TTI介质准纵波逆时偏移方法研究

发布时间：2018-05-20 10:21

本文选题：TTI介质 + 逆时偏移　；参考：《中国石油大学(华东)》2015年博士论文

【摘要】：随着油气勘探与开发难度的加深,提高复杂地质条件如岩下、断层、裂缝等介质的成像精度和分辨率等需求也越来越严峻。充分挖掘并利用地震各向异性等信息,对速度建模、成像及反演等方法的扩展和改进都具有重要价值。横各向同性(TI)介质是最常用的一种各向异性模型,它包括对称轴垂直(VTI)和倾斜(TTI)两种情况。各向异性介质弹性波方程可准确描述TI介质中波的传播特征,但对于采用双程波方程准确成像的逆时偏移应用,它不仅需要庞大的计算代价,还要求提供横波速度模型等条件,均不利于对其展开广泛应用。因此研究各向异性介质(特别是TTI介质)中单独利用准纵波进行地震成像处理具有重要意义。TTI介质的准纵波方程有两种形式,一种是耦合方程,另一种是纯波方程。本文通过运用Alkhalifah的拟声学近似,假设沿对称轴方向的横波速度为零,对两种方程的推导和数值实现进行研究。耦合方程推导有两种方式,一种是以准确的TI介质频散关系为基础,另一种则直接由弹性运动方程出发,两种不同的推导可以得到形式上相似的二阶耦合方程,它们具有足够高的精度以描述准纵波运动学特征,二阶耦合方程可在时空域中运用有限差分数值实现。然而,拟声学近似只是将对称轴方向上的横波速度值设为零,在非对称轴方向上仍存在剩余横波,剩余横波伴随有效信号参与反射,在准纵波正演模拟中构成噪音污染,并且随着数值模拟的时间推进,在介质物性变化剧烈区域易引起数值不稳定。TTI介质准纵波逆时偏移另一个研究趋势是解耦的纯准纵波方程,这是因为基于混合域时间延拓的解耦方程不仅从根本上消除了横波剩余干扰,同时还具有在长时间模拟中无频散、稳定传播的优势,它彻底不存在耦合方程面临的两个难题——横波剩余和数值不稳定,并且同样基于拟声学近似的纯波方程充分满足对精度的需求。纯波方程的基本原理是将TI介质的频散关系形式上解耦,分别得到准纵波与横波方程,然后利用混合空间波数域的数值方法求解准纵波纯波方程,实现混合域时间延拓。本文研究的准纵波方程皆以逆时偏移(RTM)的应用为目的,RTM计算包括三个步骤——正向波场延拓、逆时波场延拓,以及成像条件的运用。为得到精度和分辨率更高的偏移结果,需要处理RTM宽角成像能力引起的浅层低频噪音,本文应用成像后滤波压制偏移假象,此外还运用多组模型测试对准纵波逆时偏移的几个问题进行对比研究,包括TTI偏移算法与各向同性和VTI偏移的比较,耦合方程与纯波方程逆时偏移结果的对比分析等,通过一系列模型试算得出几点认识。最后,作为对拟声学近似方程的补充,本研究对TTI介质全弹性波动方程RTM同样给予了充分的研究和介绍。与准纵波方程的标量波场相比,弹性波方程的矢量波场存在纵横波的耦合与转换等复杂波场现象,传统的弹性波成像条件无法对纵横波准确成像,而基于纵横波分解的成像条件则能够克服此问题,得到物理意义明确的偏移结果,有利于在后续地震资料处理和解释中的应用。TTI介质中纵横波场分解不能采用各向同性介质中求取散度和旋度的直接方法,构建与局部介质参数相关的分离算子可在非均匀TTI介质中将纵横波场准确的进行分离,将其应用于TTI介质弹性波RTM对提高成像的准确度有较大作用。通过比较同一 TTI介质地质模型参数下的准纵波RTM和弹性波RTM的纵波偏移结果,可对准纵波方法精度等进行初步的评价。本文针对弹性波TTI介质逆时偏移面临的问题,提出了基于耦合方程和纯波方程的准纵波逆时偏移方法,准纵波逆时偏移可以极大的提高计算效率,模型试算表明本文方法对非均匀各向异性介质成像问题的解决具有较好的适用性和有效性,能为复杂地震勘探提供较高分辨率和信噪比的偏移剖面。目前研究尚存在一些需要完善的内容,对此相关的许多重要问题仍需做进一步深入和广泛的研究。
[Abstract]:With the deepening of the difficulty of oil and gas exploration and development, the need for improving the imaging precision and resolution of the medium, such as rock, fault and crack, is becoming more and more serious. It is of great value to fully excavate and utilize seismic anisotropy, which is of great value to the expansion and improvement of velocity modeling, imaging and inversion. TI) medium is one of the most commonly used anisotropic models, which include two cases of symmetrical axis vertical (VTI) and tilt (TTI). The elastic wave equation of anisotropic medium can accurately describe the propagation characteristics of waves in the TI medium, but it requires not only a huge calculation cost but also a large calculation cost for the accurate imaging of a double wave equation. The transverse wave velocity model and other conditions are not conducive to the extensive application of it. Therefore, it is important to study the quasi longitudinal wave in the anisotropic medium (especially TTI medium) using quasi longitudinal wave for seismic imaging processing. There are two forms of quasi longitudinal wave equation of.TTI medium. One is the coupling equation and the other is the pure wave equation. This paper uses Alkhalifa in this paper. H's onomatopoeia approximation, assuming that the velocity of the transverse wave along the symmetrical axis is zero, studies the derivation and numerical realization of the two equations. There are two ways to deduce the coupling equation. One is based on the accurate dispersion relation of the TI medium, the other is directly derived from the elastic motion equation, and the two different derivation can be obtained in the form similarity. The two order coupling equations have high precision to describe the kinematic characteristics of quasi longitudinal waves. The two order coupling equations can be realized by finite difference numerical values in the spatio-temporal domain. However, the quasi acoustic approximation only approximated the transverse wave velocity in the direction of the symmetric axis to zero, and the residual transverse waves still exist in the asymmetric axis, and the residual transverse waves are accompanied by the wave. The effect signal participates in the reflection and constitutes noise pollution in the quasi longitudinal wave forward modeling, and with the time of the numerical simulation, the other research trend of the inverse time migration of the quasi longitudinal wave of the numerical unstable.TTI medium is a decoupled pure quasi longitudinal wave equation. This is due to the decoupling square based on the time extension of the mixed domain. It not only eliminates the residual interference of the transverse wave radically, but also has the advantages of no dispersion and stable propagation in the long time simulation. It does not exist two difficult problems of the coupling equation - the residual of the transverse wave and the numerical instability, and the same pure wave equation based on the approximation of the onomatopoeia fully satisfies the demand for the precision. The basic principle is to decouple the dispersion relation of TI medium, obtain the quasi longitudinal wave and the transverse wave equation respectively, and then use the numerical method of the mixed space wave number domain to solve the quasi longitudinal wave pure wave equation and realize the mixed domain time extension. The quasi longitudinal wave equation studied in this paper is aimed at the application of the inverse time migration (RTM), and the RTM calculation includes three steps. The forward wave field extension, the inverse time wave field extension, and the application of imaging conditions. In order to obtain higher precision and resolution, we need to deal with the low frequency noise caused by the RTM wide angle imaging capability. In this paper, the post imaging filtering is used to suppress the offset false image. In addition, several groups of models are used to test several questions on the inverse time migration of longitudinal wave. The comparison study includes the comparison between the TTI migration algorithm and the isotropy and the VTI migration, the contrast analysis of the inverse time migration result of the coupling equation and the pure wave equation and so on. The results are obtained by a series of models. Finally, as a supplement to the quasi acoustic approximation equation, the research is also given to the full elastic wave equation RTM of the TTI medium. Compared with the scalar wave field of the quasi longitudinal wave equation, the vector wave field of the elastic wave equation has complex wave fields such as the coupling and conversion of the longitudinal and horizontal waves. The traditional elastic wave imaging conditions can not accurately imaging the longitudinal and transverse waves, while the imaging strip based on the vertical and horizontal wave decomposition can overcome this problem and get the physical meaning clear. The result of migration is beneficial to the application of the.TTI medium in the subsequent seismic data processing and interpretation. The vertical and horizontal wave field decomposition can not be used to obtain the divergence and curl in the isotropic medium. The separation operator related to the local medium parameters can be used to separate the vertical and horizontal wave field in the non-uniform TTI medium and apply it. The elastic wave RTM in the TTI medium has a great effect on improving the accuracy of the imaging. By comparing the longitudinal wave RTM of the same TTI medium geological model and the longitudinal wave migration of the elastic wave RTM, the accuracy of the align P-wave method is preliminarily evaluated. In this paper, the problems facing the inverse time migration of the elastic wave TTI medium are presented, and the coupling is proposed. The inverse time migration of quasi longitudinal wave in the equation and the pure wave equation and the inverse time migration of quasi longitudinal wave can greatly improve the computational efficiency. The model trial calculation shows that the proposed method has good applicability and effectiveness for the solution of the nonuniform anisotropic media imaging problem, and can provide a high resolution and signal to noise ratio offset section for complex seismic exploration. At present, there are still some problems that need to be improved. Many important issues still need further in-depth and extensive research.
【学位授予单位】：中国石油大学(华东)
【学位级别】：博士
【学位授予年份】：2015
【分类号】：P631.4;P618.13

【相似文献】