【摘要】 地震波在地下介质传播过程中,受近地表的非均质性和地层的非完全弹性影响产生散射和吸收衰减,使得地震记录具有非平稳性.因此对非平稳地震记录的研究需打破子波时不变的假设,更加真实地反映地震子波的传播特性,将有效提高地震子波估计精度.目前非平稳地震记录中提取地震子波的研究主要集中在两方面:对非平稳地震记录进行衰减补偿后采用常规方法提取地震子波;利用谱模拟或分段处理直接对非平稳地震记录进行时变子波提取.本文深入研究了衰减补偿方法与时变子波提取方法,对广泛应用于非平稳地震记录处理中的时频分析方法的优缺点进行归纳,并对谱模拟及分段处理方法的应用前景进行了展望。更多还原

【Abstract】 During the propagation of seismic wavelet in underground medium,scattering and absorbing are caused by near-surface heterogeneity and stratigraphic viscoelasticity,which lead to the nonstationarity of seismogram.Therefore,the study of non-stationary seismogram will more truly reflect the characteristics of seismic wave propagation,and have an important role in improving the resolution of seismogram.Currently there are mainly two types of treatment for timevarying wavelet extraction from non-stationary seismogram.On the one hand,using conventional methods to extract seismic wavelet after the attenuation compensation of non-stationary seismogram.On the other hand,using spectrum modeling or segment to extract time-varying wavelet directly from nonstationary seismogram.In this article,Attenuation compensation methods and time-varying wavelet extraction methods were described.Time-frequency analysis methods widely used in non-stationary seismogram processing were summarized.And the applications of spectrum modeling and segment processing were prospected.更多还原