基于牛顿-拉夫逊的拉压不同模量问题的数值求解

针对拉压模量不同引起的材料本构非线性,首先,通过引入改进的Heaviside函数将本构方程连续光滑化;然后,基于特征值与特征向量的求导策略,推导有限元求解模型中切线刚度矩阵的列式;最终,提出基于牛顿-拉夫逊迭代格式的拉压不同模量问题的数值求解算法。数值算例验证了本文算法比传统算法具有更稳定的收敛性和更高的求解精度,特别适合于工程分析中大规模计算问题的求解。     

基于牛顿-拉夫逊的拉压不同模量问题的数值求解

针对拉压模量不同引起的材料本构非线性,首先,通过引入改进的Heaviside函数将本构方程连续光滑化;然后,基于特征值与特征向量的求导策略,推导有限元求解模型中切线刚度矩阵的列式;最终,提出基于牛顿-拉夫逊迭代格式的拉压不同模量问题的数值求解算法。数值算例验证了本文算法比传统算法具有更稳定的收敛性和更高的求解精度,特别适合于工程分析中大规模计算问题的求解。     

蚁群算法求解二维拉压不同模量反问题

利用光滑函数技术对二维拉压不同模量本构关系进行光滑化处理,采用初应力方法求解二维拉压不同模量正问题的有限元方程。在此基础上,建立了基于连续域蚁群算法的二维拉压不同模量反问题的数值求解模型,考虑了区域非均质的影响,实现了对拉压弹性模量和泊松比的单一/组合识别。通过两个数值算例,对所提算法进行了数值验证,分别探讨了蚁群算法相关参数、测点分布和数据噪音等对识别结果的影响。数值验证表明,所提算法可有效地求解二维拉压不同模量反问题,并具有较好的计算精度。     

光滑函数法求解拉压不同弹性模量问题

采用光滑函数技术,对拉压不同弹性模量问题的应力应变关系进行光滑处理,可避免迭代中应力状态的判断,方便计算。同时建立了相应的基于初应力技术的有限元计算模式,仅需对刚度阵三角化一次,避免了考虑剪切刚度带来的不便。文中通过不同算例,对所提算法进行了数值验证,与解析解相比有很好符合。此外,对不同拉压模量的热应力分析进行了初步探讨。     

索网天线的参变量变分及非线性有限元方法

针对大型周边桁架式索网天线由拉索拉压模量不同引起的本构非线性和结构大变形引起的几何非线性问题,给出了基于参变量变分原理的几何非线性有限元方法. 首先针对含预应力索单元拉压模量不同分段描述的本构关系,通过引入参变量,导出了基于参变量及其互补方程的统一描述形式,避免了传统算法需要根据当前变形对索单元张紧/松弛状态的预测,提高了算法收敛性. 然后利用拉格朗日应变描述索网天线结构大变形问题,结合几何非线性有限元法,建立了基于参变量的非线性平衡方程和线性互补方程;并给出了牛顿-拉斐逊迭代法与莱姆算法相结合的求解算法. 数值算例验证了本文提出的算法比传统算法具有更稳定的收敛性和更高的求解精度,特别适合于大型索网天线结构的高精度变形分析和预测.      

THE PARAMETRIC VARATIONAL PRINCIPLE AND NON-LINEAR FINITE ELEMENT METHOD FOR ANALYSIS OF ASTROMESH ANTENNA STRUCTURES

针对大型周边桁架式索网天线由拉索拉压模量不同引起的本构非线性和结构大变形引起的几何非线性问题,给出了基于参变量变分原理的几何非线性有限元方法. 首先针对含预应力索单元拉压模量不同分段描述的本构关系,通过引入参变量,导出了基于参变量及其互补方程的统一描述形式,避免了传统算法需要根据当前变形对索单元张紧/松弛状态的预测,提高了算法收敛性. 然后利用拉格朗日应变描述索网天线结构大变形问题,结合几何非线性有限元法,建立了基于参变量的非线性平衡方程和线性互补方程;并给出了牛顿-拉斐逊迭代法与莱姆算法相结合的求解算法. 数值算例验证了本文提出的算法比传统算法具有更稳定的收敛性和更高的求解精度,特别适合于大型索网天线结构的高精度变形分析和预测.     

基于多节点微桁架元的薄板锚固区拉压杆模型

目前混凝土桥梁结构D区多依据应力迹线的走向以及应力积分来确定拉压杆模型的基本构形, 模型中杆件 的布置存在一定的困难和随意性.针对此问题, 以后张薄板锚固区为研究对象, 提出一种多尺度分区渐进结构优化算法.首先, 通过一种新的多节点微桁架元构建相应的等效基结构.然后利用此算法对不同偏心锚固力作用下的等效基结构进行分析, 以自动生成锚固区拉压杆模型.最后, 将该拉压杆模型与美国AASHTO桥梁设计规范中推荐的模型进行了对比.研究结果表明:该算法能够定量化确认拉压杆模型中杆件的位置, 特别是拉杆位置与数值分析得到的受拉区合力作用线能较好吻合.     

三维接触问题的非光滑算法

给出了一种非光滑算法直接用于求解三维摩擦接触问题的不可微非线性互补模型,不再对模型进行光滑化处理,使算法更加简单。文中对非光滑算法的收敛性给出了严格的数学证明,数值实验表明该算法列式简单,但与光滑化算法同样有效。     

两级敏度分析求解双弹性模量桁架结构反问题

利用光滑函数和有限元技术,建立了求解双弹性模量桁架结构正问题的数值模型,可方便推导刚度对位移、位移对弹性模量的敏度计算公式。在此基础上,提出了基于两级敏度分析求解双弹性模量桁架结构反问题的数值模型,正问题与反问题分别采用Newton-Raphson与Gauss-Newton算法进行求解,数值验证给出了令人满意的结果,并考虑了数据噪音的影响。     

基于边界元法的弹性动力问题并行求解

为了扩大结构弹性动力分析的规模和提高分析速度,在微机机群环境下给出了两种基于边界元法的瞬态问题并行求解算法,即并行拉普拉斯变换求解算法和并行时域求解算法.并行拉氏变换法通过拉氏变换隐去时间变量,由各结点机独立求解各自负责的变换边界元问题.并行时域法采用与时间有关的基本解,使得边界元系统矩阵可以实现时间域上的并行形成.系数矩阵采用卷帘存储,以保持负载平衡.通过矩阵向量运算的并行化实现时间步进算法的并行化.理论分析和数值试验结果表明:两种算法都具有较好的并行性能.可以用于大型问题的高效求解.     

时域自适应算法求解移动荷载下梁的多宗量反问题

通过一种时域自适应算法,建立了求解变速移动荷载下梁的多宗量反问题的数值模型,可同时识别移动荷载和梁的物性参数.正问题采用时域自适应算法和FEM建模,并可由此方便地推导敏度公式;在反问题求解中采用Levenberg-Marquardt法,计算表明该方法具有较好的抗不适定性.通过两个算例,对所提算法进行了数值验证,并探讨了噪声和测点的变化对反演结果的影响,结果令人满意.     

Investigation of Near-Surface Mechanical Properties of Materials Using Atomic Force Microscopy

As ultra-thin films or small-scale structures become widely used in electronics and biology, knowledge concerning their near-surface mechanical properties of the materials is increasingly important. Atomic force microscopy (AFM) is employed to determine near-surface elastic modulus via force-penetration curves acquired during indentation. Samples include polydimethylsiloxane (PDMS), parylene, mica, and single-crystal silicon, and indentations are performed with single-crystal silicon and silicon nitride AFM tips. An analysis algorithm based on the secant modulus method is proposed to extract the true penetration curves from the experimental displacement curves. The penetration data is then analyzed in terms of Hertzian model to estimate the elastic modulus. Three concerns in applying nanoscale AFM indentation to the measurement of the elastic modulus of an ultra-thin material are addressed. First, the effect of the lateral force caused by the inclined angle of the cantilevered probe is investigated theoretically and by numerical simulation. A second concern is local plastic deformation induced by a sharp probe tip. In this case, numerical results show a relatively small effect on the force-penetration curves if the plastic deformation is limited to the central area below the probe tip. The deviation of the elastic-plastic simulation from the elastic estimation depends on the yield strength of the material. Finally, the effect of stiffness matching between the AFM probe and the sample is a key issue that is studied numerically, and appropriate stiffness matching criteria are suggested.     

基于拓扑优化的结构弹性成像方法

弹性模量是工程材料重要的性能参数, 可以衡量物体抵抗弹性变形的能力. 弹性成像是一种通过弹性模量表征生物体组织物理特性的医学成像方法. 为了将弹性成像应用于机械装备结构的缺陷识别, 提高弹性成像的局部表征和全局识别能力, 提出一种基于拓扑优化的结构弹性成像方法. 受拓扑优化理论启发, 采用结构离散单元的相对密度(弹性模量系数)作为弹性成像参数来表征损伤程度, 建立成像参数与弹性模量的插值模型, 基于有限元模型构建损伤表征、结构模型与物理响应的映射关系. 以损伤结构和无损结构位移响应的最小二乘为目标函数, 以成像参数上下限为约束, 建立结构弹性成像的优化模型. 以伴随法推导弹性成像问题的灵敏度, 并详细给出了弹性成像反演的数值实施方式. 二维悬臂梁和米歇尔梁算例表明, 本文提出的基于拓扑优化的弹性成像方法无需先验损伤信息, 可有效实现均质/非均质、单/多缺陷结构的弹性成像, 且弹性成像结果不依赖于特定的边界条件, 进一步将弹性成像方法扩展至三维悬臂梁问题验证了其通用性.      

圆薄板材料性能随温度变化的大挠度弯曲分析

记入材料性能参数随温度的变化，导出了横向力作用下圆薄板轴对称大挠度弯曲的位移型控 制方程，并利用伽辽金方法求出了圆薄板大挠度弯曲的近似位移解；针对一个简例进行了理 论计算和有限元数值模拟，二者结果一致. 在此基础上分析了圆薄板大挠度弯曲的规律及材 料参数随温度变化的热软化效应，给出了相关结果.     

Time domain modeling of damping using anelastic displacement fields and fractional calculus

A fractional derivative model of linear viscoelasticity based on the decomposition of the displacement field into an anelastic part and elastic part is developed. The evolution equation for the anelastic part is then a differential equation of fractional order in time. By using a fractional order evolution equation for the anelastic strain the present model becomes very flexible for describing the weak frequency dependence of damping characteristics. To illustrate the modeling capability, the model parameters are fit to available frequency domain data for a high damping polymer. By studying the relaxation modulus and the relaxation spectrum the material parameters of the present viscoelastic model are given physical meaning. The use of this viscoelastic model in structural modeling is discussed and the corresponding finite element equations are outlined, including the treatment of boundary conditions. The anelastic displacement field is mathematically coupled to the total displacement field through a convolution integral with a kernel of Mittag–Leffler function type. Finally a time step algorithm for solving the finite element equations are developed and some numerical examples are presented.     

Motion of the interface of two fluids with formation of a region of two-phase flow

A study is made of the problem of the displacement of one fluid by another with the formation of a region of combined flow in the case of an elastic flow regime in the region of the displaced fluid. A self-similar solution is constructed for the flow equations averaged with respect to the vertical coordinate. A numerical algorithm is developed for determining the saturation in the region of the mixture, the pressure, and also the position and shape of the interface.Translated from Izvestiya Akademii Nauk SSSR, Mekhanika Zhidkosti i Gaza, No. 8, pp. 78–83, October–December, 1981.We thank A. A. Barmin for a helpful discussion of the work.     

结构优化半解析灵敏度分析的改进算法

提出了结构半解析灵敏度分析的改进算法,该算法实现简便,对于设计变量摄动步长具有极佳的数值稳定特性。首先,从总体角度推导静力问题半解析法灵敏度分析新算法,提出了位移与应力灵敏度列式,并给出了算法实施途径;然后,将此思路推广于自振频率、屈曲临界荷载和瞬态响应等多种问题,提出了相应的计算步骤。以梁单元与壳单元等典型结构为例,开展了多个算例测试。测试表明,改进算法计算精度和效率均有提升,特别是设计变量步长有更大的数值稳定区域,为复杂工程结构形状优化的灵敏度分析提供了新途径。     

Determination of elastic-plastic stresses and strains from measured surface strain data

A rigorous approach founded in the fundamental principles of plasticity is used to develop an accurate numerical algorithm for the determination of stresses and elastic and plastic strains from total strain data measured on a structure surface. The approach used to develop the algorithm and its relationship to both the flow theory of plasticity and recent advances in tangent stiffness-based numerical solution procedures for elastic-plastic boundary value problems are presented. Verification of the method for plane stress problems is demonstrated. A discussion of how the method can be used with measured surface displacement data is proved.     

Domain Decomposition Methods for Sensitivity Analysis of a Nonlinear Aeroelasticity Problem

Abstract

We consider the nonlinear aeroelasticity problem of the interaction between a viscous, incompressible fluid and Lin elastic solid undergoing large displacement. The non-linearities of the problem formulation include the solid and fluid governing equations. as well as thc dependence of the How geometry on the solid deformation. The resulting coupling is thus two-way. We develop domain-decomposition methods for solution and sensitivity analysis of the coupled problem. The domain decomposition is in the form of a block-Gauss-Seidel-like prcconditioncr that decomposes ihc coupled-domain problem into distinct nonovcrlapping fluid and solid subdotnain problems. The preconditioner thus enables exploitation or single-domain algorithms for solid and fluid mechanics discretization and solution. On the other hand, two-way fluid-solid coupling is retained within the residuals, which is essential for correct sensitivities. Sensitivities of field quantities can be found with little additional work beyond that required for solving the coupled fluid-solid system. The methodology developed here is illustrated by the solution of a problem of viscous incompressible flow about an infinite clastic cylinder. Sensitivities of the resulting velocity and displacement fields with respect to elastic modulus and fluid viscosity are computed.