数值分析和光弹性实验的几种混合解法

本文采用数值分析和光弹性的几种混合法解决结构的应力分析问题。如再应用图象处理和计算机技术,必将促使光弹性实验向着自动方向发展。在上述的几种混合法中,根据结构的不同类型,可以选用光弹性与最小二乘配点的混合法,光弹性与富里哀级数展开的边界值的混合法,光弹性与边界元的混合法或光弹性与边界元、差分联合的混合法。文中还谈及主应力方向的求法,文末有具体算例。本文作者已编制出通用程序,在微机上对平面应力问题一些简例作了具体计算,其结果与解析解比较令人满意。     

边界元法与光弹性实验相结合的应力分析法

本文介绍了边界无法与光弹性实验相结合进行应力分析的方法。叙述了用边界无法求解主应力的基本原理。结合光弹性实验等差线条纹图,可分离出主应力。     

正交异性光弹性应力分离的边界元法

本文对平面正交各向异性复合材料模型引入正应力线性和及边界上正应力线性和流的概念,提出从应力相容方程出发.用边界元法计算正交异性光弹性模型内任一点的正应力线性和位的方法,再与正交异性光弹性法中所给出的应力同的关系结合,即可进行正交异性光弹性应力的分离.最后,对边界元方法的精度进行了讨论.     

动态光弹性方法的主应力分离的研究

本文提出了动态光弹性、动态焦散线实验方法同边界元法结合的混合法，并用这种方法解决了动态光弹性主应力分离的问题，首先对现有的多火花高速摄影系统进行了改造，在动态实验过程中，成功地得到了不同瞬时的清晰的动态光弹性的等差线条纹和动态焦散线系列图像，这样，便可提取不同瞬时的边界应力值、全场主应力差值及边界上的外载。继而提出用Ｌａｐｌａｃｅ变换域上的边界元法来求解在冲击载荷作用下二维弹性模型全场的主应力和。最后，以受冲击载荷作用的圆盘为例，进行实验及边界元法计算，得到了分离的主应力场。     

冲击作用下含孔洞纤维增强复合材料平板的动力学实验与数值研究

作为复合材料动力学实验与数值研究的应用实例 ,实验研究采用正交异性动态光弹性方法 ,数值分析运用各向异性介质的时域边界元方法。纤维增强光弹性复合材料平板被用来模拟含孔洞的正交异性半无限域 ,用小口径步枪施加与纤维方向成 0及 90两个方向的冲击载荷 ,在正交异性动态光弹性实验中记录了应力波在孔洞周围的传播、反射与绕射过程 ,此过程被进一步转换成应力分量的变化时程 ,并与相应的时域边界元方法的数值分析结果进行了比较。     

网状系杆拱连续梁桥的光弹性应力分析

采用光弹性冻结应力法对网状系杆拱连续梁桥进行了不同工况下的应力分析，给出了边界应力分布和主应力迹线，为优化工程设计提供了参考依据．     

数字光弹性法综述

将光弹性法与计算机图像处理技术相结合 ,来自动采集光弹性数据和分析应力的方法 ,称为数字光弹性法 ,与传统光弹性法相比 ,它可以进一步提高实验速度和精度。本文详细讨论了以下两个方面 :一是光弹性条纹的细化和倍增处理技术 ;二是自动确定光弹性参数的技术 ,包括相移法、傅立叶变换法、逐步载荷法、广谱分析法和RGB光弹性法等。通过对近二十年来国内外在这些方面的研究、应用和进展作了综述 ,认为采用白光的彩色域相移技术计算光弹性等倾角 ,结合采用白光源或三色光源的相移法来确定光弹性等色线级数 ,有望成为解决静态二维和三维冻结模型薄切片应力分析的最佳方法 ;另外 ,设计一种能实时和同步采集多幅条纹图的实验装置 ,通过相移法来自动获取动态光弹性数据 ,是数字动态光弹性法很有前景的发展方向     

弹性薄板弯曲问题的一个分区横向联结的泛函

给出了一个弹性薄板弯曲问题分区横向联结的泛函。并且证明，它的驻值条件等价于全部支配方程、外边界条件、分区公共边界处的内力平衡条件和位移连续条件。     

各向异性介质波动问题的时域边界元法及实验验证

对前已建立的各向异性介质波动问题的时域边界元计算模型进行了实验验证．用单向纤维增强光弹性复合材料模拟正交异性介质，用冲击加载，加载方向与纤维方向分别成0°，90°及45°角度，进行了正交异性动态光弹性实验及动态应变测量，并同时对该模型进行了时城边界元计算．将时域边界元方法计算出的应力分量代入正交异性动态光弹性的动态应力-光性定律，得到双折射条纹级数随时间的变化曲线，将其与动态光弹性实验的结果进行比较；此外，由动态电测获得的应变响应曲线推算出应力时程；与时城边界元计算出的应为响应曲线也进行了比较．两种情况下，时域边界元的计算成果均与实验成果吻合较好，从而证明该各向异性介质波动问题的时域边界元计算模型具有较好的精度及稳定性，可用于各向异性介质的动态问题特别是波传播问题的分析研究．     

双轴应力状态下正交异性动态光弹性应力——光性定律研究

基于静态下Ｈｙｅｒ和Ｌｉｕ表述的正交异性应力－光性定律，在前文中，提出了正交异性光弹性复合材料的动态应力－光性定律并证明了其在单轴应力状态下的正确性。本文旨在进一步考察在双轴应力状态下正交异性动态应力－光性定律的适用性，采用的方法是对纤维增强光弹性复合材料制作的平板模型施加冲击荷载，加载方向与材料纤维方向分别成０°、９０°及４５°角，同时进行正交异性动态光弹性实验和动态应变测量，另外，对该模型进行相应的各向异性介质时域边界元计算。把动态应变测量推算出的应力分量以及时域边界元计算出的应力分量分别代入正交异性动态应力－光性定律，得到随时间变化的双折射条纹级数历程，将其与正交异性动态光弹性实验的结果进行比较。实验及计算结果表明，在三个加载方向下，由这三种方法得到的双折射条纹级数历程均吻合良好，从而证明了前文提出的正交异性动态应力－光性定律在双轴应力状态下的正确性。     

A Revisit to the Frozen Stress Phenomena in Photoelasticity

Stress freezing in photoelasticity is regularly employed for 3-D elastic stress analysis of geometrically complex models, wherein the models under load are soaked at the stress freezing temperature and slowly cooled to room temperature. During this process, a model may distort or fail undergoing large deformations at the stress freezing temperature owing to thermo-mechanical interactions. The contribution of thermal deformation to the model distortions is neglected in the available literature of stress freezing. This aspect of stress freezing is investigated in this paper, wherein Digital Image Correlation (DIC) is used to study the strain behavior during a complete stress freezing cycle for an epoxy made of CY230 resin cured with HY951 hardener. The results show that the thermal contributions to the model distortions at the critical temperature must be taken care of to estimate the failure margins. The distortions and failures would mainly depend upon the thermal and mechanical response of the model material and the complexity of the model.     

Two photoviscoelasticity experiments using analog data reduction

A tensile analog may conveniently be used to determine the history of the principal-stress difference from the results of some photoviscoelastic model tests. With this method, the model test is run, and the fringe-order history at the point (s) of interest in the model is recorded. A tensile specimen of the same material is placed in a loading frame and loaded with a stress history which forces it through the fringe-order history observed in the model. This stress history is then the desired principal-stress-difference history occurring in the model. The stress history at points in two nonhomogeneous models was determined with a tensile analog. The uses and limitations of the analog technique are discussed.     

Prediction of wheel performance by analysis of normal and tengential stress distributions under the wheel-soil interface

Prediction of wheel performance by analysis of normal stress distribution under the wheel-soil interface was reported by one of our research members. In this study analysis of both normal and tangential stress distributions are included for the prediction of wheel performance. A visco-elastic soil model based on a three-element Maxwell model is used to evaluate normal stress distribution under a wheel running on soft ground. The values of the parameters characterizing the visco-elastic behavior of the soil can be derived from plate penetration tests. A rigid wheel-soil interface model is used to evaluate the tangential stress distribution under the wheel-soil interface. Shear deformation modulus, cohesion and angle of internal shearing resistance of the soil are derived from shear-displacement tests. Test results indicate that both maximum normal and shear stress occur in front of the wheel axle, and the location of peak normal stress shifts backwards towards the wheel axle while that of tangential stress shifts forwards when slippage is increased from a low value. Increasing slippage also causes a decrease in normal stress and an increase in tangential stress. Coefficients of traction and tractive efficiency are low at low slippage, increase with an increase in slippage, and level off at higher slippage.     

基于简化模型仿真的内螺纹应力分析方法研究

内螺纹通常是机械零件的疲劳强度薄弱部位。由于螺纹建模困难、计算量极大而且不容易收敛,工程上普遍采用无螺纹的简化模型进行仿真。简化模型的缺点是无法反映螺纹根部应力集中,所以应力结果是不正确的。针对此问题,提出一种基于简化模型仿真的内螺纹根部应力分析方法,该方法把内螺纹根部的应力分解成近源应力分量和远源应力分量,并根据它们的特点提出近源应力转换矩阵、远源应力转换矩阵的概念以及获取方法,利用这两个矩阵可以将简化模型仿真结果转换为近源应力和远源应力,然后叠加得到螺纹根部的最大应力。计算结果显示,内螺纹应力转换法基本上达到了三维细节模型的有限元计算精度,尤其是在疲劳强度薄弱部位即孔底端第一扣啮合螺纹根部,两种方法的结果吻合良好,证明了内螺纹应力转换法的精确性和有效性。     

An elastic–plastic constitutive law for the description of uniaxial and multiaxial ratchetting

An elastic–plastic constitutive model is proposed to describe 1-D and 2-D ratchetting. The model is also able to give correct results for 2-D ratchetting when only uniaxial identification is used, while no special threshold or parameter is used for the case of non-proportional loading. The original feature of this model consist in the introduction of a ratchetting stress (material characteristic) along with the maximal stress supported in the history of loading and the plastic strain at the last unloading. In this paper uniaxial and 3-D formulations have been described based on a numerical implementation in the software Code_Aster. Uniaxial and also multiaxial identifications have been used. Simulations have been realized for proportional and non-proportional homogeneous cases, as well as for structures under anisothermal thermomechanical loading. The results of a benchmark on a structure, comparing experiment, simulations by this model and some other phenomenological models, and a polycrystalline model are presented. An analysis of error margin due to the choice of Mises criterion is exposed.     

Stress functions for central straight cracked anisotropic plates

In this paper, the plane problem for an anisotropic plate with a central straight crack in any direction is solved. The stress functions are given to represent the finite stress concentrations near the crack tips by the weight integral method. It shows that there is no stress singularity at the crack tip. The model can be used to appropriate to fracture mechanics for non-metallic materials.     

Fully coupled hydromechanical model for compacted soils

To accomplish a proper conception and further modeling of the mechanical behavior of soils, observations at the micro and macro scales need to be merged. The authors believe that the link between scales can be achieved using the effective stress concept. In this paper, we present a model that quantifies the air and water volumes contained within the pores of a solid when their pore pressure is varied. The macroscopic consequences of this are expressed in terms of a single stress that is used to formulate a simple elastoplastic constitutive model to predict volume strains and shear strength of soils.     

A correlation for yield stress fluid flow through packed beds

Relatively few correlations are available for non-Newtonian fluid flows through packed beds, even though such fluids are frequently used in industry. In this paper, a correlation is presented for yield stress fluid flow through packed beds. The correlation is developed by introducing the yield stress model in place of the Newtonian model used in deriving Erguns equation. The resulting model has three parameters that are functions of the geometry and roughness of the particle surfaces. Two of the parameters can be deduced in the limit as the yield stress becomes negligible and the model reduces to Erguns equation for Newtonian fluids. The third model parameter is determined from experimental data. The correlation relates a defined friction factor to the dimensionless Reynolds and Hedstrom numbers and can be used to predict pressure drop for flow of a yield stress fluid through a packed bed of spherical particles. Conditions for flow or no-flow are also determined in the correlation. Comparison of model calculations, between a Newtonian and a yield stress fluid for flow penetration into a packed bed of spheres, shows the yield stress fluid initially performs similar to the Newtonian fluid, at large Reynolds numbers. At lower Reynolds numbers the yield stress effect becomes important and the flow rate significantly decreases when compared to the Newtonian fluid.     

Microstructural parameters affecting creep induced load shedding in Ti-6242 by a size dependent crystal plasticity FE model

This paper is aimed at identifying critical microstructural parameters that cause local stress concentration due to load shedding between microstructural regions of varying strengths. This stress is viewed as one of the fundamental reasons for crack initiation in Ti-6242. A rate dependent, anisotropic, elasto-crystal plasticity based finite element model (CPFEM) for poly-phase Ti-6242 is used in this study to identify the critical variables responsible for localized stress concentration due to load shedding. The model can account for various microstructural features like grain size, orientation and misorientation distributions. Various microstructural variables, such as crystal orientation, misorientation, grain size, Schmid factor and composition of phases, are considered in a detailed parametric study. Critical combinations of these parameters that result in high stress due to load shedding are identified. Finally, load shedding in a microstructure model of polycrystalline Ti-6242 is discussed from the results of CPFEM simulations. The model is statistically equivalent with respect to features observed in OIM scans.