基于跨模型模态应变能变化的结构损伤识别两阶段法

为提高结构损伤识别精度,采用跨模型模态应变能变化(Cross-model Modal Strain Energy Changes,CMSEC)代替传统的模态应变能变化(Modal Strain Energy Changes,MSEC)建立先定位后定量的结构损伤识别两阶段法.与传统模态应变能(Modal Strain Energy,MSE)计算时仅利用结构损伤后的振型不同,跨模型模态应变能(CMSE)的计算同时利用结构损伤前后的振型.先利用CMSEC构建CMSECI(Cross-model Modal Strain Energy Index)损伤指标来确定损伤位置,筛选出疑似损伤构件后,再利用CMSEC建立定量识别损伤程度的灵敏度方程.采用基于奇异值截断的迭代法求解该方程,以避免迭代过程中可能出现的病态方程问题.根据公式构成可以定性判断CMSEC比MSEC更难被噪声干扰.某简支梁算例表明,本文方法在损伤定位和损伤定量两个阶段均比传统方法具有更高的识别精度.为了深入考察本文方法的优越性,采用蒙特卡洛法从宏观的相对误差上界和具体的相对误差分布这两个视角解释了本文方法优于传统方法的原因.     

基于不完备模态信息的结构损伤识别方法

针对结构损伤识别中的有限测点问题和测试噪声问题,提出一种基于模型修正法的损伤识别方法,仅利用结构的低阶频率和相应的不完备振型进行损伤识别。基于动力缩聚法构造参数化的振型扩展矩阵,解决振型不完备的问题,然后根据交叉模型交叉模态法CMCM(cross-model cross-mode)构造约束方程,并使用Hestenes-Powell增广拉格朗日乘子法求解约束优化问题,从而根据优化问题的最优解判断出损伤位置和损伤程度。在模态数据包含测试噪声的情况下,提出一种改进的CMCM方法,以减小测试噪声对损伤识别结果的影响。对一个25杆平面桁架进行数值仿真实验,结果表明,在3%的噪声水平下,仅需测得损伤结构的前5阶不完备模态,本文方法就能较准确地识别结构损伤。     

基于模态参数灵敏度的损伤方程组求解正则化方法研究

基于模态参数的结构损伤识别方法是振动损伤识别领域中应用最为广泛的方法.利用模态参数灵敏度构建结构损伤方程组,对其进行求解可以识别结构损伤位置和程度.由于实际工程中模态参数不完备性和噪声的影响,结构损伤方程易出现病态问题,直接求解可能产生错误的结果.为了解决这一问题,可以引入正则化方法进行求解.然而,各类正则化方法的基本...     

基于矩阵摄动理论识别结构损伤的筛选法

定义了结构有限元模型中的单元损伤识别参数,建立了基于矩阵摄动理论的结构损伤识别方程,结合结构损伤的特点,提出了求解损伤识别方程的筛选法,该方法可以避免优化方法和迭代方法在未损伤单元上出现的"伪损伤"现象.通过一架五跨平面桁架损伤识别的数值仿真结果,证明了该方法的有效性与优越性.     

转子系统模态参数现场识别技术研究

基于时域振动参数识别理论,本文讨论了由转子和油膜轴承构成的转子系统的模态参数现场识别问题。本文所述方法不需要施加人工激振,仅在轴承座处测量轴承座的振动和转子的绝对振动,可以在线应用。     

识别时变结构模态参数的改进子空间方法

给出了基于整体数据子空间方法的改进算法,原方法对噪声的影响极为敏感,基于原方法的关键是计算一个与原系统的广义能观阵的特征值相同的矩阵的基本事实,将求矩阵的过程转化为最小二乘的问题,然后用广义能观阵替换输出矩阵,对数据进行滤波,从而有效地降低了算法对噪声敏感性,但同时增加了计算量。利用奇异向量矩阵的正交性,将最小二乘问题中求矩阵伪逆的过程简化为求该矩阵的转置,从而降低了计算量,改进后的算法能够在提高算法对噪声的敏感性的同时降低计算量。最后利用改进后的方法辨识出一个两自由度弹簧质量块模型在三种不同的刚度变化形式下的伪模态参数,并通过与使用原方法所得的结果比较,验证了以上结论。     

利用地脉动试验识别楼房结构的模态参数

论述了在地脉动对结构物的环境激励条件下,进行结构物地脉动响应测量的试验原理及方法,提出了固有频率识别与振型识别的互补较正法．通过对实际楼房结构的地脉动响应测量识别结构的模态参数．识别结果表明该理论及试验方法行之有效．     

基于应变模态和稀疏贝叶斯学习的网架结构损伤识别

传统稀疏贝叶斯学习算法进行损伤识别时需要对每个单元进行刚度损伤系数的迭代更新,当结构单元众多时,存在计算效率低和对振型的完备性要求高等问题.本文提出了损伤识别两步法,首先利用应变模态差指标进行疑似损伤单元的判断;接着以单元刚度损伤系数为目标参数,建立结构损伤识别的多层次稀疏贝叶斯学习模型,利用稀疏贝叶斯学习算法进一步识...     

基于模态数据的长细结构小损伤精确识别研究

本文针对现有的损伤识别方法不能满足部分结构损伤识别精度要求的现状,对结构的小损伤精确识别方法开展研究．以长细结构为研究对象,对具有不同损伤位置和损伤程度的圆柱形的轻阻尼梁结构进行了数值分析和实验研究,应用数值计算方法和实验确定的特征向量和特征频率对长细结构裂缝参数进行识别计算．本文在研究过程中编制了一个创新性的预测程序,通过其一次性生成目标函数图来选择合适的初始参数,从而对识别结果进行分析．研究结果表明,应用本文提出的识别方法,裂缝位置的识别误差可以控制在0.05 %~0.28 %范围内,裂缝深度识别误差低于7 %．     

结构模态参数的时序分析(ARMAV模型)识别法

本文简述了用时间序列分析(ARMAV 模型)识别结构模态参数的原理,对建模方法进行了分析,并对建模分析中至关重要的初值选择问题进行了讨论,提出了一种有效的初值的选取方法;在模拟计算部分中,用 Wilson-θ积分法模拟了一个三个自由度的振动和低信噪比、大阻尼振动和密集模态等复杂情形,对用时序方法(ARMAV 模型)识别模态参数的能力进行了检验;最后通过对实验数据进行有效的预处理后,用时序方法(ARMAV 模型)识别出一个真实结构(圆盘)的第一、二阶模态参数,并得到结构的完整振型.     

A note on a boundary condition for spouted beds

The boundary condition, zero solids pressure at the top of a particle bed of maximum spoutable height, Hm, is shown to eliminate any resort to empiricism in the derivation of the fluid velocity in the annulus of a spouted bed for which both viscous and inertial effects are taken into account. The same boundary condition fails when applied to a spouted bed for which the bed height H 〈 Hm, especially when H 〈 0.8Hm.     

Inflow boundary condition for DNS of turbulent boundary layers on supersonic blunt cones

For direct numerical simulation (DNS) of turbulent boundary layers, gen- eration of an appropriate inflow condition needs to be considered. This paper proposes a method, with which the inflow condition for spatial-mode DNS of turbulent boundary layers on supersonic blunt cones with different Mach numbers, Reynolds numbers and wall temperature conditions can be generated. This is based only on a given instant flow field obtained by a temporal-mode DNS of a turbulent boundary layer on a flat plate. Effectiveness of the method is shown in three typical examples by comparing the results with those obtained by other methods.     

Damage detection based on the propagation of longitudinal guided wave in a bimetal composite pipe

This paper investigates the damage detection based on the propagation of guided wave in bimetal composite pipes, which can identify damage locations in both axial and circumferential directions. The feasibility of the method is showed by numerical simulations using FEM code ANSYS. Mode analysis is used to evaluate the guided wave mode and its structure, which can provide the basis of the mode selection in measurements scheme. The guided wave propagation in a damaged pipe is computed by transient analysis. 16 nodes around the pipe wall, as probes, are used to record the guided wave signal. When Pseudo Margenau—Hill distribution (PMHD) for each signal is carried out, three types of modes could be found, which are led mode, excited mode and lag mode in sequences. Based on the results, the arrival time of the excited mode could be used to locate damage in axial direction, and the energy distribution around the pipe of lag mode is consistent with the damage in circumferential direction. The simulation illustrated the possibility of detecting damage location in both axial and circumferential directions based on longitudinal ultrasonic guided waves only.     

Interval parameter perturbation method for evaluating the bounds on natural frequencies of structures with interval parameters

For structural parameters with uncertainties, interval mathematics can, in the case where the probabilistic distribution density of uncertain variables is unavailable, deal with the influence of uncertainties in structural parameters on the response of structures. In order to evaluate the region containing natural frequencies of structures with interval parameters, the interval parameter perturbation method is presented in this paper. The advantage of the present method is its computational efficiency in evaluating the region containing natural frequencies. A numerical example is used to illustrate the efficiency of the method proposed. The project is supported by National Youth Natural Science Foundation of China and National Outstanding Youth Science Foundation of China.     

基于迭代估计的三轴加速度计温度模型参数外场标定方法

为补偿加速度计标度因数和零偏的热漂移误差,设计了一种可估计内场恒温环境下参数变化量的加速度计温度参数模型,并提出了基于迭代估计的温度模型参数外场标定方法。该方法不依赖精密的惯性测量设备和温箱,仅在重力场内对加速度计进行连续多组位置观测,充分利用加速度计冷启动过程产生的热漂移误差进行模型参数辨识,解决了加速度计温度误差特性和启动温度点相关的问题。针对加速度计测量信号为高斯白噪声的特点,建立了关于温度模型参数矢量和多组位置倾斜矢量的非线性准则函数,提出了两步迭代估计方法实现两组参数矢量的分离估计。根据不同位置下三轴加速度计输出信号粗略提取倾斜矢量,解决了迭代估计算法的初值问题。重力场内通过优化分析加速度计温度模型参数对重力值的灵敏度,设计了6组位置观测编排。实验结果表明,温度误差补偿前的重力值测量最大误差为3.62×10-4g,而温度误差补偿后重力值测量误差小于1×10-5g;同时,温度误差补偿前系统3 h纯惯性导航最大定位误差为1186 m,而温度误差补偿后最大定位误差小于600 m,从而表明提出的外场标定方法的有效性。     

有界不确定参数结构位移范围的区间摄动法

当结构参数具有误差或有界不确定性时,区间数学可以在不同知识不确定变量的概率分布的情况下定量地考察不确定参数对结构响应的影响。为计算出不确定结构参数对结构位移影响范围的上下界,文中提出了的两种区骚动国方法。     

基于重力场特征参数信息熵的适配区选择方法

针对当前重力统计特征参数类别繁多,选择标准复杂而导致错选有效匹配区域的问题,利用信息熵具有能够整合多种统计参数且算法计算量小的特点,提出了一种基于特征参数信息熵的重力辅助导航适配区的选择方法。首先,在DTU10模型下将该方法与传统单一特征参数的方法进行比较,确定了传统方法的确会错误选择可匹配区,从而也反映了所提出方法的优越性;其次,在该方法划分出的匹配区和非匹配区中分别设计了8条仿真航线,匹配区中仿真航线的匹配效果明显优于非匹配区中的匹配效果。仿真结果表明了该方法的有效性。     

Plate/shell structure topology optimization of orthotropic material for buckling problem based on independent continuous topological variables

The purpose of the present work is to study the buckling problem with plate/shell topology optimiza-tion of orthotropic material.A model of buckling topology optimization is established based on the independent,con-tinuous, and mapping method, which considers structural mass as objective and buckling critical loads as constraints. Firstly, composite exponential function (CEF) and power function(PF)as filter functions are introduced to recognize the element mass,the element stiffness matrix,and the ele-ment geometric stiffness matrix.The filter functions of the orthotropic material stiffness are deduced. Then these fil-ter functions are put into buckling topology optimization of a differential equation to analyze the design sensitiv-ity.Furthermore,the buckling constraints are approximately expressed as explicit functions with respect to the design vari-ables based on the first-order Taylor expansion.The objective function is standardized based on the second-order Taylor expansion. Therefore,the optimization model is translated into a quadratic program.Finally,the dual sequence quadratic programming(DSQP)algorithm and the global convergence method of moving asymptotes algorithm with two different filter functions(CEF and PF)are applied to solve the opti-mal model.Three numerical results show that DSQP&CEF has the best performance in the view of structural mass and discretion.