基于高分辨率拉东谱的逆时偏移角度域共成像点道集提取

基于拉东谱的分析提出了一种获得高分辨率拉东谱的方法,在付出很小计算代价的同时使得原始拉东谱能量收敛到中心点处,大幅提高了谱分辨率。利用改进的高分辨率拉东谱获得地下反射界面各点法线方向概率分布,并基于该概率分布提出一种获得逆时偏移角度域共成像点道集的方法。该方法通过炮点入射波矢量方向p s和反射界面法线方向p n计算入射角度并按照概率分布进行能量投影,获得角度域共成像点道集,实现了对于有反射界面区域其能量按照入射角度正确投影,进行相干加强;对于无反射界面区域其能量随机分布,相互抵消。解决了传统的基于炮点入射波矢量方向p s、反射界面法线方向p n计算地震波入射角度时在无反射界面区域能量投影不准确带来的噪声问题。采用带有一个高速岩体的二维鲸模型进行数值实验,结果表明:本文方法有效提高了拉东谱的分辨率,利用地下反射界面法线方向概率分布生成的逆时偏移角度域共成像点道集,有效压制了道集中无反射界面区域的噪声,为后续AVO、AVA分析提供了有效工具。

Extraction of RTM common image gather in angle domain based on high-resolution Radon spectrum

A method for obtaining a high-resolution Radon spectrum was proposed.This method converges the original Radon spectral energy to a central point with small computational cost,which could improve the spectral resolution.Through the improved high-resolution Radon spectrum,the probability distribution of the normal direction at the reflection interface was obtained.Then,a method for obtaining the common-image-gather in the angle domain from reverse time migration(RTM)analysis was proposed.This method calculates the incident angle from the direction of the vector of the incident wave at the shot point and the normal direction of the reflection interface,and performs the energy projection according to the probability distribution,so as to obtain the common-image-gather in the angle domain.In this method,the coherent enhancement was realized by correctly projecting the energy of the reflected interface region according to the incident angle.On the contrary,in non-reflective interface regions,the energy was randomly distributed and canceled each other out.The proposed method solves the noise problem caused by inaccurate energy projection in the non-reflective interface region found in traditional methods.The numerical experiments showed that the proposed method improved the resolution of the Radon spectrum.Further,the RTM common-image-gather in the angle domain obtained by this method suppressed the noise of the non-reflective interface region,thereby providing a powerful tool for AVO and AVA analyses.