基于替代模型的岸桥随机有限元模型修正

研究了考虑参数随机不确定性的岸桥有限元模型修正问题。首先,假设岸桥的待修正参数和模态参数都服从正态分布,将不确定性模型修正问题转化为均值和标准差的修正问题;其次,以某岸桥为研究对象,进行风振响应实测,利用随机子空间法得到岸桥前4阶实测模态参数;最后建立岸桥的有限元模型,基于Kriging替代模型及多目标遗传算法对岸桥结构进行有限元模型修正。结果表明,考虑参数不确定性的随机有限元模型修正方法能有效修正岸桥结构参数的均值和标准差。     

面向健康诊断的混凝土斜拉桥模型设计与分析

以广东惠州合生大桥为原型,基于相似理论面向健康诊断设计制作了1∶15的大比例混凝土斜拉桥模型,分析了模型设计中的问题及解决办法。建立了原型桥和模型桥的初始有限元模型(FEM),以实测数据对其进行修正,得到了符合实际的基准有限元模型,采用基准有限元模型对模型桥和原型桥的静动力特性及损伤状态下的损伤特性进行了计算和相似性分析。结果表明,模型设计中进行了各种简化和处理后,模型桥与原型桥的静动力性能及损伤状态下的损伤特性仍具有良好的相似性,采用该模型桥进行损伤识别试验研究能够较好地反映实际,为混凝土斜拉桥模型设计提供参考。     

佛山东平大桥静动力分层次有限元模型修正研究

提出一种适用于大型桥梁结构联合静动力的分层次有限元模型修正方法，并在佛山东平大桥成桥静载试验和模态试验据的基础上，借助于ANSYS软件中的优化设计平台对该桥进行了静动力分层次有限元模型修正，结果表明该方式物理概念清晰，能充分利用静动力试验信息，减小优化规模．修正后的有限元模型的静力特性和动力特性与试验结果吻合良好，建立了该桥结构健康监测与安全评估的基准有限元模型。     

基于人工神经网络的藏式结构有限元模型修正

针对典型藏式传统建筑年代久远、建造过程无设计图纸、存在太多不确定性,通过有限元数值分析所得结果与实验结果不能较好吻合问题,建立所测结构的有限元模型,并通过人工神经网络方法,以实测结构模态参数为目标对典型藏式结构的有限元模型中部分不确定因素—梁及雀替的等效变截面梁高、材料密度及弹性模量进行修正,得到更接近真实状态的有限元模型。该模型对该典型藏式结构的损伤识别、可靠度评估具有重要意义。     

基于环境振动测量值的悬索桥结构动力模型修正

以主跨1385米的江阴长江公路大桥为背景，研究利用环境振动测量值进行悬索桥结构有限元模型修正的可行性。采用的模型修正方法以结构特值灵敏度分析为基础，通过迭代求解二次规划问题修正有限元模型参数。根据江阴长江公路大桥的设计图纸建立悬索桥三维有限元模型固有振动分析。同时利用大桥现场环境振动测量值得到一组结构实测模态参数，用以作为有限元模型修正的基准，根据结构特征值灵敏度分析结构选择了一组待修正结构参数并予以修正，修正后有限元模型的动力特征更加趋近于环境振动实测值，最后指出了模型修正不能完全消失大桥结构理论－实测频率差的可能原因。     

考虑扭转效应的钢管混凝土纤维梁模型应用研究

在已有研究的基础上,采用考虑扭转效应的钢管混凝土纤维梁模型,建立了曲线组合梁桥的杆系有限元分析模型,并同时建立了“壳-实体”精细有限元模型,多尺度有限元模型和传统纤维梁模型进行对比。对不同输入方向的四条地震波作用下的有限元模型的计算结果进行了全面的对比,包括自振特性、墩顶位移、墩柱扭转角,墩柱内力等。结果表明:采用考虑扭转效应的纤维梁模型分析曲线组合梁桥在地震作用下的受力反应,能够获得较高的精度以及较快的建模速度和求解效率,可对曲线钢-混凝土组合梁桥在纵向地震作用下的扭转效应做出准确的评估。     

基于Kriging模型和分层模型修正技术的结构边界条件识别

为得到能够准确反映结构装配关系的有限元模型边界条件，结合Kriging模型和分层模型修正技术，提出了一种结构边界条件识别方法。为削弱材料参数误差对边界条件识别的影响和改善修正不适定，利用实测自由模态频率，修正有限元模型的材料参数；以修正后的模型为基础进行边界条件识别，通过灵敏度分析确定模型的待修正边界参数，采用拉丁超立方抽样(Latin hypercube sampling, LHS)进行试验设计，并以实测约束模态频率与Kriging模型预测频率的差值最小为目标函数，利用粒子群算法求解最优参数。某包装机械摇臂连杆结构的试验结果表明，与传统的边界条件识别方法相比，所提方法具有较好的识别效果，识别得到的边界参数、结构响应等与实际结构具有较好的一致性。     

C/SiC复合材料结构件热模态模型建模和修正方法

C/SiC材料是一种热的不良导体，在防热隔热领域具有重要的潜在应用前景。本文开展了C/SiC结构件的有限元模型修正方法研究。在热弹性理论的基础上推导了各向异性材料的应力-应变关系，建立了C/SiC复合材料层板在高热流密度条件下的系统泛函。建立了考虑温度效应的C/SiC复合材料结构的动力学有限元模型。提出了基于多层级思想的C/SiC复合材料结构模型修正方法。以模态频率差值最小为修正目标，复合材料力学、热学参数为修正变量，开展模型修正。模型修正结果表明，本文提出的方法具有良好的修正效果，能够准确修正结构热环境下的复合材料参数及边界条件。     

大跨径钢桥桥道系协同作用试验模型研究

建立了考虑整桥变形的局部梁段模型,对钢桥面铺装在整桥及车载作用下的应力进行了计算,得到整桥变形的影响系数为1.17。建立初拟的两种协同作用试验方案的有限元模型,以局部梁段模型中铺装的最大应力应变值为参考值,对两种协同作用试验方案中相关结构参数进行了修正;在此基础上,为保证设计的协同作用试验模型与原型的一致性,将修正后的试验模型与局部梁段模型在控制点位应力值进行了对比;最终得到大跨径钢桥桥道系协同作用试验模型。研究成果可为钢桥面铺装室内试验模型的设计提供理论依据。     

基于极限状态法的岸桥结构抗震可靠性分析

针对集装箱码头岸边装卸桥(简称岸桥)在地震作用下的安全可靠性问题,对岸桥在极限状态下的抗震可靠性开展了研究。通过对地震参数与材料力学性能的随机性描述,获得了地震作用下各随机变量的概率分布规律和统计参数;采用有限元软件ANSYS建立了岸桥结构的力学模型,用时间历程法对地震作用下岸桥承载能力进行了计算;根据可靠性理论建立了地震作用下岸桥的可靠度计算模型,用 Monte-Carlo法计算了岸桥在极限状态下的抗震可靠度。结果表明在8级烈度地震作用下,岸桥结构整体上比较安全,但门框横梁与陆侧立柱的连接处的可靠度指标比较低,存在失效可能性;与许用应力法相比,极限状态法更具合理性。     

低压母线系统振动与声辐射特性分析

采用有限元与边界元法,以某型号母线桥为例,对低压大电流母线设备正常运行中噪声偏大的问题进行研究。建立母线桥有限元结构分析及边界元声学分析的三维模型。在求解系统固有频率及振型的基础上,应用有限元法完成了母线桥结构振动响应的数值计算。然后,将母线桥外表面的节点振动位移作为声学边界条件,采用边界元法计算了母线桥的辐射噪声特性,得到母线桥辐射声场的分布规律,确定系统中主要振源和声源的部位。根据分析结果提出了结构改进方法,计算和实验结果表明该方法控制噪声的可行性和有效性。     

地铁齿轮箱振动与噪声计算分析

运用结构有限元法和声学有限元法，对地铁某型号齿轮箱振动与噪声问题进行计算分析。建立齿轮箱三维有限元模型，采用结构有限元法计算齿轮箱模态频率和振动响应。建立齿轮箱外声场三维有限元模型，导入齿轮箱振动响应结果作为声学边界条件，并采用无限有限元法计算了辐射声压级。     

Level set topology optimization of structural problems with interface cohesion

This paper presents a finite element topology optimization framework for the design of two‐phase structural systems considering contact and cohesion phenomena along the interface. The geometry of the material interface is described by an explicit level set method, and the structural response is predicted by the extended finite element method. In this work, the interface condition is described by a bilinear cohesive zone model on the basis of the traction‐separation constitutive relation. The non‐penetration condition in the presence of compressive interface forces is enforced by a stabilized Lagrange multiplier method. The mechanical model assumes a linear elastic isotropic material, infinitesimal strain theory, and a quasi‐static response. The optimization problem is solved by a nonlinear programming method, and the design sensitivities are computed by the adjoint method. The performance of the presented method is evaluated by 2D and 3D numerical examples. The results obtained from topology optimization reveal distinct design characteristics for the various interface phenomena considered. In addition, 3D examples demonstrate optimal geometries that cannot be fully captured by reduced dimensionality. The optimization framework presented is limited to two‐phase structural systems where the material interface is coincident in the undeformed configuration, and to structural responses that remain valid considering small strain kinematics. Copyright © 2017 John Wiley & Sons, Ltd.     

基于响应面法的结构动力学模型修正

为了获得精确的结构动力学模型,提出了响应面和优化相结合的方法。利用参数化模型和优化拉丁方试验设计获取样本点构造多项式响应面模型,最小二乘法确定多项式系数并检验响应面的拟合精度。用响应面计算结果与实验结果的误差构造目标函数,自适应模拟退火算法来优化修正响应面参数,将修正后的参数值带入有限元模型得到修正模型。以欧洲航空科技组织的基准模型GARTEUR飞机模型为算例,对比修正前后模态频率,结果表明修正后的模型在测试频段和预测频段具有良好的复现和预测能力,进而验证了基于响应面法与优化方法相结合的结构动力学有限元模型修正的有效性。     

Finite Element Analysis of Mechanical Properties of 3D Four-directional Rectangular Braided Composites—Part 2: Validation of the 3D Finite Element Model

In the first part of the work, we have established a new parameterized three-dimensional (3D) finite element model (FEM) which precisely simulated the spatial configuration of the braiding yarns and considered the cross-section deformation as well as the surface contact relationship between the yarns. This paper presents a prediction of the effective elastic properties and the meso-scale mechanical response of 3D braided composites to verify the validation of the FEM. The effects of the braiding parameters on the mechanical properties are investigated in detail. By analyzing the deformation and stress nephogram of the model, a reasonable overall stress field is provided and the results well support the strength prediction. The results indicate it is convenient to predict all the elastic constants of 3D braided composites with different parameters simultaneously using the FEM. Moreover, the FEM can successfully predict the meso-scale mechanical response of 3D braided composites containing periodical structures.     

复阻尼模型结构地震时程响应研究

阻尼是结构动力分析的重要参数,对动力响应计算的结果有明显影响。目前,结构动力响应分析中的阻尼模型大都采用的是传统的Rayleigh阻尼模型,这种模型是为计算解耦而构造出的,其物理意义不是很明确。该文根据复阻尼系统理论,利用Newmark-β积分法,编制了Rayleigh阻尼和复阻尼模型的三维有限元程序。以三维钢框架结构为研究对象,计算了两种阻尼模型的结构地震时程响应和复阻尼模型的损耗因子,并讨论了不同加速度峰值和时间积分步长对复阻尼结构响应和损耗因子的影响。研究表明:在加速度峰值为2m/s2的El-Centro波作用下,两种阻尼模型的响应相差50%―100%,后者的结构动力响应远大于前者,损耗因子随应力或位移的增大而增大;复阻尼模型结构动力响应和损耗因子的稳定性和精度,与时间积分步长密切相关,不合适的时间积分步长将导致结果发散。     

Finite-element analysis of the frequency response of a metallic cantilever coupled with piezoelectric transducer

This paper is devoted to modeling the dynamic response of an electromechanical system consisting of a piezoelectric transducer glued on part of the upper surface of a metallic cantilever. The piezo works both as vibration sensor and as actuator, and the system is the basis of many vibration-control devices of current interest. A three-dimensional (3D) finite-element method (FEM) model that reproduces the physical system is proposed, and its advantages with respect to an analytical approach and to one-dimensional (1D) and two-dimensional (2D) FEM models are discussed. In sensor mode, the frequency response in terms of the voltage at the electrodes is drawn; while in actuator mode, the frequency response of acceleration and displacement at the free end of the cantilever is calculated. The 3D model has been verified through the comparison with the results from the experiment carried out at the University of L'Aquila, Italy. Furthermore, experimental inaccessible quantities such as stresses at the piezo-cantilever interface are computed in both modes as a preliminary step in the study of delamination phenomena and their impact on the performance of the system in vibration-control applications.     

斜纹机织碳纤维/环氧树脂复合材料性能及其在电动汽车轻量化设计中的应用

采用均质法和单胞有限元法, 研究了斜纹机织碳纤维/环氧树脂复合材料的三维本构模型, 并应用于轻量化电动汽车结构的设计计算中。通过光学显微镜观测获得T300碳纤维/环氧树脂复合材料层合板细观结构几何参数, 依据参数建立了代表体积元的几何模型; 结合有限元分析方法与层合板理论对代表体积元模型进行受力分析, 获得了复合材料三维本构模型参数, 并运用单向拉伸与三点弯曲物理试验对所建立的本构模型进行了验证; 通过二次开发, 把该三维本构模型用于轻量化电动汽车结构件的设计中, 并根据汽车安全测试标准对所设计的结构进行了强度校核。研究表明: 运用所建立的三维材料本构模型进行的仿真实验结果与真实拉伸和弯曲试验结果相吻合, 误差分别为4.04%和7.79%, 可用于轻量化结构的设计; 研究开发的电动汽车, 轻量化效果明显, 在满足车身强度要求下整车质量减轻了12%。      

Investigation of the ultimate residual strengthen of a worn casing by using the arc-length algorithm

The accurate prediction of the ultimate residual strength (URS) (including collapse, burst, and tensile strength) of worn casing is important in the design and safety assessment of casings for the oil and gas industry. Based on the true stress–strain constitutive relationship of P110 steel, this paper establishes a three-dimensional (3D) finite element model (FEM) to obtain the URS of worn casing by using the arc-length algorithm. The effects of worn defect length, wear percentage, drillstring diameter, and casing thickness on the URS are investigated. Moreover, three full-scale experiments on casings with different wear percentages are conducted to verify the accuracy and reliability of the 3D FEM and of the algorithm. The validation proves that the URS of worn casing can be predicted by the 3D FEM to a satisfactory degree of accuracy. The URS has an exponential relation with defect length, but a linear relation with wear percentage, drillstring diameter, and casing thickness. The effects of numerous factors on the URS are also studied. Regression analysis is conducted on the basis of 3D FEM predictions, and the corresponding design equations for worn casing are presented.     

A Numerical-analytic Method for Quickly Predicting Springback of Numerical Control Bending of Thin-walled Tube

Springback is one of important factors influencing the forming quality of numerical control （NC） bending of thin-walled tube. In this paper, a numerical-analytic method for springback angle prediction of the process was put forward. The method is based on springback angle model derived using analytic method and simulation results from three-dimensional （3D） rigid-plastic finite element method （FEM）. The method is validated through comparison with experimental results. The features of the method are as follows： （1） The method is high in efficiency because it combines advantages of rigid-plastic FEM and analytic method. （2） The method is satisfactory in accuracy, since the field variables used in the model is resulting from 3D rigid-plastic FEM solution, and the effects both of axial force and strain neutral axis shift have been included. （3） Research on multi-factor effects can be carried out using the method due to its advantage inheriting from rigid-plastic FEM. The method described here is also of general significance to other bending processes.