变异操作与萤火虫算法相结合,提出一种混合萤火虫算法。最后,通过若干试验对比分析,证实了该优化模型的可靠性及求解方法的有效性。     

基于多步迭代方法的快速总变分图像复原

针对目前总变分图像复原算法收敛速度较慢的问题,本文提出了一类基于多步迭代的快速总变分图像复原算法.提出了两种多步总变分图像复原算法:多步迭代收缩阈值算法和多步迭代加权收缩算法,并针对多步算法每次迭代需要额外估计权参数的不足,给出了一种固定权参数的多步总变分复原算法.同传统的单步总变分复原算法相比,提出算法在每次迭代过程中无需额外增加计算量,且需要很少的迭代就能达到收敛.实验结果表明:对复原不同因素引起的退化图像,文中提出算法的收敛性能远高于传统的单步迭代复原算法,从而我们提出的算法是有效的.     

等效线性化方法中系统参数求解的优化算法

等效周期和有效阻尼比的合理取值是保证等效线性化方法求解精度的关键。该研究利用在自振周期和阻尼比两维空间中的全局优化搜索手段,提出了一种确定等效线性化方法系统参数的标准粒子群优化算法。算例分析验证了优化算法求解的可行性。在保证系统参数合理取值的同时,标准粒子群优化算法的计算效率约为统计分析方法的2．0％,从而克服了统计分...     

基于信赖域技术的非单调非线性共轭梯度算法

共轭梯度算法由于其迭代简单和较小的存储在求解大规模无约束优化问题中起着特殊的作用.本文基于信赖域技术和修正拟牛顿方程,结合Zhang非单调策略,设计了一种新的求解无约束最优化问题的基于信赖域技术的非单调非线性共轭梯度算法.该算法每次迭代自动产生信赖域半径,并通过求解一个简单的子问题得到下一个迭代点,信赖域技术的应用保证...     

基于Woodbury非线性方法的迭代算法对比分析

在结构局部非线性求解过程中,刚度矩阵仅部分元素发生改变,此时切线刚度矩阵可写成初始刚度矩阵与其低秩修正矩阵和的形式,每个增量步的位移响应可用数学中快速求矩阵逆的Woodbury公式高效求解,但通常情况下迭代计算在结构非线性分析中是不可避免的,因此迭代算法的计算性能也对分析效率有重要影响。该文以基于Woodbury非线性方法为基础,分别采用Newton-Raphson(N-R)法、修正牛顿法、3阶两点法、4阶两点法及三点法求解其非线性平衡方程,并对比分析5种迭代算法的计算性能。利用算法时间复杂度理论,得到了5种迭代算法求解基于Woodbury非线性方法平衡方程的时间复杂度分析模型,定量对比了5种迭代算法的计算效率。通过2个数值算例,从收敛速度、时间复杂度和误差等方面对比了各迭代算法的计算性能,分析了各算法适用的非线性问题。最后,计算了5种算法求解基于Woodbury非线性方法平衡方程的综合性能指标。     

求解随机变分不等式问题的随机逼近向前–向后算法

由于其在交通运输、随机博弈和经济均衡等领域中的广泛应用，关于随机变分不等式数值算法的研究受到广泛关注。借助于随机逼近方法，提出了求解随机变分不等式问题的向前–向后线搜索算法，该算法每次迭代只需计算一次到闭凸集上的投影，并且不要求Lipschitz常数信息，从而避免了很多不必要的计算量。在温和的假设下，证明了算法产生的序列几乎处处收敛到随机变分不等式问题的解，以及算法基于自然残差剩余函数的次线性收敛率和迭代复杂度结果。最后，通过数值算例验证了算法的可行性和有效性。     

钢筋混凝土异形柱双向偏心承载力高效算法与程序设计

异形柱结构的发展需要一种高效、精确的承载力求解方法。目前常用的数值积分方法是将截面剖分成二维网格来进行求解,这种方法虽然比较精确,但往往存在计算速度与计算精度之间的矛盾,尤其是计算包含大量构件的结构体系。针对这一问题提出了新的截面数值积分算法。新算法采用特殊的单元剖分技术,实现了有限元法与单元内部解析求解法的有机结合,并且通过构造新的迭代变量加速了迭代运算的过程。与以往算法的对比分析表明,新算法具有更高的运算速度、运算精度以及更为广泛的适用性。     

应用最优化方法辨识指数自回归模型

本文研究应用非线性最优化原理辨识指数自回归模型的方法，以Ｎｅｗｔｏｎ一维搜索和线性回归法为基础，提出一种改进的坐标轮换算法，交替地估计模型的线性和非线性参数。实例表明，该算法对指数自回归模型具有良好的辨识效果。     

Comparison of the born iterative method and tarantola's method for an electromagnetic time-domain inverse problem

Two methods of solving the nonlinear two-dimensional electromagnetic inverse scattering problem in the time domain are considered. These are the Born iterative method and the method originally proposed by Tarantola for the seismic reflection inverse problems. The former is based on Born-type iterations on an integral equation, whereby at each iteration the problem is linearized, and its solution is found via a regularized optimization. The latter also uses an iterative method to solve the nonlinear system of equations. Although it linearizes the problem at each stage as well, no optimization is carried out at each iteration; rather the problem as a whole is posed as a (regularized) optimization. Each method is described briefly and its computational complexity is analyzed. Tarantola's method is shown to have a lower numerical complexity compared to the Born iterative method for each iteration, but in the examples considered, required more iterations to converge. Both methods perform well when inverting a smooth profile; however, the Born iterative method gave better results in resolving localized point scatterers.     

相关正态空间中改进的有限步长迭代法

针对有限步长迭代法在结构功能函数非线性程度极高时,保证收敛的初始步长难于确定的问题,提出了改进的有限步长迭代法。通过实例说明有限步长迭代法出现迂回迭代甚至不收敛的原因。为了保证每一迭代步长为最优步长,该文引入黄金分割法对步长进行一维搜索,并根据增广拉格朗日函数的极值条件构造了一个新的评价函数,给出了相关正态空间中改进的有限步长迭代法的计算步骤。数值算例表明改进的有限步长迭代法的迭代结果正确,在结构功能函数非线性程度极高时收敛性较好,迭代步数少于修正迭代法的步数。     

Bordering method for solving systems of linear equations generated by the finite element method in the plate bending problem

A combined iteration algorithm based on the bordering and conjugate gradient methods is proposed to solve systems of linear equations generated by the finite element method in the plate bending problem. The numerical results for the analysis of the convergence rate of the iterative process are presented in the solution of model problems using a classical and modified algorithm of the method of conjugate gradients. The possibility of acceleration of the iterative algorithm is shown. __________ Translated from Problemy Prochnosti, No. 4, pp. 137–145, July–August, 2007.     

Analysis of 2‒D bimodular materials and wrinkled membranes based on the parametric variational principle and co‒rotational approach

In the numerical analysis of 2?D bimodular materials, strain discontinuity is problematic, and the traditional iterative algorithm is frequently unstable. This paper develops a stable algorithm for the large?displacement and small?strain analyses of 2?D bimodular materials and structures. Geometrically nonlinear formulations are based on the co?rotational approach. Using the parametric variational principle (PVP), a unified constitutive equation is created to resolve the problem induced by strain discontinuity in the local coordinate system. Because the traditional stress iteration is not required, the local linear stiffness matrix does not need to be updated when computing the global stiffness matrix and the nodal internal force vector. The nonlinear problem is ultimately transformed into a complementarity problem that is simply solved by combing the Newton–Raphson scheme and the mature quadratic programming algorithm. Numerical examples demonstrate that the PVP algorithm presents better convergence behavior than the traditional iterative algorithm. By incorporating the concept of material modification, the new algorithm is also be successfully extended to the wrinkling analysis of thin membranes. Copyright © 2014 John Wiley & Sons, Ltd.     

Stabilizing the iterative solution to ultrasonic transmission tomography

A stable iterative reconstruction algorithm is presented for ultrasonic tomography through the uses of weighting matrices, damping factor, and the features of conjugate gradient method. The weighting matrices are chosen according to Berryman's analysis of nonlinear tomography. An implicit conjugate gradient-type iterative formula is used to solve a weighted least-squares problem. In the iterative process, a damping factor of linear increase rather than a constant is used to filter out the influences of smaller singular values' components. The results of simulated data show that our method can efficiently stabilize the iteration, and provide a robust solution.     

Second-order iterative approach to inverse scattering: numerical results

We introduce an iterative algorithm for the reconstruction of dielectric profile functions from scattered field data, in which each step corresponds to the solution of a quadratic inversion problem. This means that, at each iteration, we perform a second-order approximation of the scattering operator connecting the unknown profile to the data about a reference profile function. This procedure is then compared with a linear iterative inversion algorithm, and it is pointed out that, within a prescribed class of profile functions, the linear iterative inversion does not converge to the actual solution, whereas the proposed approach does. This feature can be explained by reference not only to the improved approximation provided by the addition of a further term for profile functions of a larger norm but also to the different classes of functions that can be reconstructed by either the linear or the quadratic model. Numerical examples of profile reconstruction in the scalar two-dimensional geometry, with far-zone scattered field data at a fixed frequency, confirm the theoretical analysis.     

Multi-grid solutions to the elastic plastic torsion problem in multiply connected domains

The elastic plastic torsion problem for an elastic, perfectly plastic cylinder with multiply connected cross section twisted around its longitudinal axis is formulated as an obstacle problem for an associated stress potential, the obstacle being defined in terms of a generalized distance function. Based upon the reformulation of the obstacle problem as an equivalent linear complementarity problem, the latter is discretized by means of finite difference techniques, and a monotonically convergent iterative scheme for its numerical solution is developed. At each step of the iteration the solution of a reduced system of discrete Poisson equations is required which is done by applying multi-grid techniques with respect to a hierarchy of grid-point sets. Combined with a suitably chosen nested iteration process this results in a computationally very efficient algorithm for the approximate solution of the elastic plastic torsion problem.     

A robust incomplete Choleski-conjugate gradient algorithm

When applied to large sparse sets of simultaneous equations, classical iterative methods may yield very poor convergence rates. This paper gives an incomplete Choleski-conjugate gradient algorithm (ICCG) which has reliably good convergence rates at the expense of computing and using at each iteration an incomplete Choleski factor of the coefficient matrix. The method is applicable to any problems in which the coefficient matrix is symmetric positive definite and is likely to be advantageous with respect to elimination when it is not possible to represent the equations in a dense band form.     

LDDA动接触力的迭代算法

概述了接触非线性问题的各种分析方法,指出了用Lagrange Discontinous Deformation Analysis(LDDA)求解缝面接触问题有较大优点,与DDA方法比较,它对接缝的处理更真实、更自然;又因为LDDA对全部接触力进行平衡迭代,所以接触力的计算精度也较高;同时,推导了LDDA的基本公式,提出了一个新的计算动接触力的迭代求解算法-改进的Uzawa迭代算法,该算法的松驰因子是依据块体接触的物理特性而非按照数学理论选用的,算例表明:LDDA理论和与之相对应的动接触力的迭代求解算法具有较快的收敛速度和较高的计算精度。为接触问题的计算提供了一个有效的方法和工具。