分阶段结构损伤诊断方法

在现有分步损伤诊断方法的基础上,提出了一种3阶段结构损伤诊断方法.通过对遗传算法的特点进行分析,提出以逐步评定并排除无损单元的思路来处理结构损伤诊断问题.提出的分步方法中,应用了多次灵敏度遗传算法计算来进行无损单元排除,同时结合残余力向量指标对损伤大致位置进行判断.算例证明,在单纯用1种方法难以完成损伤诊断的情况下,用...     

采用扩展的贝叶斯人工神经网络选择结构损伤监测模型

对采用规则的动态数据进行结构损伤监测时,模式识别是一个有效的方法,人工神经网络作为匹配模式特征的系统方式广泛应用于模式识别研究中。人工神经网络设计是影响模型识别性能和效率的最基本因素。由Lam等人提出的贝叶斯人工神经网络设计法则为单隐层前馈人工神经网络确定大量隐性神经单元提供了严格的数学手段。本文的第一个目标是对贝叶斯人工神经网络设计法则进行拓展,包括选择隐层中神经单元的传递函数。所提出的法则具有高效的特点,适用于实时人工神经网络设计。目前,许多人工神经网络设计技术需要在训练前已知人工神经网络模型的类型,因此,最基本的问题是自动选择优化的人工神经网络模型类型的技术。由于模型参数和Ritz向量一般用于描述模式的特征,本文的第二个目标是采用模式识别对结构损伤监测中这两个模式特征进行比较。为了清楚判断这两个特征,研究中采用了IASC-ASCE准则。研究结果显示:采用模型参数进行训练的人工神经网络性能略优于采用Ritz向量进行训练的人工神经网络性能。     

Localized Structural Damage Detection: A Change Point Analysis

Many current damage detection techniques rely on the skill and experience of a trained inspector and also require a priori knowledge about the structure's properties. However, this study presents adaptation of several change point analysis techniques for their performance in civil engineering damage detection. Literature shows different statistical approaches which are developed for detection of changes in observations for different applications including structural damage detection. However, despite their importance in damage detection, control charts and statistical frameworks are not properly utilized in this area. On the other hand, most of the existing change point analysis techniques were originally developed for applications in the stock market or industrial engineering processes; utilizing them in structural damage detection needs adjustments and verification. Therefore, in this article several change point detection methods are evaluated and adjusted for a damage detection scheme. The effectiveness of features from a statistics based local damage detection algorithm called Influenced Coefficient Based Damage Detection Algorithm (IDDA) is expanded for a more complex structural system. The statistics used in this study include the univariate Cumulative Sum, Exponentially Weighted Moving Average (EWMA), Mean Square Error (MSE), and multivariate Mahalanobis distances, and Fisher Criterion. They are used to make control charts that detect and localize the damage by correlating locations of a sensor network with the damage features. A Modified MSE statistic, called ModMSE statistic, is introduced to remove the sensitivity of the MSE statistic to the variance of a data set. The effectiveness of each statistic is analyzed.     

梁结构损伤检测的一种方法

提出了一种用于梁结构损伤检测的方法。首先利用损伤前后结构柔度矩阵的改变来确定损伤发生的位置,再使用频率灵敏度法来计算损伤程度。以简支梁为例验证了该方法,结果表明:该方法是可行的,能够用于工程实践中。     

基于最小秩方法的结构损伤识别

针对结构损伤识别中的最小秩方法存在的问题,经过研究发现,对测试模态进行关于质量矩阵的正交归一化可保证反演后刚度矩阵的对称性;提出了一种迭代修正算法,可保持反演结果的稀疏性;基于模态力余量,定义了一种损伤指标来预先判定结构损伤单元的位置,并可据此选取合适的测试模态阶数进行反演计算。数值试验结果表明,改进后的方法在考虑测试模态误差的情况下可对结构的损伤进行精确的定位和标定。     

Multi‐objective Optimization of Sensor and Excitation Layouts for Frequency Response Function‐Based Structural Damage Identification

Abstract: The accuracy of many damage identification methods depends significantly on the quality of measurements collected by sensors, such as accelerometers, concerning the response characteristics of a structure. Often the number of sensors used to collect measurements is limited due to available funds, equipment, and access. In addition, the excitation location can significantly affect a sensor's ability to collect quality measurement information. Therefore, both the location and number of sensors and the location of the excitation must be optimized to maximize the quality of information collected. A multi‐objective optimization approach is presented that minimizes the number of sensors specified while maximizing the sensitivity of the frequency response functions (FRFs) collected at each specified sensor location with respect to all possible damaged structural elements. The multiple Pareto‐optimal sensor/excitation layouts obtained aid in determining the number of sensors required to obtain an effective level of measurement information. The benefit of using Pareto‐optimal sensor/excitation layouts is investigated by using the optimized layouts to collect measurement information for a FRF‐based structural damage identification method. Trial results confirm that an increase in damage identification accuracy and efficiency is achieved when Pareto‐optimal sensor/excitation layouts are used instead of nonoptimal layouts. In addition, the Pareto‐optimal layouts improved damage identification accuracy in noisy measurement environments due to increasing the quality of measurements collected.     

某建筑火灾后结构受损检测及加固

火灾对建筑结构的影响很大,依据某钢筋混凝土剪力墙结构建筑物受火后的受损检测,探究了火灾后建筑物检测内容、分析了受火构件的受力特性以及构件受损综合评定方法,随后对受损构件进行加固处理。为类似工程提供参考,以提高对防火重要性的认识。     

数字图像相关技术在结构损伤检测中的应用

结构损伤识别技术是结构健康监测的核心技术,也是健康监测的重难点之一。工程测量中受测量环境和技术的影响,很难获得完备的实验数据。论文利用一种新型测量技术,数字图像相关性(Digital Image Correlation,以下简称"DIC")技术,来提高模态应变能的识别精度。首先运用DIC技术和传统加速度传感器两种测量手段来获取固有频率和模态振型,实验结果表明,DIC技术获取的频率和振型与加速度传感器基本一致;随后对网壳结构进行基于DIC的模态应变能损伤识别分析,参数识别法获得模态振型及利用振型拟合得到的转角代入损伤识别指标进行损伤识别,结果表明,基于DIC的模态应变能损伤识别法能够有效进行损伤定位识别,且识别精度较好。     

建筑结构损伤识别的一种方法

将建筑结构简化为层剪切模型,基于模态频率对刚度的灵敏度,运用最小二乘原理,采用迭代寻优的方法,根据被测试的建筑物的固有频率,识别出系统的层间刚度.算例表明迭代速度快、精度高,通过和损伤前刚度的比较,能够准确地识别出结构损伤的位置和损伤量.     

基于应变指标的桥梁损伤识别方法

为了更好地实现桥梁的损伤识别、确保结构安全,探讨了应变指标(弯曲正应变和剪切应变)在桥梁损伤识别中的应用,给出了具体的应用方式,并通过多个数值模拟实例进行了验证;考虑到该指标的显著局部性,给出了测点优化布置的方法;考虑实际应用时的具体情况,分析了该方法识别结果的存在性、惟一性和稳定性问题,给出了有效的处理方法,使之有了很强的实用性。结果表明:该方法能直接发现损伤,对小损伤比较敏感,而且处理方式简单、结果可靠。     

基于应变差值类影响线的桁架损伤识别

面向基于静力测试数据的结构损伤识别问题,发展一种基于应变差值类影响线来识别损伤的方法,该方法利用事先规定应变测量单元的桥梁结构在单位移动荷载作用下的应变影响线,通过差值、求导计算,提出以应变差值、一阶导和二阶导影响线作为损伤指标的损伤诊断方法,一点测量多点激励,直接对所得影响线图进行观察即可准确定位损伤。本文分别以钢桁梁桥和星海跨海悬索大桥主桁架简化模型为例,分析了距离损伤杆远近不同的腹杆做测点识别损伤的效果,结果表明:该方法可以有效识别桁架腹杆损伤。     

径向基函数随机响应面法

针对随机响应面法对非正态分布响应与标准正态分布输入之间的复杂非线性隐函数拟合不够理想的问题,基于径向基函数在杂散数据拟合方面的优异性能,提出使用径向基函数替换Hermite多项式来解决复杂非线性隐函数拟合问题。以若干个非线性解析函数和钢管混凝土肋拱极限承载力不确定性问题作为算例,验证该方法对非正态分布响应拟合的精确性和对工程问题的适用性。算例结果表明,基于径向基函数随机响应面法对高度非线性的响应与输入隐函数拟合较好;在多参数钢管混凝土拱极限承载力不确定性问题中,精度较高,且比Hermite多项式样本点数量少。     

空间谱估计在Lamb波的结构损伤检测中的应用

考虑到阵列信号处理中空间谱估计能得到信号的波达方向的特点,本文将空间谱估计引入到基于Lamb波的结构损伤检测中。选择了最具代表性的多重信号分类算法进行分析,利用其对波达方向的估计对损伤进行定位。首先介绍了多重信号分类算法的特点及Lamb波的基本原理,然后通过在铝板上开小孔模拟损伤进行仿真。选取结构上布置的传感器阵列中的一个压电片激励Lamb波,随后收集阵列中所有传感器的波信号,对接收信号采用多重信号分类算法进行波达方向分析,可以从中比较准确地估计出结构损伤发生的角度,仿真结果表明空间谱估计中对波达方向的估计可以应用于结构的损伤检测中。     

基于卷积神经网络的结构损伤识别研究进展

为系统梳理基于卷积神经网络的工程结构损伤识别方法的发展脉络和研究现状,分别从结构损伤的识别目的和在不同类型结构中的应用两方面进行了归类、分析和评价。介绍了卷积神经网络的基本结构和评价指标,回顾了卷积神经网络的研究和应用历程。在损伤的识别目的方面,主要针对混凝土结构损伤的分类、定位和分割,详细介绍了基于不同类型卷积神经网络的结构损伤识别方法,即基于分类的方法、基于回归的方法和像素级的图像分割算法; 分析了各类方法所使用的卷积神经网络模型的结构特点、计算流程、训练方法和损伤识别性能。在不同类型结构的损伤识别方面,分析了卷积神经网络在砌体结构、钢结构桥梁和古建筑木结构裂缝识别中的应用。最后,基于对卷积神经网络优缺点的思考,提出了发展建议和展望。结果表明:训练样本中结构损伤的多样性对模型的损伤识别效果影响较大; 现有基于卷积神经网络的损伤分割方法模型参数较多,计算量大; 采用数据增广和迁移学习方法可有效防止模型过拟合,提高模型训练效率; 针对微小损伤和不同类型结构损伤的识别,此类方法的性能有待提高。     

基于频率平方变化比的桥梁结构损伤检测

频率平方变化比方法能对结构微小损伤进行定位识别。论文使用频率平方变化比对对钢桁架桥梁结构进行损伤检测,并通过对一个钢架桥梁的有限元分析证明了该方法的正确性。     

基于静力虚拟变形法的结构损伤识别研究

为了提高大型结构损伤识别的计算效率,引入静力虚拟变形法（VDM）,并结合序列二次规划（SQP）算法实现损伤定位和损伤定量。首先,基于VDM的基本原理,推导了损伤因子与虚拟变形的关系;其次,建立损伤应变与实际损伤应变的目标函数,并利用SQP算法优化目标函数,实现了结构损伤识别的快速计算;最后,以某实际大桥有限元模型为例,对其吊杆的损伤识别进行了数值模拟研究。设计了基于恒荷载的实时监测和基于车辆静荷载的定期检测2种工况对该方法进行验证。结果表明:该方法能够快速准确地识别出损伤发生的位置和程度。     

Damage Pattern Recognition for Structural Health Monitoring Using Fuzzy Similarity Prescription

Abstract:   Structural health monitoring (SHM) is a systematic method for non-destructive evaluation of a structure's performance by sensing, extracting, patterning, and recognizing features of the structural response. Most SHM approaches focus on statistical analysis for damage identification considering only random uncertainties. This article introduces a method that allows accommodating other types of uncertainties due to ambiguity, vagueness, and fuzziness which are statistically non-describable. The proposed method deals primarily with epistemic uncertainty. The method improves damage identification by performing damage pattern recognition using fuzzy sets. In this approach, healthy observations are used to construct a fuzzy set representing healthy performance characteristics. Additionally, the bounds on the similarities among the structural damage states are prescribed. Thus, an optimal group of fuzzy sets representing damage states such as little, moderate, and severe damage can be inferred as an inverse problem from healthy observations only. Piecewise linear functions are used as fuzzy membership functions representing the states of healthy and damaged. The optimal group of damage fuzzy sets is used to classify a set of observations at any unknown state of damage using the principles of fuzzy pattern recognition based on maximum approaching degree. A case study for damage pattern recognition of a model steel bridge is presented and discussed. The approach is capable of identifying damage patterns accurately.     

Finite Element Model Updating Using Evolutionary Strategy for Damage Detection

Abstract: Structural health monitoring through the use of finite element model updating techniques for dispersed civil infrastructures usually deals with minimizing a complex, nonlinear, nonconvex, high‐dimensional cost function with several local minima. Hence, stochastic optimization algorithms with promising performance in solving global optimization problems have received considerable attention for finite element model updating purposes in recent years. In this study, the performance of an evolutionary strategy in the finite element model updating approach was investigated for damage detection in a quarter‐scale two‐span reinforced concrete bridge system which was tested experimentally at the University of Nevada, Reno. The damage sequence in the structure was induced by a range of progressively increasing excitations in the transverse direction of the specimen. Intermediate nondestructive white noise excitations and response measurements were used for system identification and damage detection purposes. It is shown that, when evaluated together with the strain gauge measurements and visual inspection results, the applied finite element model updating algorithm of this article could accurately detect, localize, and quantify the damage in the tested bridge columns throughout the different phases of the experiment.     

既有刚架拱桥结构损伤检测及研究

刚架拱桥是在双曲拱桥、桁架拱桥基础上结合斜腿刚构的特点发展而来的复合结构桥型,兼有拱与斜腿刚架的力学特性和优势.通过对一运营十多年既有刚架拱桥的结构损伤调查研究,同时开展静载试验研究,测定桥梁应力、变形等结构整体行为,评估该桥承载能力等实际工作状态,为加固设计提供依据.     

Structural Damage Detection with Automatic Feature‐Extraction through Deep Learning

Structural damage detection is still a challenging problem owing to the difficulty of extracting damage‐sensitive and noise‐robust features from structure response. This article presents a novel damage detection approach to automatically extract features from low‐level sensor data through deep learning. A deep convolutional neural network is designed to learn features and identify damage locations, leading to an excellent localization accuracy on both noise‐free and noisy data set, in contrast to another detector using wavelet packet component energy as the input feature. Visualization of the features learned by hidden layers in the network is implemented to get a physical insight into how the network works. It is found the learned features evolve with the depth from rough filters to the concept of vibration mode, implying the good performance results from its ability to learn essential characteristics behind the data.