基于支持向量机的梁桥多位置损伤识别研究
|
摘要
为实现更准确的梁桥多位置损伤识别,从理论角度分析得出某位置的损伤会对周边位置的位移和曲率产生一定影响。曲率模态差的突变是损伤的典型特征,以曲率模态差变化率作为量化突变的参数,并将其归一化处理后输入支持向量机中进行桥梁多位置损伤识别。以某多跨连续刚构桥为例,将多处典型截面发生10%,30%和50%刚度折减情况下的前2阶竖向振动的曲率模态差变化率以及各位置的损伤状态作为特征向量去训练支持向量机。之后将预测集中包含20%和40%刚度折减的曲率模态差变化率输入训练好的支持向量机中去识别各位置的损伤情况,其识别准确率达到99.68%。
In order to identify multi position damage of beam bridge more accurately, through the theoretical analyzing, it is concluded that the damage of a certain position will have a certain effect on the displacement and curvature of the surrounding position. However, the sudden change of curvature modal difference is the typical feature of damage, and the change rate of curvature modal difference is used to quantize the sudden change. Then normalize and input them into the support vector machine to identify multi position damage. Taking a multi-span continuous rigid frame bridge as an example, typical cross section appears 10%, 30% and 50% stiffness reduction.The change rate of curvature modal difference of the 1 st and 2 nd ordered vertical vibration and the damage state of each position are used as eigenvectors to train the support vector machine. After that, the change rate of curvature modal difference with 20% and 40% stiffness reduction in the prediction set is input into the trained support vector machine to identify the damage of each position, and the recognition accuracy is 99.68%. It is proved that this method has good accuracy and generalization in damage identification.
In order to identify multi position damage of beam bridge more accurately, through the theoretical analyzing, it is concluded that the damage of a certain position will have a certain effect on the displacement and curvature of the surrounding position. However, the sudden change of curvature modal difference is the typical feature of damage, and the change rate of curvature modal difference is used to quantize the sudden change. Then normalize and input them into the support vector machine to identify multi position damage. Taking a multi-span continuous rigid frame bridge as an example, typical cross section appears 10%, 30% and 50% stiffness reduction.The change rate of curvature modal difference of the 1 st and 2 nd ordered vertical vibration and the damage state of each position are used as eigenvectors to train the support vector machine. After that, the change rate of curvature modal difference with 20% and 40% stiffness reduction in the prediction set is input into the trained support vector machine to identify the damage of each position, and the recognition accuracy is 99.68%. It is proved that this method has good accuracy and generalization in damage identification.
引文
[1]HOU Xianbin,MU Xiuwen.Research on framework of health degree and damage identification of bridge structure based on neural network and genetic algorithm[J].Revista de la Facultad de Ingenieria,2017,32(16):173-179.
[2]肖书敏,闫云聚,姜波澜.基于小波神经网络方法的桥梁结构损伤识别研究[J].应用数学和力学,2016,37(2):149-158.XIAO Shumin,YAN Yunju,JIANG Bolan.Damage identification for bridge structures based on the wavelet neural network method[J].Applied Mathematics and Mechanics,2016,27(4):684-691.
[3]李延强,张阳.改进分层遗传算法在斜拉桥主梁损伤识别中的应用[J].土木建筑与环境工程,2014,36(6):41-47.LI Yanqiang,ZHANG Yang.Application of improved hierarchic genetic algorithm to damage detection of the main girder for cable-stayed bridge[J].Journal of Civil,Architectural&Environmental Engineering,2014,36(6):41-47.
[4]GENG Xiangyi,LU Shizeng,JIANG Mingshun,et al.Anew method to identify bridge bearing damage based on radial basis function neural network[J].Earthquake and Structures,2016,11(5):841-859.
[5]XIONG Chunbao,LU Huali,ZHU Jinsong.Reality of virtual damage identification based on neural networks and vibration analysis of a damaged bridge under a moving vehicle[J].Neural Computing and Applications,2018,29(5):1331-1341.
[6]夏敬婵,段敬民,赵洪波,等.基于模态参数和支持向量机的桁架结构损伤识别研究[J].工业建筑,2011,41(7):16-20.XIA Jingchan,DUAN Jingmin,ZHAO Hongbo,et al.The truss structure damage identification research based on the modal parameters and support vector machine[J].Industrial Construction,2011,41(7):16-20.
[7]何浩祥,闫维明,彭凌云.基于支持向量机的钢筋混凝土桥梁损伤识别[J].公路交通科技,2008,25(3):65-69.HE Haoxiang,YAN Weiming,PENG Lingyun.Damage identification of reinforced concrete bridge based on support vector machine[J].Journal of Highway and Transportation Research and Development,2008,25(3):65-69.
[8]李忠献,齐怀展,朱劲松.基于模态曲率法的大跨度斜拉桥损伤识别[J].地震工程与工程振动,2007,27(4):122-126.LI Zhongxian,QI Huaizhan,ZHU Jingsong.Damage identification for long-span cable-stayed bridge based on modal curvature method[J].Journal of Earthquake Engineering and Engineering Vibration,2007,27(4):122-126.
[9]刘春城,刘佼.基于支持向量机的大跨度拱桥损伤识别方法研究[J].振动与冲击,2010,29(7):174-178.LIU Chuncheng,LIU Jiao.Damage identification of a long-span arch bridge based on support vector machine[J].Journal of Vibration and Shock,2010,29(7):174-178.
[10]宋刚.异形桥梁优化设计与损伤识别方法研究[D].长春:吉林大学,2015.SONG Gang.Research on optimization design and damage identification of irregular-shaped bridge[D].Changchun:Jilin University,2015.
[11]张燕君,王会敏,付兴虎,等.基于粒子群支持向量机的钢板损伤位置识别[J].中国激光,2017,44(10):1-7.ZHANG Yanjun,WANG Huimin,FU Xinghu,et al.Identification of steel plate damage position based on particle swarm support vector machine[J].Chinese Journal of Lasers,2017,44(10):1-7.
[12]Kourehli S S.Structural damage diagnosis using incomplete static responses and LS-SVM[J].Inverse Problems in Science and Engineering,2017,25(3):418-433.
[13]周绮凤,宁永鹏,周青青,等.一种基于支持向量机的结构损伤识别方法[J].厦门大学学报(自然科学版),2013,52(1):57-62.ZHOU Qifeng,NING Yongpeng,ZHOU Qingqing,et al.Structural damage detection based on support vector machine[J].Journal of Xiamen University(Natural Science),2013,52(1):57-62.
[2]肖书敏,闫云聚,姜波澜.基于小波神经网络方法的桥梁结构损伤识别研究[J].应用数学和力学,2016,37(2):149-158.XIAO Shumin,YAN Yunju,JIANG Bolan.Damage identification for bridge structures based on the wavelet neural network method[J].Applied Mathematics and Mechanics,2016,27(4):684-691.
[3]李延强,张阳.改进分层遗传算法在斜拉桥主梁损伤识别中的应用[J].土木建筑与环境工程,2014,36(6):41-47.LI Yanqiang,ZHANG Yang.Application of improved hierarchic genetic algorithm to damage detection of the main girder for cable-stayed bridge[J].Journal of Civil,Architectural&Environmental Engineering,2014,36(6):41-47.
[4]GENG Xiangyi,LU Shizeng,JIANG Mingshun,et al.Anew method to identify bridge bearing damage based on radial basis function neural network[J].Earthquake and Structures,2016,11(5):841-859.
[5]XIONG Chunbao,LU Huali,ZHU Jinsong.Reality of virtual damage identification based on neural networks and vibration analysis of a damaged bridge under a moving vehicle[J].Neural Computing and Applications,2018,29(5):1331-1341.
[6]夏敬婵,段敬民,赵洪波,等.基于模态参数和支持向量机的桁架结构损伤识别研究[J].工业建筑,2011,41(7):16-20.XIA Jingchan,DUAN Jingmin,ZHAO Hongbo,et al.The truss structure damage identification research based on the modal parameters and support vector machine[J].Industrial Construction,2011,41(7):16-20.
[7]何浩祥,闫维明,彭凌云.基于支持向量机的钢筋混凝土桥梁损伤识别[J].公路交通科技,2008,25(3):65-69.HE Haoxiang,YAN Weiming,PENG Lingyun.Damage identification of reinforced concrete bridge based on support vector machine[J].Journal of Highway and Transportation Research and Development,2008,25(3):65-69.
[8]李忠献,齐怀展,朱劲松.基于模态曲率法的大跨度斜拉桥损伤识别[J].地震工程与工程振动,2007,27(4):122-126.LI Zhongxian,QI Huaizhan,ZHU Jingsong.Damage identification for long-span cable-stayed bridge based on modal curvature method[J].Journal of Earthquake Engineering and Engineering Vibration,2007,27(4):122-126.
[9]刘春城,刘佼.基于支持向量机的大跨度拱桥损伤识别方法研究[J].振动与冲击,2010,29(7):174-178.LIU Chuncheng,LIU Jiao.Damage identification of a long-span arch bridge based on support vector machine[J].Journal of Vibration and Shock,2010,29(7):174-178.
[10]宋刚.异形桥梁优化设计与损伤识别方法研究[D].长春:吉林大学,2015.SONG Gang.Research on optimization design and damage identification of irregular-shaped bridge[D].Changchun:Jilin University,2015.
[11]张燕君,王会敏,付兴虎,等.基于粒子群支持向量机的钢板损伤位置识别[J].中国激光,2017,44(10):1-7.ZHANG Yanjun,WANG Huimin,FU Xinghu,et al.Identification of steel plate damage position based on particle swarm support vector machine[J].Chinese Journal of Lasers,2017,44(10):1-7.
[12]Kourehli S S.Structural damage diagnosis using incomplete static responses and LS-SVM[J].Inverse Problems in Science and Engineering,2017,25(3):418-433.
[13]周绮凤,宁永鹏,周青青,等.一种基于支持向量机的结构损伤识别方法[J].厦门大学学报(自然科学版),2013,52(1):57-62.ZHOU Qifeng,NING Yongpeng,ZHOU Qingqing,et al.Structural damage detection based on support vector machine[J].Journal of Xiamen University(Natural Science),2013,52(1):57-62.