移动荷载作用下梁裂缝识别的小波方法研究

利用小波分析对简支梁的裂缝进行识别。通过对带裂缝简支梁在移动荷载作用下的跨中响应用Mexicanhat小波进行连续小波变换，从小波系数的模极大值点有效地得到荷载经过裂缝的时间，从而识别裂缝位置，从模极大值处的Lipschitz指数判断裂缝深度，Lipschitz指数随裂缝深度的增加而减小。同时讨论了荷载移动速度和裂缝位置对Lipschitz指数的影响。通过分析和仿真计算获得满意结果，在梁结构损伤诊断中具有较高的应用价值。     

基于小波分析的Timoshenko梁裂缝识别研究

采用等效转动弹簧代替梁内的不扩展横向裂缝,研究Timoshenko裂缝梁的横向振动特性,建立了一种与有限元分析相结合的、基于模态参数的小波分析识别Timoshenko梁内裂缝的方法。以一简支梁为例,通过建立含横向不扩展裂缝的Timoshenko梁的有限元模型,用Lanczos法对结构的模态进行了计算分析,求出了基本振型和转角模态。分别应用mexh小波和db小波为母小波对二者做小波变换,进行多尺度分析,通过小波系数模极大值位置识别出梁内的裂缝。并对识别结果进行对比,发现识别Timoshenko梁裂缝时,基于转角模态小波变换的方法对小波基、尺度的要求较低,变换后的小波系数线更为平滑,奇异性特征更为明显,故运用转角模态小波变换来识别Timoshenko梁裂缝,较之运用基本振型小波变换的方法更为方便、有效。该方法对Timoshenko梁裂缝识别的工程应用具有参考价值。     

基于加速度响应连续小波变换的线性时变结构瞬时频率识别

推导了函数积分运算的连续小波变换计算方法。借助此法,假设线性时变结构的质量系数为时不变常数,或者其随时间变化的规律已被经验掌握,仅利用线性时变结构自由振动的加速度响应信号,就可以计算出速度响应和位移响应信号的连续小波变换值。将一小段时刻点的线性代数方程组构造成最小二乘问题,求解最小二乘解识别出结构的时变阻尼和时变刚度,从而确定时变结构的瞬时频率。通过对5层剪切梁楼房模型和3自由度密集模态时变结构瞬时频率的识别,验证了识别方法的正确性、有效性和抗噪声能力.     

小波变换在结构损伤识别特征提取中的应用

小波变换用来进行特征提取是近年来模式识别领域中较新的一个研究方向。不同于传统的特征提耿方法，它借助信号处理领域常用的小波变换这一数学工具，对特征库进行再次加工，经过去粗取精而得到更利于分类的新的信息。针对结构损伤识别特征提取的特点，发展了小波变换对基本特征库进行分频段处理的双谱相辅分析技术，并开发了相应的软件系统，通过算例表明，该方法可以达到降低噪声，凸显结构损伤信息的目的。     

基于小波散射卷积神经网络的结构损伤识别

损伤识别是结构状态评估领域的关键问题之一，对确保结构安全性有重要意义。深度学习算法在基于振动的结构损伤识别方面带来了许多突破，但从海量数据中挖掘结构损伤关键信息仍是亟待解决的技术难题。该研究提出了基于一维卷积神经网络(one-dimensional-convolutional neural network, 1D-CNN)深度学习的结构多类型损伤识别模型，采用小波散射变换对1D-CNN架构第一层卷积滤波器进行替换，通过散射系数实现输入层原始数据降维与特征提取，结合CNN卷积层、激活层和池化层实现监测数据特征增强处理。在此基础上，结合1D-CNN全连接层与Softmax函数实现特征数据分类，从而实现结构多类型损伤定位与定量高效识别。通过钢桁架结构和斜拉桥两种数值模型对上述框架进行了验证。结果表明：与普通卷积神经网络模型相比，基于小波散射卷积神经网络的结构损伤识别精度显著提升，损伤分类准确率达95.0%以上。随着传感数据环境噪声比例的增加，小波散射卷积神经网络损伤分类准确率虽略有下降，但仍保持较高精准度，说明该方法具有较强的鲁棒性抗噪能力。     

结构损伤识别的小波包分析试验研究

在结构服役期内，由于局部损伤的累积发展将导致结构刚度等特性的降低，可能导致灾难性的事故。传统的基于模态参数的损伤识别方法有时效果并不理想。提出基于小波包变换的能量变化率指标，并用其进行损伤识别的方法。首先将所得的结构响应信号进行小波包分解，然后通过小波包能量变化率指标进行损伤定他。通过三种不同损伤工况的梁体室内试验，证明所提出的损伤指标可以准确地识别损伤位置。     

基于WTMM的爆破振动信号奇异性分析

为更细致的研究爆破振动信号中携带的能反映场地特征和爆破特征的重要信息，首先进行了完整场地、减震沟和预裂缝条件下的爆破振动试验，在获得的试验数据的基础上，运用小波变换模极大值（WTMM）方法对试验数据进行分析处理，得到了如下结论。小波变换模极大值线能够延伸至最小细节尺度处，模极大值的变化反映了此尺度时信号能量的变化趋势；由小波变换模极大值曲线图可以看出，该爆破试验场地上的爆破振动信号在lba（a为小波变换尺度）为4．5—6．0的范围内具有稳定的李氏指数，并且该指数能够很好地刻画爆破地震波携带的场地特征等奇异性信息；Lips—chitz指数曲线图较为直观地反映了爆破场地的差异，同时得到了三种条件下的李氏指数分别为-3．69～-1．88、-3．54～-1．71、-4．14～-1．58，验证了减震沟比预裂缝有更好的减震效果。     

利用加速度响应连续小波变换的时变系统物理参数识别

基于小波理论,推导了函数积分运算的连续小波变换计算方法,应用此方法仅用线性时变结构的振动加速度响应信号,就可计算出速度和位移响应信号的连续小波变换值,并将振动微分方程组转换成用小波表示的线性方程组,求解不同时刻的线性代数方程组识别出时变结构的物理参数(质量、刚度和阻尼)。以5自由度时变结构为仿真算例,利用添加噪声的采集信号,识别了突变、线性变化和周期变化3种不同类型的时变物理参数,算例验证了识别方法的正确性、有效性和抗噪声能力。最后还研究了多尺度分析对参数识别的影响,给出了尺度参数区间选取和离散的依据。     

基于小波变换的结构模态参数识别

及时、准确地识别出结构的模态参数是结构健康监测与损伤识别的重要前提。小波分析是众多识别方法中较优越的一种，因其在时一频两域都具有表征信号局部特征的能力，近年来这一方法在线性及非线性系统的参数识别中开始应用。探讨了基于小波脊（Ridge）与小波骨架（Skeleton）的模态参数识别方法，针对小波变换中遇到的边端效应问题，提出了基于自回归滑动平均模型（ARMA）的“预测延拓”方法，并以美国土木工程师学会（ASCE）提供的Benchmark模型为例进行了数值模拟。结果表明，本文提出的方法可以有效地抑制小波边端效应，通过小渡变换可以准确地识别出结构的模态参数。     

基于CACO算法的结构多损伤识别

结构多损伤识别是结构健康监测领域的一个具有挑战性的研究课题,数学上,它通常可以转化为连续函数的约束优化问题。介绍连续优化蚁群算法(CACO)的基本原理、从离散ACO到CACO的实现、寻优路径的构建、以及信息素更新和挥发的数学建模,尝试探寻将CACO算法应用于结构多损伤识别问题的可行性并进行数值仿真和实验研究。通过两层刚架多损伤数值仿真以及三层建筑框架结构4种损伤的实验研究,结果表明:采用CACO算法对结构多损伤进行识别不但能够准确定位结构多损伤,而且还可以有效识别其损伤程度。由此可见,CACO算法应用于结构多损伤识别的效果是显而易见的。     

连铸坯表面裂纹产生原因分析

铸坯表面质量的好坏直接影响最终轧制产品,针对包钢CSP所生产的Q345B热轧板坯出现的裂纹缺陷,阐述了影响铸坯表面纵裂纹、横裂纹形成的原因及防止表面裂纹产生的技术措施。     

Texture Analysis for Crack Detection in Fracture Mechanics

In the context of the metal-mechanic industry, there is a strong demand for material behavior measuring methods in the presence of cracks under stress conditions. This behavior is usually quantified by means of fracture toughness tests using parameters such as the stress intensity factor K, the CTOD, or the J integral. Regardless of the parameters used to quantify fracture toughness, all tests require knowledge of the correct length of the existing fatigue pre-crack in the sample that generated the failure. Hence, most laboratories have adopted visual measurement methods using a stereoscopic magnifying glass or a profile projector. Additional techniques, such as the use of heat tinting, help users and researchers to visualize the stable crack growth front. With the improvement in image processing and computer vision techniques, the present paper proposes the application of a new method for border detection by texture analysis, in order to get the corresponding contour of the crack front in postmortem analyses of SE(B) samples with high precision. The results suggest that the proposed method could be applied with high precision to images of fracture toughness tests for crack length measurement, having achieved a discrimination rate of 98%. The results also suggest that the method can be applied to samples that have not undergone heat tinting.     

小波包分析和自组织神经网络在连铸下渣自动检测中的应用

提出了基于小波包分析和自组织神经网络的钢水连铸下渣自动检测方法.利用小波包分析的高时频分辨率特性来提取特征参数,并利用自组织神经网络进行特征参数的自动聚类,实现正常浇注和下渣信号的自动识别.实验表明该方法可有效地进行下渣的自动检测.     

Advanced Fatigue Crack Detection Using Nonlinear Self-Sensing Impedance Technique for Automated NDE of Metallic Structures

This article reports the application of a nonlinear impedance technique under a low-frequency vibration to detect contact-type structural defects such as fatigue cracks. If the contact-type damage is developed within the structure due to the low-frequency dynamic load, the vibration can cause a nonlinear fluctuation of the structural impedance because of the contact acoustic nonlinearity (CAN). This nonlinear effect can lead to amplitude modulation and phase modulation of the current flow. The nonlinear characteristics of the structural impedance can be extracted by observing the coupled electromechanical impedance of a piezoelectric active sensor and utilizing nonlinear wave modulation spectroscopy. Experimentally, a low-frequency vibration was applied to a notched coupon at a certain natural frequency by a shaker, so that a nonlinear fatigue crack can be artificially formed at the notch tip. Then, the nonlinear features are extracted based on a self-sensing impedance measurement from a host structure under a low-frequency vibration. The damage metric was established based on the nonlinear fluctuation of the impedance due to the CAN.     

On Crack Detection in Tuned and Mistuned Repeating Structures Using the Modal Assurance Criterion

Repeating structures in the form of multiple‐bladed rotors are used widely in turbomachinery. Mistuning in turbomachinery is caused by small differences in individual blade properties because of inevitable material and manufacturing variations, resulting in the splitting of vibration modes of the tuned system. Modal characteristics of the blades are quite sensitive to the level of mistuning present inside the structure. In addition, the existence of damage also results in changed dynamics of the complete system. This paper introduces a modal assurance criterion (MAC)‐based approach for the investigation of small defects such as cracks in a repeating structure. In order to understand the key issues involved, initial work involved a numerical study of a simple comb‐like repeating structure, followed by a detailed numerical and experimental investigation of a tuned and mistuned bladed disc. Changes to the system mode shapes and mode order arising from damage are related to the location and severity of damage. Damage, in the form of small, open cracks, is modelled using different techniques such as follows: material removal, monotonic reduction in the modulus of elasticity of selected elements at the required location and mass modification. Damage indices based on differences in the MAC that give a measure of the change in the mode shapes are introduced. MAC matrices are obtained using a reduced number of data points. The damage index is obtained from the Frobenius norm of the MAC matrix subtracted from the AutoMAC of a reference tuned model without crack. A clear correlation between the damage indices and the crack depth/location is shown. Application of this approach to the limited data obtainable from developing techniques such as blade tip timing is also explained.     

基于连续小波变换的奇异性检测与故障诊断

小波奇异性检测方法作为一种有效的信号局部奇异程度的定量描述方法,已在许多领域展开应用。针对当前普遍采用的基于二进小波变换的奇异性检测方法的不足,建立了基于连续小波变换的奇异性检测方法,具有更细致的局部奇异性刻画能力。本文将这种方法应用在压缩机气阀的故障诊断中,充分显示了该方法的有效性。     

Spatial Wavelet Approach to Local Matrix Crack Detection in Composite Beams with Ply Level Material Uncertainty

Wavelet coefficients based on spatial wavelets are used as damage indicators to identify the damage location as well as the size of the damage in a laminated composite beam with localized matrix cracks. A finite element model of the composite beam is used in conjunction with a matrix crack based damage model to simulate the damaged composite beam structure. The modes of vibration of the beam are analyzed using the wavelet transform in order to identify the location and the extent of the damage by sensing the local perturbations at the damage locations. The location of the damage is identified by a sudden change in spatial distribution of wavelet coefficients. Monte Carlo Simulations (MCS) are used to investigate the effect of ply level uncertainty in composite material properties such as ply longitudinal stiffness, transverse stiffness, shear modulus and Poisson’s ratio on damage detection parameter, wavelet coefficient. In this study, numerical simulations are done for single and multiple damage cases. It is observed that spatial wavelets can be used as a reliable damage detection tool for composite beams with localized matrix cracks which can result from low velocity impact damage.     

Feature Extraction for Crack Detection in Rain Conditions

The solution of the features selection problem is critical for robust detection of crack signals in noisy environment, varying from short-time impulses such as raindrops to the wide-band white Gaussian noise. In this paper, two novel feature selection methods were used to reduce an initial set of 90 features, 67 estimated in the time domain and 23 in the frequency domain, decreasing significantly the memory requirements and the computational complexity of a Radial-Basis-Function (RBF) cracks detector. The evaluation process is carried out in a database including of more than 6000 cracks, raindrops and simultaneous crack and raindrops signals. Additive white Gaussian noise is used to distort the real signals at ?20 to 20 dB Signal to Noise Ratio (SNR). The experimental results show that the number of features can be reduced to approximately 25, without affecting the classification rate of cracks and raindrops in the noisy signals, if the SNR is better than 0 dB. In noise-free environment a classification rate of 91% for a single crack/raindrop event is achieved using only five features. A different set of five features reaches a rate of 85% at 10 dB SNR.     

Multiphysics Modeling and Analysis of the Photoinductive Imaging Effect for Crack Detection

Numerical multiphysics modeling of the photoinductive imaging (PI) effect was performed with a 2-D transient to characterize corner cracks at the edge of a specimen with a bolt hole. We present how the finite-element method (FEM) can be utilized to model the PI effect and observe the influence of critical factors on the coil probe impedance for a rectangular crack in the Ti-6Al-4V specimen. As anticipated, the proposed model can show that the PI method has a higher spatial resolution in the defect in 2-D models compared to the conventional eddy current testing method. The FEM simulation results for 0.25-, 0.50-, and 0.75-mm rectangular notches are shown and discussed. The effects of coil current frequency, laser-point temperature, and lift-off distance on the PI signal are also examined and analyzed. We demonstrate that the PI effect is a novel sensing method for characterizing the geometric shape of cracks and that the enhanced output signals of the coil probe can also be obtained given an appropriate quantity of factors.