基于转角模态和小波神经网络的连续梁损伤识别研究

基于小波奇异性检测原理和神经网络非线性映射能力,结合结构基本模态参数,提出了一种结合小波神经网络与结构转角模态的损伤识别方法．首先,建立三跨连续梁的有限元模型获取结构模态参数,并对其进行Mexihat小波变换,通过系数图突变点判断结构损伤位置．然后,将小波系数模特征向量作为BP神经网络的输入,分别研究了该方法在单损伤和多损伤工况下的识别能力．最后将不同工况下神经网络预测值与结构实际损伤程度进行对比,得到单处损伤预测误差平均值为0.22%,多处损伤预测误差平均值分别为0.22%和0.18%,结果表明该方法在结构损伤识别方面的有较高有效性及精确度．     

利用振动模态测量值和神经网络方法的结构损伤识别研究

提出了一种基于模态测量参数和神经网络的结构损伤检测方法，建造了两种输入方式的BP神经网络，即自振频率以及结合自振频率与振型，并讨论了不同数量的输入信息对结构损伤检测精度和计算效率的影响。证明了输入的参数越多，神经网络就越聪明，训练的收敛速度越快；以及在保证一定的测量精度的情况下，基于频率与振型的损伤识别结果要好于基于频率的检测结果。最后，通过对3层框架模型的4种损伤工况下的结构损伤检测结果的分析，认为利用模态测量参数和神经网络方法能够准确地识别结构损伤的位置，而且能较精确地识别结构损伤的大小。     

组合神经网络在结构损伤检测中的应用

子结构的动态响应变化与整体结构相比，对结构内部损伤反应更为敏感。组合神经网络可以克服单个神经网络功能的单一局限性，实现更加全面综合的仿真识别功能。本文首先运用双协调自由界面模态综合法对结构进行模态分析，获取各子结构及整体结构的模态信息。然后，通过组合BP神经网络将损伤子结构与整体结构的模态频率变化率组合起来进行结构损伤检测。该方法在改善网络训练性能的同时，提高了检测结果的准确性和可靠性。文章最后通过数值算例验证了该方法的可行性和有效性。     

既有钢桁梁桥损伤识别与诊断方法研究

从结构动力学基本理论出发，对既有钢桁梁桥的损伤识别与诊断方法进行了研究. 通过选取钢桁梁桥损伤前、后的固有频率作为特征参数，并应用BP神经网络方法和MATLAB方法对桥梁损伤识别和诊断过程进行了深入细致地分析，提出了一套完整的桥梁损伤识别和诊断过程流程图. 最后，用一实际钢桁梁桥的损伤识别进行了数值模拟，计算结果证明了该方法的准确性和有效性.     

灰色相关性分析在结构静力损伤识别中的应用

基于灰色理论的相关性分析方法，首次提出了灰色曲率关联系数的概念并将其应用到结构的静力损伤识别中，提出了对局部损伤十分敏感的静态位移曲率置信因子SDCACi，通过该因子的大小对各节点所连接的单元是否会发生损伤进行精确的判断，然后运用最小二乘法对损伤区域的损伤程度进行识别．并将该方法应用于两端固支梁的损伤识别中，由识别结果可以证明：不论测量数据(用有限元仿真计算并考虑了测量误差)的多少，该方法对结构中的单损伤和多损伤都能进行准确的定位，因此该方法在大型结构及复杂结构的损伤识别中具有广阔的应用前景．     

基于振动信号神经网络层合板分层损伤检测研究

基于振动信号应用神经网络研究层合板分层损伤的检测方法. 对层合板分层损伤区域, 采 用相同坐标不同节点建立了分层损伤处的有限元模型; 通过数值模拟提取结构无损和 不同程度面积分层损伤的全局振动标识量; 重点研究神经网络对层合板分层损伤位置 和损伤程度的检测技术. 研究表明, 用结构全局振动标识量作为人工神经网络的输入, 对 层合板结构分层损伤检测是一种很有效的工程实用技术, 可应用于实际结构的在线损伤检测.     

基于固有频率向量的结构损伤检测方法

基于损伤前后结构固有频率的变化,引入了固有频率向量及固有频率向量置信准则的概念,提出了基于固有频率向量的结构损伤检测方法.首先建立完好结构的有限元模型,并模拟出结构的若干种损伤模式,分别求出完好结构与各损伤模式结构的固有频率向量,从而构建出结构损伤特征数据库,随后对具有某种未知损伤的结构,根据该结构的固有频率向量与损伤数据库中每一个固有频率向量之间的相关程度--固有频率向量置信准则,由最大置信准则来判定对应的损伤模式.应用该方法对一个八层剪切空间框架模型的损伤进行了仿真检测,对于无测量误差的情况,在任何一层的支柱损伤大于最低可检损伤的情况下,都能准确的检测出损伤的位置及程度;考虑了固有频率测量的随机误差之后,应用损伤检测概率进行评估,该方法也具有较高的损伤检测概率.     

基于遗传神经网络的锚杆极限承载力预测的研究

针对BP人工神经网络具有易陷入局部极小等缺陷,提出了将遗传算法与神经网络结合,同时优化网络结构的权值与阈值的思想,建立了基于遗传算法的锚杆极限承载力预测的遗传神经网络模型。该模型以低应变动测的5个变量作为输入变量来对锚杆极限承载力进行预测,并与BP神经网络预测结果进行比较。数值算例表明,遗传神经网络在锚杆极限承载力预测中具有较高的计算效率和识别精度。     

改进BP网络的海底地形辅助导航算法

鉴于传统的迭代最近点算法存在着易陷入局部最优的缺陷和实时性不好的问题,提出了一种将BP神经网络引入迭代最近点算法中进行地形匹配的新方法。针对传统BP算法存在的局部极小和收敛速度慢等缺点,采用自适应学习方法、引入动量因子、可变化的学习率因子和可调激活函数等措施进行了BP算法的改进。仿真结果表明,改进后的算法可以在一定程度上克服由于局部收敛带来的匹配失效问题,能够获得很好的匹配效果,同时也解决了在实时性上存在的突出问题。     

基于移动载荷响应的多跨连续桥梁损伤检测

应用有限元法将连续桥梁结构离散为两跨弱耦合欧拉-伯努利梁模型,研究了该弱耦合梁系统由于局部损伤而引起的振动模态局部化现象.利用Newmark直接积分法求出在移动车载作用下桥梁的动态响应,并进一步推导出动态响应桥梁对物理参数(如杨氏模量)的时域灵敏度.在反问题中利用基于响应灵敏度的有限元模型修正法识别出桥梁的局部损伤,并讨论了人工测量噪声对损伤识别结果的影响.算例表明该方法能够快速有效地检测具有相重频率的弱耦合系统的局部损伤,并具有精度高、对测量噪声不敏感等特点.     

Damage detection in structures using modified back-propagation neural networks

A nonparametric structural damage detection methodology based on neural networks method is presented for health monitoring of structure-unknown systems. In this approach appropriate neural networks are trained by use of the modal test data from a ‘healthy’ structure. The trained networks which are subsequently fed with vibration measurements from the same structure in different stages have the capability of recognizing the location and the content of structural damage and thereby can monitor the health of the structure. A modified back-propagation neural network is proposed to solve the two practical problems encountered by the traditional back-propagation method, i.e., slow learning progress and convergence to a false local minimum. Various training algorithms, types of the input layer and numbers of the nodes in the input layer are considered. Numerical example results from a 5-degree-of-freedom spring-mass structure and analyses on the experimental data of an actual 5-storey-steel-frame demonstrate that neural-networks-based method is a robust procedure and a practical tool for the detection of structural damage, and that the modified back-propagation algorithm could improve the computational efficiency as well as the accuracy of detection Project supported by the National Natural Science Foundation of China (No. 59908003) and the Natural Science Foundation of Hubei Province (No. 99J035).     

Recognition of gas–liquid two-phase flow patterns based on improved local binary pattern operator

A new method to pattern recognition of gas–liquid two-phase flow regimes based on improved local binary pattern (LBP) operator is proposed in this paper. Five statistic features are computed using the texture pattern matrix obtained from the improved LBP. The support vector machine and back-propagation neural network are trained to flow pattern recognition of five typical gas–liquid flow regimes. Experimental results demonstrate that the proposed method has achieved better recognition accuracy rates than others. It can provide reliable reference for other indirect measurement used to analyze flow patterns by its physical objectivity.     

Combined neural network and reduced FRF techniques for slight damage detection using measured response data

Summary  This paper deals with structural damage detection using measured frequency response functions (FRF) as input data to artificial neural networks (ANN). A major obstacle, the impracticality of using full-size FRF data with ANNs, was circumvented by applying a data-reduction technique based on principal component analysis (PCA). The compressed FRFs, represented by their projection onto the most significant principal components, were used as the ANN input variables instead of the raw FRF data. The output is a prediction of the actual state of the specimen, i.e. healthy or damaged. A further advantage of this particular approach is its ability to deal with relatively high measurement noise, which is a common occurrence when dealing with industrial structures. The methodology was applied to detect three different states of a space antenna: reference, slight mass damage and slight stiffness damage. About 600 FRF measurements, each with 1024 spectral points, were included in the analysis. Six 2-hidden layer networks, each with an individually-optimised architecture for a specific FRF reduction level, were used for damage detection. The results showed that it was possible to distinguish between the three states of the antenna with good accuracy, subject to using an adequate number of principal components together with a suitable neural network configuration. It was also found that the quality of the raw FRF data remained a major consideration, though the method was able to filter out some of the measurement noise. The convergence and detection properties of the networks were improved significantly by removing those FRFs associated with measurement errors. Received 9 March 2000; accepted for publication 12 December 2000     

LOCATING BEAM DAMAGE BASED ON INITIAL FLEXURAL STIFFNESS DETECTION

应变模态变化率指标在服役梁结构的损伤定位方面已有应用,但现有研究大多忽视了梁上的初始局部抗弯刚度具有离散性的情况,因此难以区分真实损伤和初始离散性造成的局部刚度变化,对于实际梁结构的应用效果不佳. 先提出了一种通过求解线性方程组来得到梁上各区间真实初始抗弯刚度的方法,然后采用应变模态变化率指标来进行损伤定位. 研究结果表明,该方法可以处理梁上初始局部抗弯刚度具有离散性的情况,实现准确的损伤定位.     

基于独立分量分析的多源动态载荷识别方法

提出了基于独立分量分析的多源动态载荷识别方法, 解决了在结构系统未知的情况下载荷波形的识别问题. 该方法基于结构在多源动态载荷作用下, 其响应是载荷与对应的结构脉冲响应卷积的原理, 并假设载荷源相互统计独立. 与既有的载荷识别方法相比,该识别方法特点表现在: 结构质量, 刚度及阻尼等信息可以完全未知, 但以实际载荷间的独立性为优化目标; 用互信息来度量识别载荷间的独立性, 通过梯度下降算法取消识别载荷间的各阶相关性, 使识别载荷间基本满足相互独立; 从波形的角度来进行载荷识别.通过数值仿真表明: 该方法对测点, 噪声, 不同载荷形式及不同结构有较好的鲁棒性; 识别载荷与实际载荷在归一化条件下, 识别载荷与实际载荷相关性系数约为1.      

基于扩展有限元的结构内部缺陷(夹杂)的反演分析模型

传统的结构检测方法一般需要钻孔取样,对结构本身有一定的破坏作用,而无损检测方法在检测过程中不破坏结构本身,这项技术的重要性日益显著. 结合扩展有限元法和人工蜂群智能优化算法的优点,建立了结构内部缺陷(夹杂)的反演分析模型,为结构的无损检测技术提供了一条新的途径.扩展有限元法通过引入非连续位移模式可以在不重新划分网格的情况下通过改变水平集函数反映缺陷(夹杂)的位置及大小,避免了反演分析每次迭代过程中的网格重剖分,人工蜂群智能优化算法在每次迭代中都采用全局和局部搜索,找到最优解的概率大大增加并可很好地避免局部最优,因此,扩展有限元法与人工蜂群智能优化算法的结合有效地减少了反演分析的计算工作量. 通过若干算例的分析表明:建立的反演分析模型能准确地探测结构内部存在的单个缺陷(夹杂).      

APPLICATION OF STOCHASTIC FINITE ELEMENT METHOD TO STRENGTH AND STABILITY ANALYSIS OF EARTH DAMS

This paper applies the stochastic finite element method to analyse thestatistics of stresses in earth dams and assess the safety and reliability of the dams.Formulations of the stochastic finite element method are briefly reviewed and theprocedure for assessing dam's strength and stability is described.As an example,adetailed analysis for an actual dam-Nululin dam is performed.A practical method forstudying built-dams based on the prototype observation data is described.     

A 2-D continuous wavelet transform of mode shape data for damage detection of plate structures

A two-dimensional (2-D) continuous wavelet transform (CWT)-based damage detection algorithm using “Dergauss2d” wavelet for plate-type structures is presented. The 2-D CWT considered in this study is based on the formulation by Antoine et al. (2004). A concept of isosurface of 2-D wavelet coefficients is proposed, and it is generated to indicate the location and approximate shape or area of the damage. The proposed algorithm is a response-based damage detection technique which only requires the mode shapes of the damaged plates. This algorithm is applied to the numerical vibration mode shapes of a cantilever plate with different types of damage to illustrate its effectiveness and viability. A comparative study with other two 2-D damage detection algorithms, i.e., 2-D gapped smoothing method (GSM) and 2-D strain energy method (SEM), is performed, and it demonstrates that the proposed 2-D CWT-based algorithm is superior in noise immunity and robust with limited sensor data. The algorithm is further implemented in an experimental modal test to detect impact damage in an FRP composite plate using smart piezoelectric actuators and sensors, demonstrating its applicability to the experimental mode shapes. The present 2-D CWT-based algorithm is among a few limited studies in the literature to explore the application of 2-D wavelets in damage detection, and as demonstrated in this study, it can be used as a viable and effective technique for damage identification of plate- or shell-type structures.     

基于BP神经网络与小冲杆试验确定在役管道钢弹塑性性能方法研究

为了能够在不停输油气工况下获得在役管道材料的弹塑性力学性能, 提出了一种人工智能BP (back-propagation)神经网络、小冲杆试验与有限元模拟相结合,通过确定材料真应力-应变曲线从而获得材料弹塑性力学性能的方法. 首先,通过系统改变Hollomon公式中的参数$K$, $n$值,获得457组具有不同弹塑性力学性能的假想材料本构关系, 其次,将得到的本构关系代入经试验验证的含有Gurson-Tvergaard-Needleman(GTN)损伤参数的小冲杆试验二维轴对称有限元模型,通过有限元计算得到了与真应力-应变曲线一一对应的457条不同假想材料的载荷-位移曲线,最终将两组数据作为数据库输入BP神经网络进行训练,建立了同种材料小冲杆试验载荷-位移曲线与真应力-应变曲线之间的关联关系.通过此关联关系,可利用试验得到的小冲杆载荷-位移曲线获取在役管道钢的真应力-应变曲线,从而确定其弹塑性力学性能.通过对比BP神经网络得到的X80管道钢真应力-应变曲线与单轴拉伸试验的结果以及引用现有文献中不同材料的试验数据对此关系进行验证,证明了该方法的准确性与广泛适用性.