一种基于神经网络的模态参数识别方法
|
摘要
提出一种利用神经网络结合自回归滑动平均时间序列分析来提取结构或系统模态参数的方法。这种方法首先运用神经网络计算出自回归滑动平均时间序列分析中的未知系数,然后提取出系统的模态参数。数值仿真及实验结果表明,该方法具有很高的识别精度、对噪声不敏感和具有很强的抗噪性的特性,这几点是传统时域辨识方法不具备的。文中提出的方法适合提取出实际系统的模态参数,对实际测量信号分析具有较好的参考价值。
A novel artificial neural network(ANN)-based technique employing auto regression moving average(ARMA) time series method will be used to extract structural modal parameters.It takes advantage of the output-only vibration signals to obtain the modal parameters.Firstly,the method uses ANN to calculate the unknown coefficients in ARMA form,and then,it extracts the modal parameters from the calculated matrix before.Both numerical evaluations and experimental results demonstrate that the ANN-based method identifies structural modal parameter accurately and robustly.Furthermore,it has a more robust performance than the traditional time domain method.Thus the proposed method is very meaningful to the real measured vibration signals,and it is very suitable to extract the modal parameters from the real system.
A novel artificial neural network(ANN)-based technique employing auto regression moving average(ARMA) time series method will be used to extract structural modal parameters.It takes advantage of the output-only vibration signals to obtain the modal parameters.Firstly,the method uses ANN to calculate the unknown coefficients in ARMA form,and then,it extracts the modal parameters from the calculated matrix before.Both numerical evaluations and experimental results demonstrate that the ANN-based method identifies structural modal parameter accurately and robustly.Furthermore,it has a more robust performance than the traditional time domain method.Thus the proposed method is very meaningful to the real measured vibration signals,and it is very suitable to extract the modal parameters from the real system.
引文
[1]李炜明,朱宏平,夏勇.某框架损伤识别参数的特征分析[J].振动与冲击,2008,27(3):108-112.LI Wei Ming,ZHU HongPing,XIA Yong.Character analysis of proper parameters in damage identification of frame structure[J].Journal of Vi-bration and Shock,2008,27(3):108-112(In Chinese).
[2]李忠献,杨晓明,丁阳.基于结构响应统计特征的神经网络损伤识别方法[J].工程力学,2007,24(9):1-7.LI ZhongXian,YANG XiaoMing,DING Yang.Damage identification neural network method based onstatistical property of structural responses[J].Engineering Mechanics,2007,24(9):1-7(In Chinese).
[3]罗武,赵美英,万小鹏.基于模态应变能比与神经网络的复合材料结构损伤辨识[J].机械强度,2006,28(1):146-149.LUO Wu,ZHAO MeiYing,WANXiaoPeng.Composite structural damage monitoring based onratio of modal strain energy and neural networktech-nique[J].Journal of Mechanical Strength,2006,28(1):146-149(In Chinese).
[4]孙宗光,高赞明,倪一清.基于神经网络的损伤构件及损伤程度识别[J].工程力学,2006,23(3):18-22.SUNZongGuang,GAO ZanMing,NI YiQing.Identification of damaged members and damage extent in bridge deck by neural network[J].Engi-neering Mechanics,2006,23(3):18-22(In Chinese).
[5]罗璇,程伟.利用模态编码进行结构损伤位置识别的自联想存储器神经网络方法[J].航空学报,2008,29(1):60-65.LUO Xuan,CHENG Wei.Auto-associated memory neural network of structure damage position detection using mode coding[J].Acta Aeronau-tica et Astronautica Sinica,2008,29(1):60-65(In Chinese).
[6]寇立夯,金峰.基于HHT方法的结构模态参数识别[J].水利水电科技进展,2008,28(3):45-49.KOULiHang,JIN Feng.Modal parameter identification for engineering structures based on HHTmethod[J].Advancesin Science and Technolo-gy of Water Resources,2008,28(3):45-49(In Chinese).
[7]YANGJ N,LEI Ying,PANShuwen,et al.Systemidentification of lin-ear structures based on Hilbert-Huang spectral analysis:part1normal modes[J].Earthquake Engineering Structure Dynamics,2003,32(9):1443-1467.
[8]陈隽,徐幼麟.HHT方法在结构模态参数识别中的应用[J].振动工程学报,2003,16(3):383-388.CHENJuan,XUYouLin.Application of HHTfor modal parameter iden-tificationto civil structures[J].Journal of Vibration Engineering,2003,16(3):383-388(In Chinese).
[9]罗光坤,张令弥.Morlet小波用于环境激励下的模态参数识别研究[J].地震工程与工程振动学报,2007,27(4):109-115.LUO GuangKun,ZHANG LingMi.Study on identification of modal pa-rameters under ambient excitation based on Morlet wavelet[J].Journal of Earthquake Engineering and Engineering Vibration,2007,27(4):109-115(In Chinese).
[10]柳小勤,岳林,朱如鹏.连续小波变换应用于密集模态参数识别[J].南京航空航天大学学报,2007,39(4):496-500.LIUXiaoQin,YUELin,ZHURuPeng.Continuous wavelet transformfor close modal parameter estimation[J].Journal of Nanjing University of Aeronautics and Astronautics,2007,39(4):496-500(In Chinese).
[11]Staszewski WJ.Identification of dampingin MDOFsystems using time-scale decomposition[J].Journal of Sound and Vibration,1997,203(2):283-305.
[12]李中付,华宏星,石银明,等.遗传算法在结构模态参数辨识中的应用[J].振动、测试与诊断,2001,21(4):245-248.LI ZhongFu,HUA HongXing,SHI YinMing,et al.Application of Ge-netic Algorithmin modal parameter identification for engineering struc-tures[J].Journal of Vibration,Measurement and Diagnosis,2001,21(4):245-248(In Chinese).
[13]Jiu-Bin Tan,Yan Liu,Lei Wang,et al.Identification of modal parame-ters of a systemwith high damping and closely spaced modes by combin-ing continuous wavelet transformwith patternsearch[J].Mechanical Sys-temand Signal Processing,2008,22:1055-1060.
[14]朱启兵,刘杰,李允公,等.利用Laplace小波和粒子群算法实现阻尼和频率精确计算[J].东北大学学报:自然科学版,2006,27(7):794-797.ZHUQiBing,LIUJie,LI YunGong,et al.Exact evaluation of damping andfrequency by Laplace wavelet and PSO[J].Journal of Northeastern University:Natural Science Edition,2006,27(7):794-797(In Chi-nese).
[15]欧进萍,何林,肖仪清.基于ARMA模型和自由振动提取技术的海洋平台结构参数识别[J].应用数学与力学,2003,24(4):398-404.OUJinPing,HELin,XIAOYiQing,Parameter identificationin offshore platformusing ARMA model and technology of extracting free vibration signal[J].Applied Mathematics and Mechanics,2003,24(4):398-404(In Chinese).
[16]高维成,李冀龙,邹经湘.多维ARX模型在高层建筑模态识别中的应用[J].哈尔滨工业大学学报,2003,35(12):1422-1425.GAO WeiCheng,LI JiLong,ZOUJingXiang.Action of multiple ARX model for modal identification of high-rise buildings[J].Journal of Har-bin Institute of Technology,2003,35(12):1422-1425(In Chinese).
[17]Smail M,Thomas M.ARMA models for modal analysis:effect of model orders and sampling frequency[J].Mechanical Systemand Signal Pro-cessing,1999,13(6):925-941.
[18]Jian Zhang,Tadanobu Sato,Susumu Iai,et al.Apattern recognition technique for structural identification using observed signals:Linear case studies[J].Engineering Structures,2008,30:1439-144.
[2]李忠献,杨晓明,丁阳.基于结构响应统计特征的神经网络损伤识别方法[J].工程力学,2007,24(9):1-7.LI ZhongXian,YANG XiaoMing,DING Yang.Damage identification neural network method based onstatistical property of structural responses[J].Engineering Mechanics,2007,24(9):1-7(In Chinese).
[3]罗武,赵美英,万小鹏.基于模态应变能比与神经网络的复合材料结构损伤辨识[J].机械强度,2006,28(1):146-149.LUO Wu,ZHAO MeiYing,WANXiaoPeng.Composite structural damage monitoring based onratio of modal strain energy and neural networktech-nique[J].Journal of Mechanical Strength,2006,28(1):146-149(In Chinese).
[4]孙宗光,高赞明,倪一清.基于神经网络的损伤构件及损伤程度识别[J].工程力学,2006,23(3):18-22.SUNZongGuang,GAO ZanMing,NI YiQing.Identification of damaged members and damage extent in bridge deck by neural network[J].Engi-neering Mechanics,2006,23(3):18-22(In Chinese).
[5]罗璇,程伟.利用模态编码进行结构损伤位置识别的自联想存储器神经网络方法[J].航空学报,2008,29(1):60-65.LUO Xuan,CHENG Wei.Auto-associated memory neural network of structure damage position detection using mode coding[J].Acta Aeronau-tica et Astronautica Sinica,2008,29(1):60-65(In Chinese).
[6]寇立夯,金峰.基于HHT方法的结构模态参数识别[J].水利水电科技进展,2008,28(3):45-49.KOULiHang,JIN Feng.Modal parameter identification for engineering structures based on HHTmethod[J].Advancesin Science and Technolo-gy of Water Resources,2008,28(3):45-49(In Chinese).
[7]YANGJ N,LEI Ying,PANShuwen,et al.Systemidentification of lin-ear structures based on Hilbert-Huang spectral analysis:part1normal modes[J].Earthquake Engineering Structure Dynamics,2003,32(9):1443-1467.
[8]陈隽,徐幼麟.HHT方法在结构模态参数识别中的应用[J].振动工程学报,2003,16(3):383-388.CHENJuan,XUYouLin.Application of HHTfor modal parameter iden-tificationto civil structures[J].Journal of Vibration Engineering,2003,16(3):383-388(In Chinese).
[9]罗光坤,张令弥.Morlet小波用于环境激励下的模态参数识别研究[J].地震工程与工程振动学报,2007,27(4):109-115.LUO GuangKun,ZHANG LingMi.Study on identification of modal pa-rameters under ambient excitation based on Morlet wavelet[J].Journal of Earthquake Engineering and Engineering Vibration,2007,27(4):109-115(In Chinese).
[10]柳小勤,岳林,朱如鹏.连续小波变换应用于密集模态参数识别[J].南京航空航天大学学报,2007,39(4):496-500.LIUXiaoQin,YUELin,ZHURuPeng.Continuous wavelet transformfor close modal parameter estimation[J].Journal of Nanjing University of Aeronautics and Astronautics,2007,39(4):496-500(In Chinese).
[11]Staszewski WJ.Identification of dampingin MDOFsystems using time-scale decomposition[J].Journal of Sound and Vibration,1997,203(2):283-305.
[12]李中付,华宏星,石银明,等.遗传算法在结构模态参数辨识中的应用[J].振动、测试与诊断,2001,21(4):245-248.LI ZhongFu,HUA HongXing,SHI YinMing,et al.Application of Ge-netic Algorithmin modal parameter identification for engineering struc-tures[J].Journal of Vibration,Measurement and Diagnosis,2001,21(4):245-248(In Chinese).
[13]Jiu-Bin Tan,Yan Liu,Lei Wang,et al.Identification of modal parame-ters of a systemwith high damping and closely spaced modes by combin-ing continuous wavelet transformwith patternsearch[J].Mechanical Sys-temand Signal Processing,2008,22:1055-1060.
[14]朱启兵,刘杰,李允公,等.利用Laplace小波和粒子群算法实现阻尼和频率精确计算[J].东北大学学报:自然科学版,2006,27(7):794-797.ZHUQiBing,LIUJie,LI YunGong,et al.Exact evaluation of damping andfrequency by Laplace wavelet and PSO[J].Journal of Northeastern University:Natural Science Edition,2006,27(7):794-797(In Chi-nese).
[15]欧进萍,何林,肖仪清.基于ARMA模型和自由振动提取技术的海洋平台结构参数识别[J].应用数学与力学,2003,24(4):398-404.OUJinPing,HELin,XIAOYiQing,Parameter identificationin offshore platformusing ARMA model and technology of extracting free vibration signal[J].Applied Mathematics and Mechanics,2003,24(4):398-404(In Chinese).
[16]高维成,李冀龙,邹经湘.多维ARX模型在高层建筑模态识别中的应用[J].哈尔滨工业大学学报,2003,35(12):1422-1425.GAO WeiCheng,LI JiLong,ZOUJingXiang.Action of multiple ARX model for modal identification of high-rise buildings[J].Journal of Har-bin Institute of Technology,2003,35(12):1422-1425(In Chinese).
[17]Smail M,Thomas M.ARMA models for modal analysis:effect of model orders and sampling frequency[J].Mechanical Systemand Signal Pro-cessing,1999,13(6):925-941.
[18]Jian Zhang,Tadanobu Sato,Susumu Iai,et al.Apattern recognition technique for structural identification using observed signals:Linear case studies[J].Engineering Structures,2008,30:1439-144.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心