基于离散谐小波变换的地震波时变谱估计及非平稳地震波人工合成

英文篇名：Time-varying spectrum estimation and artificial non-stationary ground motion simulation via dyadic harmonic wavelet transform
中文刊名：地震学报
英文刊名：Acta Seismologica Sinica
作者：樊剑 ; 吕超 ; 张辉
英文作者：Fan Jian1 ; 2) ; Lü Chao1)Zhang Hui1)1)College of Civil Engineering and Mechanics ; Huazhong University of Science and Technology ; Wuhan 430074 ; China2)Hubei Key Laboratory of Control Structure ; Huazhong University of Science and Technology ; Wuhan 430074 ; China
中文关键词：谐小波变换 ; 时变谱 ; 非平稳地震波
英文关键词：harmonic wavelet transform ; time-varying spectrum ; non-stationary ground motion
出版日期：2009-05-15
机构：华中科技大学土木工程与力学学院;华中科技大学控制结构湖北省重点实验室;
年：2009
期：03
出版单位：地震学报

摘要

采用在频域内具有极好紧支性和盒形特性的谐小波作为母小波,推导了基于离散谐小波变换的地震波时变谱,并给出了不同尺度下能量随时间变化的包络函数的近似解析表达式.文中以Northridge地震波为例分析了时频分布特征.研究发现,对于自振频率为f的结构,其地震反应与输入地震波的时变谱在f处的时域最大值(即maxt∈Tp[p(f,t)])以及地震波在f频率点附近的信号分量在时域内的能量集中程度有很大的关系;与db4小波基相比,利用谐小波作为母小波的小波变换,其频域具有较好的分辨率,但时域分辨率却较差.最后提出了两种基于离散谐小波逆变换的人工非平稳地震波仿真方法.
Explicit mathematical expression of time-varying spectrum of non-stationary ground motion is derived by using the harmonic wavelet transform.The choice of harmonic wavelet is made because it is defined by a simple explicit formula and they are exactly boxlike representation in frequency domain.The time-frequency characteristic of seismic wave of Northridge earthquake is analyzed through numerical examples.It is showed that the seismic response of the structure with natural frequency f is remarkably related to the time domain maximum of time-varying spectrum of input ground motion and the energy concentration degree of the signal component at the scale corresponding to the frequency.The frequency domain resolution is better and the time domain resolution is worse by using the harmonic wavelet than by using the db4 wavelet,but the explicit formula of time-frequency spectrum can be obtained by using the harmonic wavelet.Finally,the method of generating artificial non-stationary ground motion is presented based on dyadic harmonic wavelet.

引文

曹晖,赖明,白绍良.2004.地震地面运动的局部谱密度描述及其估计方法[J].世界地震工程,20(2):43--49.
    樊剑,刘铁,胡亮.2007.基于现代时频分析技术的地震波时变谱估计[J].振动与冲击,26(1):79--82.
    彭玉华.2003.小波变换与工程应用[M].北京:科学出版社:67--78.
    石春香,罗奇峰.2003.时程信号的Hilbert-Huang变换与小波分析[J].地震学报,25(4):398--405.
    Cohen A著.1995.白居宪译.1998.时频分析:理论与应用[M].西安:西安交通大学出版社:68--70.
    Deodatis G,Shinozuka M.1988.Auto-regressive model for nonstationary stochastic processes[J].J Eng Mech ASCE,114(11):1995--2012.
    Newland D E.1994.Wavelet analysis of vibration,PartⅠ:Theory[J].J Vibr Acoust,116(4):409--416.
    Spanos P D,Failla G.2004.Evolutionary spectra estimation using wavelets[J].Journal of Engineering Mechanics,130(9):952--960.
    Spanos P D,Tezcan J,Tratskas P.2005.Stochastic processes evolutionary spectrumestimation via harmonic wavelets[J].Comput Methods App Mech Engng,194(12):1367--1383.
    Spanos P D,Giaralisb A,Politis N P.2007.Time-frequency representation of earthquake accelerograms and inelasticstructural response records using the adaptive chirplet decomposition and empirical mode decomposition[J].Soil Dy-namics and Earthquake Engineering,27(6):675--689.
    Wen Y K,Gu P.2004.Description and simulation of.non-stationary processes based on Hilbert spectra[J].Journal ofEngineering Mechanics,130(9):942--951.