基于S变换的地震波时频分析及人工调整
|
摘要
首先推导了离散S变换的基本算法,定义了反映地震波时频特性的时变幅值谱和时变能量谱;以K obe地震波为例,基于S变换分析了该地震波的时频特性,并对采用不同窗口函数的分析结果进行了对比,分析发现当适当地调整窗口参数,复窗口的广义S变换对非平稳地震波有较好的时频分辨率;利用S变换的时频滤波性质,对实际地震波进行了任意时间和频率段的人工调整;以与抗震规范相容的功率谱为目标谱,提出了利用S变换合成非平稳地震波的新方法,通过实例分析发现,人工合成的地震波功率谱与目标谱符合很好,而且继承了母波的非平稳特性;还研究了地震波的时频特性对单自由度体系弹性反应谱的影响。
The formulas of discrete S-transform are derived,and the time-varying amplitude spectrum and time varying energy spectrum based on S-transform are given.The time-varying energy spectrum of Kobe wave is estimated by using S-transform with different window function,numerical results show that the improved Stransform is the effective method when choosing the time window parameter properly.Earthquake records can be adjusted with time-frequency filtering based on Stransform,and a new procedure is presented to generate an accelerograms whose power spectrum is compatible with a target spectrum by using actual earthquake records as parents.It is showed that the power spectrums of reconstructed accelerograms have good matching with the target spectrum and preserve the non-stationary features of the parents accelerograms.Finally,the effects of the non-stationary frequency characteristic of ground motions on the response spectrum are investigated.
The formulas of discrete S-transform are derived,and the time-varying amplitude spectrum and time varying energy spectrum based on S-transform are given.The time-varying energy spectrum of Kobe wave is estimated by using S-transform with different window function,numerical results show that the improved Stransform is the effective method when choosing the time window parameter properly.Earthquake records can be adjusted with time-frequency filtering based on Stransform,and a new procedure is presented to generate an accelerograms whose power spectrum is compatible with a target spectrum by using actual earthquake records as parents.It is showed that the power spectrums of reconstructed accelerograms have good matching with the target spectrum and preserve the non-stationary features of the parents accelerograms.Finally,the effects of the non-stationary frequency characteristic of ground motions on the response spectrum are investigated.
引文
[1]陈健云,李静,周晶,等.地震动频谱对小拱坝非线性响应的影响.振动工程学报[J],2003,16(2):207-211.
[2]Junjie Wang,Lichu Fan,Shie Qian,et al.Simulationsof non-stationary frequency content and its impor-tance to seismic assessment of structures[J].Earth-quake Engineering and Structural Dynamics,2002,31:993—1 005.
[3]Jun Iyama,Hitoshi Kuwamura.Application ofwavelets to analysis and simulation of earthquake mo-tions[J].Earthquake Engineering and Structural Dy-namics,1999,28:255—272.
[4]Sushovan Mukherjee,Vinay K Gupta.Wavelet-basedcharacterization of design ground motions[J].Earth-quake Engineering and Structural Dynamics,2002,31:1 173—1 190.
[5]Kurtis Gurley,Ahsan Kareem.Application of wavelettransforms in earthquake,wind,and ocean engineer-ing[J].Engineering Structures,1999,21:149—167.
[6]Spanos P D,Jale Tezcan,Petros Tratskas.Stochasticprocesses evolutionary spectrum estimation via har-monic wavelets[J].Comput.Methods Appl.Mech.Engrg.,2005,194:1 367—1 383.
[7]Spanosa P D,Giaralisb A,Politis N P.Time-fre-quency representation of earthquake accelerogramsand inelastic structural response records using theadaptive chirplet decomposition and empirical modedecomposition[J].Soil Dynamics and Earthquake En-gineering,2007,27:675—689.
[8]Wen Y K,Gu Ping.Description and simulation ofnon-stationary processes based on Hilbert spectra[J].Journal of Engineering Mechanics,2004,130:942—951.
[9]Stockwell R G.A basis for efficient representation ofthe S-transform[J].Digital Signal Processing,2007,17:371—393.
[10]Pinnegar C R.Time-frequency and time-time filteringwith the S-transform and TT-transform[J].DigitalSignal Processing,2005,15:604—620.
[11]Pinnegar C R,Mansinha L.Time-local Fourier analy-sis with a scalable,phase-modulated analyzing func-tion:the S-transform with a complex window[J].Signal Processing,2004,84:1 167—1 176.
[12]陈学华,贺振华.改进的S变换及在地震信号处理中的应用[J].数据采集与处理,2005,20(4):449—453.
[13]薛素铎,王雪生,曹资.基于新抗震规范的地震动随机模型参数研究[J].土木工程学报,2003,36(5):5—10.
[2]Junjie Wang,Lichu Fan,Shie Qian,et al.Simulationsof non-stationary frequency content and its impor-tance to seismic assessment of structures[J].Earth-quake Engineering and Structural Dynamics,2002,31:993—1 005.
[3]Jun Iyama,Hitoshi Kuwamura.Application ofwavelets to analysis and simulation of earthquake mo-tions[J].Earthquake Engineering and Structural Dy-namics,1999,28:255—272.
[4]Sushovan Mukherjee,Vinay K Gupta.Wavelet-basedcharacterization of design ground motions[J].Earth-quake Engineering and Structural Dynamics,2002,31:1 173—1 190.
[5]Kurtis Gurley,Ahsan Kareem.Application of wavelettransforms in earthquake,wind,and ocean engineer-ing[J].Engineering Structures,1999,21:149—167.
[6]Spanos P D,Jale Tezcan,Petros Tratskas.Stochasticprocesses evolutionary spectrum estimation via har-monic wavelets[J].Comput.Methods Appl.Mech.Engrg.,2005,194:1 367—1 383.
[7]Spanosa P D,Giaralisb A,Politis N P.Time-fre-quency representation of earthquake accelerogramsand inelastic structural response records using theadaptive chirplet decomposition and empirical modedecomposition[J].Soil Dynamics and Earthquake En-gineering,2007,27:675—689.
[8]Wen Y K,Gu Ping.Description and simulation ofnon-stationary processes based on Hilbert spectra[J].Journal of Engineering Mechanics,2004,130:942—951.
[9]Stockwell R G.A basis for efficient representation ofthe S-transform[J].Digital Signal Processing,2007,17:371—393.
[10]Pinnegar C R.Time-frequency and time-time filteringwith the S-transform and TT-transform[J].DigitalSignal Processing,2005,15:604—620.
[11]Pinnegar C R,Mansinha L.Time-local Fourier analy-sis with a scalable,phase-modulated analyzing func-tion:the S-transform with a complex window[J].Signal Processing,2004,84:1 167—1 176.
[12]陈学华,贺振华.改进的S变换及在地震信号处理中的应用[J].数据采集与处理,2005,20(4):449—453.
[13]薛素铎,王雪生,曹资.基于新抗震规范的地震动随机模型参数研究[J].土木工程学报,2003,36(5):5—10.
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