基于地震波形反演的砂泥岩薄互层结构表征研究

发布时间：2018-05-14 22:28

本文选题：薄互层 + 地震反演　；参考：《吉林大学》2017年博士论文

【摘要】：地球物理勘探方法在地震勘探以及开发过程中的应用至关重要,对于我国油气资源的可持续发展和国民经济建设也是不可或缺的。如今的勘探目标逐渐聚焦在地质情况复杂和岩性复杂的油气藏上(张晶,2013)。随着石油勘探开发工作的不断深入以及勘探目标的日趋复杂化,对地下储层的预测精度的要求也在逐渐变高。我国石油勘探的目标以才油气田为主。新生代含油气盆地大多以薄砂层、泥岩沉积为主,含有少量的薄层碳酸盐岩、页岩及膏盐层,地层岩性和厚度横向变化较大,有很多薄互砂泥岩间互形成的多旋回多油层层状储层(蔡希玲,2007)。近年来,在本文研究区,开发逐步进入中后期,石油勘探工作己经进入到了精细勘探阶段,工作的关键在于确定薄储层的空间展布规律及其性质。薄互层指的是不同岩性的薄层交互出现的一种地质现象,相邻两层的岩性不同,比如砂泥岩薄互层。对于砂泥岩薄互层相对较发育的地区,由于薄层之间地震波的传播效应和调谐干涉作用以及储层在在横向上的多变性,使得地震波的振幅、频率和相位都受到不同程度的影响,这和薄互层组中单层厚度的变化、层速度的变化以及密度等因素的变化有着一定的关系,导致地震波的波场特征随之变得更加的复杂化。目前地震勘探中,对薄层甚至薄互层的研究占有越来越重要的地位。为了充分了解薄互层结构的地震响应,本文开展了Zoeppritz方程,Aki-Richards近似方程,Shuey二项近似方程、Shuey三项近似方程和传播矩阵地震响应特征的研究。首先从五种正演算法的理论入手,分析了Zoeppritz方程及其近似式和传播矩阵算法本质的区别:基于单界面反射的精确的Zoeppritz方程仅考虑界面的反射效应;虽然几种近似式计算的侧重点都有不同,但是由于算法理论上的限制,地震合成记录都不能很好的还原薄互层结构;广义传播矩阵理论属于反射率方法,基于界面连续性条件,在频率域计算每个频率的反射系数,能够充分考虑地震波反射、透射过程中的动力学特征,与其它方法相比更适于薄互层地震响应的计算。文章用两组不同岩性和结构的薄互层储层对以上几种正演方法进行对比分析,来验证以上结论,为之后的反演算法提供正演模拟。为了充分了解研究区的岩石物理关系,本文预测了研究区4口井砂、泥岩横波速度;建立了研究区4口井中砂、泥岩孔隙度与纵波阻抗、横波阻抗、纵横波速度岩石物理模版;研究发现,传统线性或非线性拟合的方式,不适用于研究区测井数据。为解决这一问题,给出了用于分析的统计岩石物理模板,与传统方法相比,能够较好的解释砂泥岩岩石物理关系。为之后正演模拟和反演算法提供先验信息。本文在提出针对薄互层的反演算法前,应用现阶段已经比较成熟的广义S变换进行了薄互层理论模型、实际数据时频分析。计算结果表明,时频分析能够反映薄互层反射特征,但由于分辨率的限制,在薄互层精细结构直接预测方面存在困难;研究了频变AVO算法,频散属性DP和DG能够预测与流体类型有关性质,但是由于时频分析分辨率的限制,算法不适用于薄互层结构识别;根据以上分析,要解决该薄互层结构研究区的问题,将进一步开发基于全频带地震信息的薄互层地震反演技术,并用于实际地震数据,验证方法的有效性根据以上的分析研究,本课题针对研究区致密砂泥岩薄互层地震识别与描述问题,首先分析研究区致密油储层测井数据,建立标志层以下薄互层中砂、泥岩孔隙度与弹性参数的岩石物理关系;之后,提取并设计研究区典型薄互层地质模型,通过传播矩阵理论计算典型砂泥岩薄互层结构的频变AVO地震响应特征;同时,开发测井尺度高精度合成地震记录算法;最后,开发基于波形对比的反演技术,反演标志层以下砂泥岩薄互层结构,所得结果可用于有利储层预测。
[Abstract]:The application of geophysical exploration method is very important in the process of seismic exploration and development. It is also indispensable for the sustainable development of oil and gas resources in China and the construction of national economy. The exploration targets are gradually focused on the oil and gas reservoirs with complex geological conditions and complex lithology (Zhang Jing, 2013). The requirements for the prediction accuracy of the underground reservoir are also becoming higher and higher. The target of oil exploration in China is mainly oil and gas fields. Most of the Cenozoic Oil and gas basins are mainly thin sand beds and mudstone deposits, with a small amount of thin layer carbonate rocks, shale and gypsum salt layers, stratigraphic lithology and thickness transverse. In recent years, the development of petroleum exploration has entered the stage of fine exploration, and the key of the work is to determine the spatial distribution law and its properties of thin reservoirs. Thin interbeds refer to these in recent years. A geological phenomenon that interacts with thin layers of different lithologies, with different lithologies in two adjacent layers, such as thin sandstone and mudstone interbeds. For areas with relatively developed sand mudstone thin interbeds, the amplitude, frequency and phase of seismic waves are caused by the propagation and tuning interference of the thin layers of seismic waves and the transversal variability of the reservoir. It is affected by different degrees, which has a certain relationship with the change of single layer thickness, the change of layer velocity and the change of density in the thin interlayer group, which leads to the more complicated characteristics of the wave field of seismic waves. In order to fully understand the seismic response of thin interbedded structures, the Zoeppritz equation, Aki-Richards approximation equation, Shuey two term approximate equations, Shuey three approximation equations and the seismic response characteristics of propagation matrix are studied in this paper. First, the Zoeppritz equation and its approximation and propagation matrix algorithm are analyzed from the theory of five forward algorithms. The qualitative difference: the exact Zoeppritz equation based on the single interface reflection only considers the reflection effect of the interface; although the emphasis of several approximate calculations is different, the seismic synthesis records can not reduce the thin interbed structure well because of the limitation of the algorithm, and the generalized propagation matrix theory belongs to the reflectivity method based on the interface. The reflection coefficient of each frequency in the frequency domain is calculated in the frequency domain, which can fully consider the seismic wave reflection and the dynamic characteristics in the transmission process. Compared with the other methods, it is more suitable for the calculation of the seismic response of thin interbed. This paper compares and analyses the above several forward methods with two groups of thin interbedded reservoirs with different lithology and structure. In order to fully understand the rock physical relations in the study area, this paper predicted 4 wells in the study area and the shear wave velocity of mudstone in the study area, and established the sand in 4 wells in the study area, the porosity of the mudstone and the P-wave impedance, the shear wave impedance and the longitudinal shear wave velocity. In order to solve this problem, a statistical rock physical template is given to solve this problem. Compared with the traditional method, it can better explain the physical relationship between sand and mudstone, and provide a priori information for the forward modeling and inversion algorithm. Before the algorithm, the thin interbed theory model and real time frequency analysis are carried out with the more mature generalized S transform at the present stage. The results show that the time frequency analysis can reflect the reflection characteristics of thin interbed, but because of the limitation of the resolution, there are difficulties in the direct prediction of thin interbed fine structure. The frequency variant AVO algorithm and the frequency dispersion are studied. DP and DG can predict the nature of fluid type, but due to the limitation of the resolution of time frequency analysis, the algorithm is not suitable for thin interbed structure recognition. The validity of the method is based on the analysis of the above analysis. In view of the problem of seismic identification and description of thin mudstone thin interbeds in the research area, this paper first analyzes the log data of the dense oil reservoir in the study area, establishes the rock physical relationship between the sand in the thin interbeds below the mark layer, the porosity of the mudstone and the number of elastic parameters, and then extracts and designs the research. The typical thin interbed geological model is used to calculate the frequency variant AVO seismic response characteristics of the typical sand mudstone thin interbedded structure through the propagation matrix theory. At the same time, the high precision synthetic seismic record algorithm is developed by developing the well logging scale. Finally, the inversion technique based on the waveform contrast is developed to retrieve the thin interbed structure of sand and mudstone below the mark layer, and the results can be used for the results. Reservior prediction.

【学位授予单位】：吉林大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：P631.4

【相似文献】