1、第43卷第3期2023年6 月文章编号：10 0 0-130 1(2 0 2 3)0 3-0 18 9-0 9地震工程与工程振动EARTHQUAKE ENGINEERING AND ENGINEERING DYNAMICSVol.43 No.3Jun.2023D0I:10.13197/j.eeed.2023.0319基于损伤指数矩阵对层状管道结构损伤识别的有限元模拟王晓初1-2，曲灵芝，李赢3.4（1.沈阳大学建筑工程学院，辽宁沈阳110 0 44；2.沈阳大学辽宁省环境岩土工程重点实验室，辽宁沈阳110 0 44；3.防灾科技学院土木工程学院，河北廊坊0 6 52 0 1；4.中国地震局建筑
2、物破坏机理与防御重点实验室，河北三河0 6 52 0 1)摘要：为研究基于损伤指数对层状管道结构损伤程度识别的可行性，创造性地采用小波包分解方法，以损伤指数为参数，分别研究损伤径向深度与损伤指数的关系和多种轴向位置损伤对损伤指数的影响。使用有限元分析软件ABAQUS,建立在同一轴向位置且多种径向损伤深度的层状管道结构模型，对压电元件位置提取的压电传感信号进行5层小波包分解，计算损伤指数值，将其组成损伤指数矩阵。针对同一轴向损伤位置，结合多种损伤程度组成损伤指数矩阵，分析损伤径向深度与损伤指数的关系并建立拟合曲线。随后改变损伤轴向位置且径向损伤深度同样递增，建立有限元模型，并获得相应的损伤指数拟
3、合曲线。结果表明：在同一损伤轴向位置处，损伤指数拟合曲线都随着损伤径向深度的增加而呈现出先减后增的变化趋势；当改变损伤轴向位置时，损伤与信号接收位置轴向距离的增加使损伤指数的变化幅度减缓，但不改变曲线的总体趋势。研究结果表明：针对传感信号采用小波包分解方法，获得的损伤径向深度与损伤指数的拟合曲线可用于层状管道结构损伤程度的识别,并且该损伤程度识别方法受与信号接收位置轴向距离的限制较小，为快速诊断层状管道结构损伤程度提供了有效的有限元分析方法。关键词：层状管道结构；损伤指数矩阵；压电超声导波；损伤识别；小波包分解中图分类号：TU393.3文献标识码：AFinite element simulat
4、ion for damage identification of layeredpipe structures based on damage index matrixWANG Xiaochu.ou Linghil,.L Yingn(1.School of Architecture and Civil Engineering,Shenyang University,Shenyang 110044,China;2.Liaoning Province KeyLaboratory of Environmental Geotechnical Engineering,Shenyang Universit
5、y,Shenyang 110044,China;3.School of CivilEngineering,Institute of Disaster Prevention,Langfang 065201,China;4.Key Laboratory of Building Collapse Mechanismand Disaster Prevention,Institute of Disaster Prevention,China Earthquake Administration,Sanhe 065201,China)Abstract:To study the feasibility of
6、damage degree of the layered pipeline structures based on damage index,thewavelet packet decomposition method is adopted creatively,and the damage index is used as the parameter.Twoaspects are studied respectively,one is the relationship between the radial depth of the damage and the damageindex,and
7、 the other is the influence of various axial positions on the damage index.Finite element analysis software收稿日期：2 0 2 2-0 2-19；修回日期：2 0 2 2-0 5-10基金项目：辽宁省教育厅科学研究-青年“育苗”项目（2 0 2 0 JYT06）；中央高校基本科研业务费专项（ZY20220201）；河北省高等学校科学研究项目（QN2023146）；廊坊市市级科技计划项目（2 0 2 2 0 130 91）；辽宁省自然科学基金项目（2 0 2 1-BS-278）Suppor
8、ted by:Scientific Research-Youth“Cultivation Foundation of Education Department of Liaoning Province(2020JYT06);FundamentalResearch Funds for the Central Universities(ZY20220201);Science and Technology Project of Hebei Education Department(QN2023146);Langfang Municipal Science and Technology Plan Pr
9、oject(2022013091);Natural Science Foundation of LiaoningProvince(2021-BS-278)作者简介：王晓初（196 7 一），男，教授，博士，主要从事土木工程结构健康监测研究。E-mail：w a n g x i a o c h u 12 6.c o m通讯作者：李赢（198 5），女，副教授，博士，主要从事结构健康监测和诊断方法研究。E-mail:liying-190ABAQUS is used to establish the layered pipeline structure model with the same axi
10、al position and various radialdamage depths.Through the wavelet packet decomposition,the piezoelectric sensing signal extracted from thepositions of the piezoelectric element is decomposed into five layers,and the damage index value is calculated,andit is formed into a damage index matrix.For the sa
11、me axial damage position,combined with the damage indexmatrix that is composed of various damage degrees,not only the relationship between damage radial depth and damageindex is analyzed,but also the fitting curve is established.Subsequently,the damage axial position is changed,andthe radial damage
12、depth is also increased.The finite element models are established,and the fitting curve ofdamage index is obtained.The results show that:at the same damage axial position,the damage index fitting curveshows a trend of first decreasing and then increasing with the increase of the radial depth of the
13、damage.When theaxial position of the damage is changed,the increase of the axial distance between the damage and the signalreceiving position slows down the variation of the damage index but does not change the overall trend of the curve.This paper shows that the wavelet packet decomposition method
14、is used for the sensing signal,and the obtainedfitting curve of the damage radial depth and the damage index can be used to identify the damage degree of thelayered pipeline structure.The damage degree identification method is less limited by the axial distance from thereceiving position of signal a
15、nd provides a finite element analysis method for rapidly diagnosing the damage degreeof the layered pipeline structure.Key words:layered pipeline structures;damage index matrix;piezoelectric guided waves;damage identification;wavelet packet decomposition地震工程与工程振动第43卷0引言管道结构因具有运输距离短、不受地形限制、不占用地表空间等优势
16、，被广泛应用于输送石油、天然气、水等物质。在实际使用中,直接接触运输物质的管道结构层容易发生损伤，而超声导波因频散现象较小并且能进行长距离检测，被应用于管道损伤检测。通过ABAQUS有限元分析软件，模拟导波在层状管道结构的传播性质，并采用小波包分解法，分析接收信号的特点，基于损伤指数矩阵实现层状管道结构的损伤识别。在以往研究中,学者们采用小波分解方法进行信号处理和结构损伤识别研究。WANG 等1通过对入侵信号进行小波分解，得到不同频段的能量比。将提取到的能量比作为入侵类型的特征,从而实现对人侵源的检测与识别。IRADIRATU等2 进行了电机轴承损伤实验,对接收信号进行小波分解并计算出平均信号功率。研究表明，当轴承上存在损伤时，平均功率值趋于增加，说明小波分解方法可对轴承进行损伤检测。黄擎等3提出了一种名为WAVEDEC CNN的水声目标识别方法，该识别方法将原始信号进行小波分解，再输人改进后的CNN中，从而实现对目标的识别,实验证明该方法提高了识别率。CHUMCHU4 使用离散小波分解对捕获的声学信号进行预处理，再通过人工神经网络对预处理信号进行建模，从而确定供水系统中的泄漏位置。P