基于叠前CMP道集的Q值估计方法研究

发布时间：2018-12-13 09:34

【摘要】：近年来,地震勘探从传统的构造勘探向岩性勘探转变,而在岩性油气藏勘探阶段亟需解决的问题主要有:有利储层识别、裂缝检测、岩性及含流体属性分析等。品质因子值与地下介质信息密切相关,将其用于对地震波衰减频散现象的辨识与分析,有助于提高储层描述及油藏预测精度。准确估计地震波衰减对提高地震成像分辨率、储层预测及裂缝检测的可靠性具有重要作用。因此,Q值估计问题一直是介质非弹性研究领域的重点课题之一。本文首先讨论了粘弹性介质理论和地震波吸收衰减机制及其影响因素,介绍了几种时频分析方法及其各自的适用条件。进而论述了几种常规的Q值估计方法,比较了常用的峰值频率法、质心频率偏移法、谱比法的优缺点及各自的适用条件,并研究了基于时变褶积模型的常Q模型地震波衰减模拟。Q本身是一个有用的物性参数,但是却很少被准确的测量出来。传统的基于叠后地震资料的Q值估算精度较低、适用性差;基于VSP资料进行的Q值估计准确度高,但由于VSP资料覆盖范围小,实用性不强。为了克服基于常规叠后资料和VSP资料Q值估计方法的缺陷,本文提出了一种基于叠前CMP道集进行Q值估计的方法。该方法采用时频分辨率可调的广义S变换,对经过简单预处理的叠前CMP道集进行时频分析,由于广义S变换的时窗大小可以随频率调节,减小了由于薄层引起的空间谱干扰的影响。并对其进行振幅谱畸变校正,消除非双曲动校和动校拉伸的影响。对于实际资料,在偏移距窗口内选取相邻的几道做一个部分叠加,提高了信噪比。选取合适的参考轴,由频谱比对数与频率呈线性关系即可反演得到谱比斜率,由谱比斜率随偏移距平方的线性关系估算出零偏移距处的射线平均品质因子,再由层剥离法估计层品质因子,还可得到无衰减的AVO信息。模型数据和实际资料的应用都取得了很好的效果,证明基于叠前CMP道集的Q值估计方法简单有效、准确性高,可用于含油气检测。
[Abstract]:In recent years, seismic exploration has changed from traditional structural exploration to lithologic exploration. The main problems that need to be solved in lithologic oil and gas reservoir exploration stage are: identification of favorable reservoir, fracture detection, lithology and fluid-bearing attribute analysis and so on. The value of quality factor is closely related to the information of underground medium. It can be used to identify and analyze the attenuation and dispersion phenomenon of seismic wave, which is helpful to improve the reservoir description and reservoir prediction accuracy. Accurate estimation of seismic wave attenuation plays an important role in improving the resolution of seismic imaging, reservoir prediction and reliability of fracture detection. Therefore, Q-value estimation is always one of the key topics in the field of inelastic media. In this paper, the theory of viscoelastic medium, the mechanism of seismic wave absorption and attenuation and its influencing factors are discussed, and several time-frequency analysis methods and their respective applicable conditions are introduced. Furthermore, several conventional Q estimation methods are discussed. The advantages and disadvantages of peak frequency method, centroid frequency offset method, spectral ratio method and their respective applicable conditions are compared. The seismic wave attenuation simulation of constant Q model based on time-varying convolution model is studied. Q itself is a useful physical parameter, but it is seldom measured accurately. The traditional Q value estimation based on post-stack seismic data has low accuracy and poor applicability, and the accuracy of Q value estimation based on VSP data is high, but because of the small coverage range of VSP data, its practicability is not strong. In order to overcome the defects of the method of Q value estimation based on conventional poststack data and VSP data, a method of Q value estimation based on prestack CMP gather is proposed in this paper. In this method, the time-frequency analysis of prestack CMP gathers is carried out by using generalized S-transform with adjustable time-frequency resolution. Because the window size of generalized S-transform can be adjusted with frequency, The influence of spatial spectral interference caused by thin layer is reduced. The amplitude spectrum distortion is corrected to eliminate the influence of non-hyperbolic and dynamic calibration. For the actual data, several adjacent channels in the offset window are selected for a partial stack, which improves the signal-to-noise ratio (SNR). By selecting the appropriate reference axis, the spectral slope can be obtained from the linear relationship between the frequency and the frequency, and the ray-average quality factor at the zero offset can be estimated from the linear relation between the spectral ratio slope and the square of the offset. The layer quality factor can be estimated by the layer stripping method, and the AVO information without attenuation can be obtained. The application of the model data and the actual data shows that the Q value estimation method based on prestack CMP gathers is simple, effective and accurate, and can be used for oil and gas detection.
【学位授予单位】：中国石油大学(华东)
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P631.4

【参考文献】