特征波反演成像理论框架

发布时间：2018-01-24 01:32

本文关键词： 反演成像特征波场非线性问题凸问题数据域反演像域反演　出处：《石油物探》2017年01期 　论文类型：期刊论文

【摘要】：地震波反演成像的核心问题是将解一个非线性(较)强的反问题转化为提一个更凸的反问题并进行求解。存在强非线性性的主要原因是实测数据与要反演的模型参数之间的关系远非线性,其次是由于包含模型参数的控制方程不能很好地预测实际数据。因此,提出了特征波反演(characteristic wave inversion,CWI)成像的理论框架,基本思想是:不追求对实测波场全部波现象的模拟,而是模拟其中的部分特征波场;不一定追求波形逼近,但要尽可能利用到达时(相位)的逼近。相对于全波形反演(full waveform inversion,FWI),特征波反演由四个基本步骤组成:1特征波场(characteristic wave field,CWF)的提取;2波动理论的透射波层析成像;3波动理论的一次反射波层析成像;4最小二乘叠前深度偏移成像。特征波场提取是其中重要的环节,包含三重含义:1波现象的分解(譬如,矢量波分解成标量波以及一次波和多次波分解);2时空局部的、单震相的、带方向的带限波场的分解;3同相轴上地震子波的分解(譬如,提取子波的达到时、相位等)。特征波场提取基于压缩感知的框架进行,其它三个线性化的参数反演环节,首先考虑的是针对地下介质参数层状分布时的反射波反演成像,然后再考虑针对散射和绕射波的反演成像。数据域特征波反演在估计低波数速度信息时仅依赖同相轴上子波的到达时或/和相位信息,需尽量排除振幅对到达时和相位估计的影响。像域中的背景速度反演仅适宜基于到达时的反演,基于像的幅值的反演在理论上是不合理的。高波数参数估计时,首先进行方位角度反射系数的估计,在此基础上进行散射强度的估计。CWI技术系列是推进经典FWI走向实用化的正确途径,初步数值试验结果证明了上述判断。
[Abstract]:The core problem of seismic wave inversion imaging is to solve a nonlinear problem. The strong inverse problem is transformed into a more convex inverse problem and solved. The main reason for the strong nonlinearity is that the relationship between the measured data and the parameters of the model to be retrieved is far nonlinear. Secondly, because the governing equation including the model parameters can not predict the actual data well, a characteristic wave inversion is proposed. The basic idea of CWI imaging is: not to simulate all the wave phenomena in the measured wave field, but to simulate some of the characteristic wave fields; It is not necessary to pursue waveform approximation, but it is necessary to make use of the time of arrival (phase) approximation as much as possible, as opposed to full waveform inversion FWI. The characteristic wave inversion consists of four basic steps: 1. Characteristic wave field CWFs extraction. 2the transmission wave tomography of wave theory; 3Primary reflection wave tomography of wave theory; (4) least square prestack depth migration imaging. Feature wave field extraction is an important part of this process, which includes decomposition of the phenomenon of triple meaning: 1 wave (e.g., vector wave decomposition into scalar wave, primary wave and multiple wave decomposition); (2) decomposition of band-limited wave fields in local space-time, single-shock phase and band direction; (3) decomposition of seismic wavelet on cophase axis (for example, when extracting wavelet, phase, etc.) characteristic wave field extraction is based on compression-sensing frame, and the other three linearized parameter inversion links. The first consideration is the reflection wave inversion imaging for the layered distribution of underground media parameters. Then the inversion imaging of scattering and diffraction waves is considered. The data domain characteristic wave inversion only depends on the arrival or / and phase information of wavelet in the same phase axis when estimating the velocity information of low wave number. The background velocity inversion in the image domain is only suitable for the time-of-arrival inversion. The inversion based on the amplitude of the image is not reasonable in theory. Based on the estimation of azimuth angle reflection coefficient, the estimation of scattering intensity. CWI technology series is the correct way to promote the practical application of classical FWI. The preliminary numerical results prove the above judgment.
【作者单位】：波现象与反演成像研究组(WPI)同济大学海洋与地球科学学院;
【基金】：国家自然科学基金项目(41374117) 国家油气重大专项(2011ZX05005-005-008HZ,2011ZX05006-002,2011ZX05023)联合资助~~
【分类号】：P631.4
【正文快照】： extraction.Wave equation based transmitted wave tomography is based on the first arrival and early arrival traveltime.The inver-sion imaging based on the disturbance wave field firstly deal with the reflection wave from the layered strata,and then the sc

【相似文献】