运用统计小波的光纤光栅结构健康监测技术

针对振动型结构健康监测方法的特点,搭建了基于非平衡M-Z干涉仪和相位载波解调技术的光纤光栅损伤识别系统。运用小波包分解振动信号,建立了基于小波包节点能量相对变化率之和的结构损伤识别指标;介绍了统计过程控制原理,推导了使用均值-极差控制图分析损伤识别指标,识别结构连续损伤的过程。实验测试了铝制简支梁结构处于健康状态和3种损伤状态下的各40次振动信号。信号时域图显示各状态振动信号持续时间均约为0.05ms,幅值基本相同。依据结构健康状态下的统计过程控制限(12.85,41.35)进行了均值-极差控制图损伤识别分析,结果表明,搭建的损伤识别系统能连续地对结构进行健康监测。     

排列熵算法在水工结构损伤诊断中的应用

提出一种基于排列熵算法(permutation entropy,简称PE)的水工结构损伤诊断方法。首先,运用小波阈值-经验模态分解(empirical mode decomposition,简称EMD)降噪方法对原始信号进行降噪,减小环境噪声对结构损伤特征信息的干扰,提高信号的信噪比;其次,运用排列熵算法检测降噪后信号的复杂度,并计算其排列熵值。通过不同工况下信号熵值变化规律的对比,实现水工结构损伤的诊断。将该方法应用于泄流激励下悬臂梁模型的试验研究,结果表明,正常无损状态下结构振动信号的排列熵值最大;结构发生损伤时,其熵值降低,且损伤程度越大,熵值越小;排列熵对结构的初期损伤比较敏感;结构未发生损伤时,不同工况下的排列熵基本不变,说明排列熵能够有效确定结构的损伤,且具有较高的诊断精度。     

基于随机森林的建筑结构损伤识别方法

针对利用分类器对建筑结构进行损伤识别的问题,引入一种新的组合分类器算法——随机森林,提出基于小波包分解和随机森林的结构损伤识别方法。首先,采用小波包对结构在不同损伤程度和位置上的振动加速度信号进行分解,得到各个频带上的总能量;然后,利用各频带上能量值存在着差异性作为输入到分类器的特征向量;最后,训练随机森林模型并对建筑结构的损伤位置和损伤程度进行识别。应用该方法对一座8层剪切型钢框架结构进行损伤判别,并与BP神经网络和支持向量机方法进行对比,结果表明该方法具有较好的识别精度与稳定性。     

基于PCA-PDBNs的故障检测与自学习辨识

如何提高工业过程故障识别的准确性及其算法训练的效率一直是故障检测与辨识研究领域的重点和热点。将深度学习方法引入该领域,结合粒子群优化(PSO)算法和深度信念网络(DBNs),提出了一种基于PSO的DBNs辨识方法(即PSODBNs,PDBNs),使用该方法对复杂函数的拟合进行了数值仿真。实验结果表明,相比于基本的DBNs模型,经PSO算法对网络参数优化后的DBNs模型获得了更好的函数逼近效果,具有更高的辨识精度。为验证该方法在实际工业过程故障检测中的可行性,结合主元分析(PCA),提出了一种PCA-PDBNs模型,并将此应用于田纳西-伊斯曼(TE)过程的故障检测中,结果表明,基于PCA-PDBNs方法降低了故障检测模型的复杂度,进一步提高了对未知故障类型的辨识精度,取得了较好效果。     

An approach to the problem of damage identification in an elastic rod based on the Timoshenko beam model

The solution to the problem of damage identification in a cantilever based on minimization of a special functional of the discrepancy between the solution (eigenfrequencies and vibration modes) that was obtained analytically based on the Timoshenko beam model and using other methods, in particular, finite-element analysis (FEA), is presented. The results of the numerical calculations of the position and depth of a cut in a cantilever are compared with the data of the full-scale experiment, which verified the high degree of their reliability. The advantages of the proposed model as compared to FEA and the well-known model of A.D. Dimoragonas, which lie in the higher efficiency of calculations and the reliability of the reconstruction of the shapes of different vibrational modes of a damaged structure, are demonstrated. Owing to these advantages, the model can be used in optimization algorithms and applied for teaching neural networks. The correlation between the efficiency of damage identification and the position of the points of exciting load application on a tested structure is revealed. The proposed method for identification of damage can be used in systems that are intended for the diagnosis of aviation and building structures.     

Improvement of damage detection methods based on experimental modal parameters

The objective of this paper is to introduce a new method for structural damage detection based on experimentally obtained modal parameters. The new method is suitable for detection of fatigue damage occurring in an aluminium cantilever beam. The damage has been practically realised as saw cuts of different sizes and at different locations. The first step of analysis included an attempt of damage identification with the most often used damage indicators based on measured modal parameters. For that purpose special signal processing technique has been proposed improving the effectiveness of indicators tested. However the results obtained have not been satisfactory. That was the motivation for defining new damage indicators (frequency change based damage indicator, Hybrid Damage Detection method), utilising the change of natural frequencies and any mode shape (measured or modelled) as the measurement of frequencies is much less time consuming in comparison to total mode shape measurement. It has been shown that the proposed technique is suitable for damage localisation in beam-like structures.     

Three-steps-meshing based multiple crack identification for structures and its experimental studies

Multiple crack identification plays an important role in vibration-based crack identification of structures. Traditional crack detection method of single crack is difficult to be used in multiple crack diagnosis. A three-step-meshing method for the multiple cracks identification in structures is presented. Firstly, the changes in natural frequency of a structure with various crack locations and depth are accurately obtained by means of wavelet finite element method, and then the damage coefficient method is used to determine the number and the region of cracks. Secondly, different regions in the cracked structure are divided into meshes with different scales, and then the small unit containing cracks in the damaged area is gradually located by iterative computation. Lastly, by finding the points of intersection of three frequency contour lines in the small unit, the crack location and depth are identified. In order to verify the effectiveness of the presented method, a multiple cracks identification experiment is carried out. The diagnostic tests on a cantilever beam under two working conditions show the accuracy of the proposed method: with a maximum error of crack location identification 2.7% and of depth identification 5.2%. The method is able to detect multiple crack of beam with less subdivision and higher precision, and can be developed as a multiple crack detection approach for complicated structures.     

Structural damage identification of a cantilever beam using excitation force level control

Most studies in damage identification so far have concentrated on comparing modal parameters of a structure with an open crack with those of an intact structure. In this study, a new damage identification method for beam-like structures with a fatigue crack is proposed, which does not require comparative measurement on an intact structure but several measurements at different level of excitation forces on the cracked structure. The idea comes from the fact that dynamic behavior of a structure with a fatigue crack changes with the level of the excitation force. In other words, a beam with a real fatigue crack would behave as an intact beam at low excitation forces due to the crack closure. The 2nd spatial derivatives of frequency response functions along the longitudinal direction of a beam are used as the sensitive indicator of crack existence. Then, weighting function is employed in the averaging process in frequency domain to account for the modal participation of the differences between the dynamic behavior of beam with a fatigue crack at the low excitation and one at the high excitation. Subsequently, a damage index is defined such that the location and level of the crack may be identified. Finally, it is shown that damage identification method using the proposed damage index is very successful through experiment and finite element analysis.     

基于应变模态的模态应变能损伤识别方法

传统模态应变能计算需要完备模态振型信息,而模态振型信息中存在转角自由度难以准确获取的问题,为解决该问题,开展基于应变模态的模态应变能损伤识别研究,实现了结构损伤的定量识别。首先,通过基于应变与位移之间的联系,推导出应变模态与位移模态之间的转换矩阵;其次,利用应变模态代替位移模态计算单元模态应变能,建立基于灵敏度分析的损伤识别方程组;最后,根据奇异值截断法求解该方程组识别结构损伤。以一两端固支梁结构为对象,开展数值仿真和实验研究。结果表明,该方法可以有效识别出结构的损伤位置和损伤程度,相对于基于振型扩充的模态应变能损伤识别方法,具有更好精度和抗噪性能。     

冷轧过程断带故障的诊断研究

在冷轧过程中,断带故障是冷轧工序的主要生产故障之一。针对冷轧过程断带故障的特点,提出一种基于核主元分析(KPCA)非线性特征提取和最小二乘支持向量机(LSSVM)分类的故障诊断方法。此方法采用KPCA理论将冷轧过程原始空间数据映射到高维空间,并在高维空间进行主元分析,从而降维、去相关性,得到冷轧过程非线性特征向量。将降维后的特征主元作为LSSVM输入进行训练和识别,根据LSSVM的输出结果判断冷轧过程工作状态与故障类型。仿真结果表明:基于KPCA非线性特征提取和LSSVM分类的故障诊断方法计算速度快,能有效地提取冷轧过程断带故障特征,识别断带故障类型。     

LAMOST平衡系统的传感器优化布置

针对模态测试中的传感器优化布置问题,提出一种新的主元分析和组合MAC联合算法。首先,将多种传感器布置方法的结果形成原始特征数据矩阵,利用主元分析获得空间测点的主元和综合评价值,据此可以合理地大幅减少候选测点数目;其次,使用组合MAC法计算所有组合的MAC矩阵,取最大非对角元最小的一组作为最终选择;然后,以悬臂梁为例,分别用联合算法和4种传统算法对其进行传感器优化布置,通过3个评价准则的数值比较,证明联合算法的优越性;最后,将联合算法应用在望远镜LAMOST的平衡系统上,得到了两个方向上的传感器布置方案,进一步验证了联合算法的可行性。     

藏式古建筑木结构损伤识别的数值模拟

为保护藏式古建筑木结构,针对实际中众多结构存在的梁、柱构件损伤问题进行了损伤识别理论的研究。基于实地勘查结果,对梁、柱构件的残损类型和残损机理进行了汇总和分析。以某典型藏式古建筑多层梁柱排架结构为损伤识别对象,在建立其有限元模型的基础上,采用振动响应灵敏度损伤识别方法对该结构梁、柱构件的3种不同损伤状态进行了数值模拟识别。为了模拟古建筑结构构件之间物理参数离散性大的特点,给每个构件的弹性模量添加了随机误差。识别时,为了减小离散的物理参数对结果收敛性和准确性的影响,采用修正的正则化方法求解识别方程。结果表明,该方法可以克服测量噪声的影响,准确识别出藏式古建筑木结构梁、柱构件的损伤,为该类型结构实际的损伤识别提供了理论依据。     

简支梁桥损伤程度识别的模拟和试验

在结合钢桁桥损伤程度识别方法的基础上,提出了适用于简支梁结构的两种损伤程度识别方法:整体振型的相关系数法和保证准则法,将其应用到实验室简支梁结构上分别进行数值模拟和试验。脉冲激励下的结果表明,两种方法能较准确地识别损伤单元的等效损伤程度,具有很强的抗噪能力。最后,探讨了激励对提出方法的影响,为工程应用奠定了基础。     

梁类结构多裂纹微弱损伤的小波有限元定量检测方法

提出了一种定量检测梁类结构多裂纹参数的方法。利用适宜求解奇异性问题的小波有限元法,从动力学正问题入手,对裂纹梁进行有限元建模,获得裂纹故障在结构固有频率上反映的本质征兆,并利用曲面拟合技术绘制出以裂纹位置和深度作为变量的固有频率变化率曲面,然后对整个裂纹梁进行剖分,迭代求解出每个剖分单元上的结构损伤系数。损伤系数为正的单元诊断为裂纹单元,在每个裂纹单元上求出裂纹对应的前三阶固有频率变化率,并分别将其作为输入参数代入固有频率变化率曲面,做出前三阶模态的频率变化率等高线,最后通过三条等高线的交点预测出裂纹存在的位置和深度,算例分析验证了该算法的有效性。     

Decision tree and PCA-based fault diagnosis of rotating machinery

After analysing the flaws of conventional fault diagnosis methods, data mining technology is introduced to fault diagnosis field, and a new method based on C4.5 decision tree and principal component analysis (PCA) is proposed. In this method, PCA is used to reduce features after data collection, preprocessing and feature extraction. Then, C4.5 is trained by using the samples to generate a decision tree model with diagnosis knowledge. At last the tree model is used to make diagnosis analysis. To validate the method proposed, six kinds of running states (normal or without any defect, unbalance, rotor radial rub, oil whirl, shaft crack and a simultaneous state of unbalance and radial rub), are simulated on Bently Rotor Kit RK4 to test C4.5 and PCA-based method and back-propagation neural network (BPNN). The result shows that C4.5 and PCA-based diagnosis method has higher accuracy and needs less training time than BPNN.     

The research on structural damage identification using rough set and integrated neural network

A huge amount of information and identification accuracy in large civil engineering structural damage identification has not been addressed yet. To efficiently solve this problem, a new damage identification method based on rough set and integrated neural network is first proposed. In brief, rough set was used to reduce attributes so as to decrease spatial dimensions of data and extract effective features. And then the reduced attributes will be put into the sub-neural network. The sub-neural network can give the preliminary diagnosis from different aspects of damage. The decision fusion network will give the final damage identification results. The identification examples show that this method can simplify the redundant information to reduce the neural network model, making full use of the range of information to effectively improve the accuracy of structural damage identification.     

基于应变模态分析的长输油气管线损伤检测仿真研究

基于应变模态方法,运用有限元分析软件ANSYS对石油工业中的压力管道损伤识别进行了初步的研究。ANSYS的模态分析可以求得结构在损伤前后的位移模态,将其转换为对应阶数的应变模态,并对比应变模态损伤前后的变化,作出应变模态差值曲线以此来判断损伤的存在和位置以及损伤程度。计算结果表明：利用应变模态差值曲线能比较准确地识别出管道结构模型的损伤存在和损伤位置,并依此定性地识别出管道结构模型的损伤程度。采用此方法进行管道结构进行损伤识别比传统的损伤检测方法更加准确、方便。     

基于邻域属性重要度与主成分分析的齿轮箱故障特征约简

为有效降低齿轮箱故障特征的维数并提高诊断效率,提出了基于邻域属性重要度与主成分分析法相结合的齿轮箱故障特征约简方法,并利用支持向量机和BP神经网络对诊断的准确率进行对比分析。针对齿轮箱中具有不同程度裂纹的齿轮,选取其时域、频域和基于希尔伯特变换的36个特征;将邻域模型引入到特征属性的约简,构造前向贪心算法,以邻域属性重要度较大的9个特征作为特征集,提取累积贡献率达到95%以上的主成分,分别输入支持向量机和BP神经网络分类器中进行分类识别,并与不经过特征优选的主成分特征融合相对比。结果表明,采用基于邻域属性重要度与主成分分析法相结合的特征约简方法,既可以降低齿轮箱故障特征的维数,又不影响对其运行状态的表征,有助于识别不同裂纹水平的齿轮,与不经过特征优选直接进行融合的方法相比,所提出方法诊断准确率更高,训练时间更短。     

改进粒子群优化的LSSVM结构损伤评定

为改善结构动力损伤的识别效果,提出了刚度变化指标构架下改进粒子群算法优化的最小二乘支持向量机的结构损伤评估方法。首先,通过由试验技术修正的有限元模型来计算刚度变化指标(stiffness variation index,简称SVI),并进行损伤定位;然后,在SVI基础上,利用改进粒子群算法优化最小二乘支持向量机的超参数,建立结构损伤评估优化模型,计算损伤大小。将该方法用于起重机主梁的损伤评定,研究结果表明,该方法具有较高的精度和效率,能准确地判断结构的实际性态,是一种有效的评估手段。     

基于频率变化率的结构刚度非均匀退化识别

针对不利环境作用、损伤等易造成结构局部损伤且刚度退化程度不均匀的问题,以受弯梁为研究对象,从构件动力特性入手,综合考虑损伤前后的模态挠度曲率和固有频率变化,提出了基于频率变化率的刚度非均匀退化识别方法。首先,在柔度矩阵的基础上推导模态挠度曲率,通过损伤前后模态挠度曲率的改变量识别损伤位置参数,判定损伤区域;其次,对损伤区域进行节段划分,从欧拉-伯努利梁的动力方程出发建立损伤程度、损伤区域位置参数与固有频率之间的矩阵函数,实现直接利用频率值变化评估构件不同区域损伤程度。研究结果表明,该方法能很好地识别结构局部损伤位置和损伤程度,尤其是对于结构局部刚度不均匀退化的评估具有明显的优势。