地震绕射波分离方法研究进展

随着油气田勘探和开发的需求,利用地震绕射波成像探测地下不规则地质体边界和小尺度构造已经成为地震偏移成像的一个热点研究方向.从勘探地球物理学家发现地震绕射波开始,学者们致力于研究如何在常规反射地震记录中分离并提取出绕射波.过去半个多世纪,关于绕射波的研究取得了显著进展并成功应用到实际生产.本文总结了以往关于绕射波分离技术的研究成果,介绍绕射波的概念及其运动学和动力学特征,按照技术方法进行总结和分类并通过选择具有代表性的实例来向读者展示不同方法的分离效果.     

局部倾角滤波和预测反演联合分离绕射波

地震绕射波源于介质非连续性,从地震记录中将绕射波分离出来并进行成像,其结果对研究诸如碳酸盐岩缝洞储层这类复杂非均质储层具有重要意义.对炮集记录进行平面波分解,在地层倾角不大的假设下,反射波和绕射波同相轴在平面波分解剖面上存在较大的倾角差异.基于此,我们提出分步进行绕射波分离的方法:(1)利用局部倾角滤波方法将绕射波的较大倾角信息成分分离出来,此时,余下的部分包含有反射波和残留的低倾角绕射波信息;(2)利用频率-空间域预测反演方法从上述含有反射波和残留的低倾角绕射波信息中分离出残留绕射波成分;(3)将两次分离的绕射波信息相加得到最终的绕射波估计.用该方法能够得到相对完整的绕射波信息,有效地克服了靠单一的倾角差异进行绕射波分离时明显损失低倾角信息,从而影响绕射波成像结果横向分辨率这一问题.理论与实际资料试算验证了该方法的有效性.     

地震绕射波弱信号U-net网络提取方法

绕射波携带大量小尺度非均匀地质体信息,对于提高地震勘探分辨率具有重要意义.绕射波能量远小于反射波,在地震记录中常被强反射波掩盖,因此分离并单独成像绕射波,为探测小尺度非均匀地质体的关键问题.传统绕射波分离方法受限于理论模型假设,对陡倾角反射波去除效果不佳,且易对绕射波造成损伤.基于经典编码-解码框架下的U-net网络和注意力机制,本文提出了一种绕射波智能分离方法,通过编码器自动提取地震数据中的绕射波特征,再由解码器恢复绕射波,从而隐性去除反射波.该方法作为端到端的机器学习,训练后的U-net网络可自适应地分离绕射波.本文通过数值模拟数据与实际数据构建训练数据集,利用训练后的U-net网络分离绕射波,并将结果偏移成像.数值模型测试和实际资料应用表明,融合了注意力机制的U-net网络能够有效压制反射波能量,保留绕射波动力学特征,克服了传统绕射波分离方法难以去除陡倾角反射的局限性,其提取的绕射波弱信号特征较为完整,能够进一步提高地震成像分辨率,在小尺度断裂刻画上具有优势.     

一种基于KPCA的绕射信息分离方法

在油气勘探的过程中,需要对叠后地震数据中的绕射波信息进行提取,从而得到地下小尺度地质体的信息.本文在前人提出的使用主成分分析法(PCA)提取绕射波信息的基础上,进一步发展使用了基于核的主成分分析技术(KPCA)的提取绕射波信息方法,即通过构建数据模型,选取核函数进行KPCA运算,并利用不同的主分量进行信号重构,从而达到将绕射波和反射波分离的目的.相对于传统PCA,这种方法对于地震数据中弯曲的同相轴,以及倾斜的界面或者倾斜的同相轴有着更好的识别能力,在提取过程中能够提供更加精细的绕射波信息.     

基于PWD的绕射波波场分离成像方法综述

绕射波成像方法能有效地对地下小尺度不连续地质体如断层、裂缝、粗糙岩丘边缘、河道等高精度成像,成像优劣的关键在于如何有效地从原始波场记录中分离反射波和绕射波波场.目前,绕射波分离及绕射波成像方法已逐步成为研究热点.本文系统地回顾了绕射波分离及成像方法的发展历程,分析了反射波和绕射波波场特征,并着重介绍了PWD(平面波分解)解构滤波器及其在叠前、叠后分离并成像绕射波的方法原理,探讨了绕射波成像方法的应用前景.     

深层碳酸盐岩岩溶风化壳溶洞型油气藏地震波场的理论研究

碳酸盐岩洞缝型油气藏是我国乃至世界的重要油气藏类型之一，属典型的复杂非均匀介质。其中，岩溶洞穴型油气藏是最容易发现的。为了解决这类储层的勘探开发及预测问题，需要从理论上对它们的地震波场进行研究。我们利用随机介质模型和非均匀介质弹性波动方程计算了不同高度和宽度的充满流体藏层的合成地震剖面。结果表明当绕射波偏移为珠子状的同相轴时即使远小于四分之一波长的溶洞在常规带宽的地震剖面上也是可分辨的，绕射波是溶洞高度和宽度的函数。我们引入一个宽度-振幅因子。可以用这个宽度-振幅因子从对无限宽溶洞计算的绕射波计算有限宽度的溶洞的绕射振幅。     

平面波预测滤波分离绕射波方法研究（英文）

常规地震资料处理通常将地震波场当作反射波处理,绕射波的存在会降低处理成果的分辨率,因此将其当做干扰噪音进行压制。然而,绕射波往往包含了非常重要的地质信息(裂缝、溶洞、断层等特殊构造),即使偏移将绕射波归位,因为绕射波能量较弱也会被反射同相轴遮盖。将反射波和绕射波进行分离并单独成像处理,将提高绕射目标构造的成像精度。根据在平面波域反射波拟线性、绕射波拟双曲的形态差异,本文利用平面波预测滤波技术对平面波炮记录进行波场分离。首先,采用平面波解构滤波器估算地震同相轴的局部倾角,利用局部倾角信息预测并提取反射波,然后从全波场中减去反射波,间接分离出绕射波,对绕射波成像可获得高分辨率的绕射目标体成像结果。通过对2D SEG盐丘模型进行波场分离与成像,分析表明:本文采用的平面波预测滤波技术解决了平面波解构滤波的极性反转问题,保持了绕射波原有的相位信息,保证了波场分离后的绕射波场保真度,成像结果表明用平面波预测滤波方法提取的绕射波进行偏移可改善非均质目标体的成像精度。     

频率域中的合成地震记录

波动理论的合成地震记录是地震资料处理和解释中的重要研究课题。 目前,制作波动理论的合成记录主要是用Trorey的克希霍夫方法,正在研究的有有限差分法和有限元素法。本文提出一种新的方法,在频率域中制作波动理论的合成记录,给出了与克希霍夫绕射波方程等价的频率域中的绕射波方程。在国产DJS-11型计算机上,应用两种绕射波方程实现了合成地震记录,并且进行了比较,认为频率域方法是有效的,可以用于波动方程偏移和地震资料解释,进一步工作有可能获得多层变速模型的波动理论记录。     

论折射法勘探中的绕射波射线轨迹

根据超声波三维地震模拟实验的结果,研究了绕射波射线轨迹的问题。对绕射波的运动学特征进行了讨论,并引用极值光程律进行了解释。实验和理论的分折指出,绕射点的位置经常按照一定的规律偏离震源--接收点联线的一侧。 本文还提出了“双曲线叠加法”,从而在实际较复杂的情况下,可以用它迅速地确定绕射点的空间位置及波的旅行时间。 文中提出绕射波的运动学特征,从中可以提供一些有关断层或绕射棱结构的情况。至于绕射波的动力学特征及机制将另文阐述。所获的结果及结论有助于物探资料的解释。     

全反射SH地震波的Goos-Hanchen效应动校正时差

利用SH地震波(偏振化方向垂直入射面的横波)在地层界面反射系数的附加相角导出了SH波Goos-Hanchen效应所引起的横向偏移和横向偏移渡越时间,给出了Goos-Hanchen效应正常时差公式,讨论了Goos-Hanchen效应对反射SH波正常时差的影响,绘出了横向偏移、横向偏移渡越时间、Goos-Hanchen效应正常时差及Goos-Hanchen效应正常时差校正量曲线.数值算例表明:对掠入射波或入射角在临界角附近的入射波,SH反射波的横向偏移、横向偏移渡越时间非常大,Goos-Hanchen效应对正常时差会产生较大的测量误差,在其他角度的入射波,横向偏移(横向偏移渡越时间)与波长(周期)为同一个数量级.横向偏移效应对SH反射波的传播走时影响是不可忽略的,因此在实际的地震资料处理中应进行横向偏移效应误差校正.     

MULTIPLE REFLECTIONS AND HEAD WAVES IN THE GULF OF SUEZ*

Inferior reflection quality in the Gulf of Suez at the target depth interval is attributable in part to surficial multiple reflections. An excellent example of the latter is observed on a typical seismic line in the northern portion of the Gulf. An increase in prominence of the multiple reflections appears associated with decreasing depth to a dipping highvelocity layer. Inversion of a second-order polynomial time-distance function, fitted to the observed refraction onset time-distance values, gives the velocity-depth function for sediments between the water bottom and a high-velocity layer. Velocities thus determined increase non-linearly with depth from a value near water velocity at the water bottom. Depths to the high-velocity layer are obtained from the associated head-wave linear time-distance function and by ray tracing in the overlying sediments. As the high-velocity layer approaches the water bottom from sub-water depths exceeding 0.6 km to a depth of 56 m, intensity of the multiple reflections increases to the extent of completely dominating individual records to a time of at least 3 s. The estimated plane-wave normal-incident reflection coefficient at the top of the high-velocity layer increases with decreasing depth to this layer, approaching 0.5 at the shallowest depth. This strong reflection coefficient further substantiates the existence of multiple reflections between the high-velocity layer and water layer. However, existence of water-layer multiples cannot be ruled out. The estimated water-bottom reflection coefficient is approximately 0.3, a substantial value. Multiple reflections of considerably less intensity are apparent where the high-velocity layer is deepest, and it is likely that such are waterlayer multiple reflections. Unfortunately, water-layer multiple reflections and multiple reflections between the water surface and high-velocity layer cannot be separated by their coincidence with time-distance (normal moveout) curves, the configuration of each visibly matching the curves equally well.     

基于倾角-偏移距域道集的绕射波成像

地震绕射波是地下非连续性地质体的地震响应,绕射波成像对地下断层、尖灭和小尺度绕射体的识别具有重要的意义.在倾角域共成像点道集中,反射波同相轴表现为一条下凸曲线,能量主要集中在菲涅耳带内,绕射波能量则比较发散.由于倾角域菲涅耳带随偏移距变化而存在差异,因此本文提出一种在倾角-偏移距域道集中精确估计菲涅耳带的方法,在各偏移距的倾角域共成像点道集中实现菲涅耳带的精确切除,从而压制反射波.在倾角-偏移距域道集中还可以分别实现绕射波增强,绕射波同相轴相位校正,因此能量弱的绕射波可以清晰地成像.在倾角域共成像点道集中,反射波同相轴的最低点对应于菲涅耳带估计所用的倾角,因此本文提出一种在倾角域共成像点道集中直接自动拾取倾角场的方法.理论与实际资料试算验证了本文绕射波成像方法的有效性.     

OPTIMUM HYPERBOLIC MOVEOUT FILTERS WITH APPLICATIONS TO SEISMIC DATA*

Optimum multichannel filters can be designed to process seismic events falling on hyperbolic moveout curves using the conventional least-squares method. Contrary to the linear moveout filters, autocorrelation and crosscorrelation functions inherent in the normal equations have to be computed numerically. However, computation times of filter coefficients are comparable to linear moveout operators. For a given source-receiver geometry and assuming straight ray-path, relative moveout of a seismic reflection event is dependent on the two way arrival time and rms velocity. Consequently, to avoid overlapping of pass and reject moveout windows, hyperbolic moveout filters have to be designed over time gates rather than for the whole record lengths. Hyperbolic and hyperbolic-linear moveout filters applied to synthetic and field seismic reflection traces show good signal-to-noise (S/N) ratio improvements. Results of some combined synthetic and field data examples are presented.     

Compositional variation and the origin of deep crustal reflections

Identifying the sources of crustal reflections is essential for deriving more geological information from deep crustal reflection profiles. Theoretical and model studies help place constraints on the role of compositional variation in producing deep crustal reflections. Analysis of laboratory-derived velocities and densities of rock types typical for the continental crust indicates that reflection coefficients are generally small, but significantly, 17% of the possible reflection coefficients have magnitudes between 0.1 and 0.2. Comparison between reflection coefficients derived from well logs and reflections observed in associated seismic profiles indicates that constructive interference associated with geological layering is at least as important as the magnitude of the reflection coefficients for producing detectable reflections. Constructive interference can increase reflection amplitude by two to three times but is limited to a relatively narrow range of layer thickness. For a typical 10–40 Hz seismic wavelet and typical crustal velocities of about 6 km/s, constructive interference occurs for layer thickness ranging between about 35 and 80 m. Layers thinner than 35 m interfere destructively. If reflections result from compositional variation, seismic models of hypothetical and observed geologic relations provide analogs for interpreting complex reflection patterns observed in deep crustal reflection profiles. Such models show reflection patterns similar to those observed in the reflection profiles. The models indicate that the reflections could originate in the complexly deformed and intruded terranes that are common in the crystalline crust and it may not be necessary to appeal to unobserved phenomena such as special lamellae or fluid-filled fractures to explain the reflections.     

各向同性薄层反射理论地震图

本文在各向同性介质假设下,计算了薄层顶、底反射P波和PS波的理论地震图.理论模拟发现薄层的反射P波与PS波是一复合波,包括层内的多次透射和反射转换波型,且具有类似单一界面反射的、脉冲式的波形特征.两类薄层反射整体振幅随着薄层厚度的降低而缩小;在地震子波主频40 Hz条件下,2 m左右厚度的薄层反射相比单一界面反射具有等同的振幅水平.1 m以下极薄层仅有弱反射甚至无反射;薄层反射复合波振幅与炮检距的关系依然成立,但不惟一;单界面反射AVO原理、方法不适用于薄层反射解释与反演.     

Diffraction separation by plane-wave prediction filtering

Seismic data processing typically deals with seismic wave reflections and neglects wave diffraction that affect the resolution. As a general rule, wave diffractions are treated as noise in seismic data processing. However, wave diffractions generally originate from geological structures, such as fractures, karst caves, and faults. The wave diffraction energy is much weaker than that of the reflections. Therefore, even if wave diffractions can be traced back to their origin, their energy is masked by that of the reflections. Separating and imaging diffractions and reflections can improve the imaging accuracy of diffractive targets. Based on the geometrical differences between reflections and diffractions on the plane-wave record; that is, reflections are quasi-linear and diffractions are quasi-hyperbolic, we use plane-wave prediction filtering to separate the wave diffractions. First, we estimate the local slope of the seismic event using planewave destruction filtering and, then, we predict and extract the wave reflections based on the local slope. Thus, we obtain the diffracted wavefield by directly subtracting the reflected wavefield from the entire wavefield. Finally, we image the diffracted wavefield and obtain high-resolution diffractive target results. 2D SEG salt model data suggest that the plane-wave prediction filtering eliminates the phase reversal in the plane-wave destruction filtering and maintains the original wavefield phase, improving the accuracy of imaging heterogeneous objects.     

全反射SH地震波的Goos-Hänchen效应动校正时差

利用SH地震波（偏振化方向垂直入射面的横波）在地层界面反射系数的附加相角导出了SH波Goos-Hänchen效应所引起的横向偏移和横向偏移渡越时间,给出了Goos-Hänchen效应正常时差公式,讨论了Goos-Hänchen效应对反射SH波正常时差的影响,绘出了横向偏移、横向偏移渡越时间、Goos-Hänchen效应正常时差及Goos-Hänchen效应正常时差校正量曲线.数值算例表明：对掠入射波或入射角在临界角附近的入射波,SH反射波的横向偏移、横向偏移渡越时间非常大,Goos-Hänchen效应对正常时差会产生较大的测量误差,在其他角度的入射波,横向偏移（横向偏移渡越时间）与波长（周期）为同一个数量级.横向偏移效应对SH反射波的传播走时影响是不可忽略的,因此在实际的地震资料处理中应进行横向偏移效应误差校正.     

3D MODELLING OF DIFFRACTIONS OBSERVED ON DEEP REFLECTION LINE DEKORP 2-S1

DEKORP 2-S is the first profile carried out in the German continental reflection seismic programme. Besides numerous reflections in the lower crust, the seismic section is characterized by strongly curved events that are interpreted as diffractions. These diffractions occur as clusters, mainly in two areas of the profile: the Dinkelsbühl and the Spessart area. This paper deals with the Dinkelsbühl diffractions where three-dimensional control is available. The control is provided by two additional profiles P-1 and Q-40 which run parallel and perpendicular to the main line, DEKORP 2-S. The type and the location of the diffractors are determined by traveltime-modelling using crustal velocity functions derived from in-line wide-angle observations. A model with inclined line diffractors provides the best fit to the data for all three profiles. Projections of these line diffractors to the surface show that they are aligned parallel to the strike direction of the Variscides. This suggests that the diffractions are associated with the suture zone between the Saxothuringian and Moldanubian geological provinces.     

Diffraction imaging in depth

High resolution imaging is of great value to an interpreter, for instance to enable identification of small scale faults, and to locate formation pinch-out positions. Standard approaches to obtain high-resolution information, such as coherency analysis and structure-oriented filters, derive attributes from stacked, migrated images. Since they are image-driven, these techniques are sensitive to artifacts due to an inadequate migration velocity; in fact the attribute derivation is not based on the physics of wave propagation. Diffracted waves on the other hand have been recognized as physically reliable carriers of high- or even super-resolution structural information. However, high-resolution information, encoded in diffractions, is generally lost during the conventional processing sequence, indeed migration kernels in current migration algorithms are biased against diffractions. We propose here methods for a diffraction-based, data-oriented approach to image resolution. We also demonstrate the different behaviour of diffractions compared to specular reflections and how this can be leveraged to assess characteristics of subsurface features. In this way a rough surface such as a fault plane or unconformity may be distinguishable on a diffraction image and not on a traditional reflection image.
We outline some characteristic properties of diffractions and diffraction imaging, and present two novel approaches to diffraction imaging in the depth domain. The first technique is based on reflection focusing in the depth domain and subsequent filtering of reflections from prestack data. The second technique modifies the migration kernel and consists of a reverse application of stationary-phase migration to suppress contributions from specular reflections to the diffraction image. Both techniques are proposed as a complement to conventional full-wave pre-stack depth migration, and both assume the existence of an accurate migration velocity.     

Quadratic normal moveouts of symmetric reflections in elastic media: A quick tutorial

The design of reflection traveltime approximations for optimal stacking and inversion has always been a subject of much interest in seismic processing. A most prominent role is played by quadratic normal moveouts, namely reflection traveltimes around zero-offset computed as second-order Taylor expansions in midpoint and offset coordinates. Quadratic normal moveouts are best employed to model symmetric reflections, for which the ray code in the downgoing direction coincides with the ray code in the upgoing direction in reverse order. Besides pure (non-converted) primaries, many multiply reflected and converted waves give rise to symmetric reflections. We show that the quadratic normal moveout of a symmetric reflection admits a natural decomposition into a midpoint term and an offset term. These, in turn, can be be formulated as the traveltimes of the one-way normal (N) and normal-incidence-point (NIP) waves, respectively. With the help of this decomposition, which is valid for propagation in isotropic and anisotropic elastic media, we are able to derive, in a simple and didactic way, a unified expression for the quadratic normal moveout of a symmetric reflection in its most general form in 3D. The obtained expression allows for a direct interpretation of its various terms and fully encompasses the effects of velocity gradients and Earth surface topography.