储层物性参数叠前地震反演方法研究

发布时间：2018-06-27 21:03

本文选题：储层物性参数 + 弹性阻抗　；参考：《中国石油大学(华东)》2015年硕士论文

【摘要】：储层物性参数是储层描述的重要参数,对储层的综合评价起着至关重要的作用。因此,基于地震数据实现储层物性参数预测已成为目前广为重视的研究课题之一。本文将储层物性参数的联合反演方法研究作为目标,以岩石物理学作为指导依据,分析储层参数与地震反射特征之间存在的变化关系,研究一种利用弹性阻抗预测储层物性参数的反演方法。本文研究了一种在贝叶斯理论框架指导下,利用弹性阻抗预测储层物性参数的方法,该方法在充分利用了弹性阻抗反演的稳定性的前提下,联合应用贝叶斯理论,蒙特卡洛随机抽样技术,统计岩石物理模型,期望最大化算法等理论最终实现储层参数的预测。该方法主要利用贝叶斯理论推导出待反演的目标函数,依据对测井资料的统计分析建立物性参数的先验分布以及能够表征弹性阻抗与储层物性参数关系的统计岩石物理模型,并通过蒙特卡洛随机抽样技术获取弹性阻抗与物性参数的联合分布,最后通过期望最大化算法求取后验概率分布中最大位置处所对应的物性参数值,即为最终反演结果。在物性参数反演的过程中,充分利用了弹性阻抗反演的稳定性,达到稳定反演的目的,提高了反演的精度,但是由于弹性阻抗作为联系地震数据和物性参数的中间参数,仍属于间接反演,并非从地震数据直接反演物性参数,因此在反演过程中依然有累计误差的存在。将该反演方法进行实际资料分析测试,其测试效果较为理想,进而将其应用于实际工区中的碎屑岩储层中,进行储层预测,也同样得到了较为理想的预测结果,证明了此反演方法具有实用价值。
[Abstract]:Reservoir physical parameter is an important parameter in reservoir description and plays an important role in comprehensive evaluation of reservoir. Therefore, the prediction of reservoir physical parameters based on seismic data has become one of the most important research topics. In this paper, the study of joint inversion of reservoir physical parameters is taken as the objective, and rock physics is taken as the guide to analyze the relationship between reservoir parameters and seismic reflection characteristics. An inversion method for predicting reservoir physical parameters using elastic impedance is studied. In this paper, a method of predicting reservoir physical parameters with elastic impedance under the guidance of Bayesian theory framework is studied. Under the premise of making full use of the stability of elastic impedance inversion, the Bayesian theory is applied jointly. Monte Carlo random sampling technique, statistical rock physical model and expectation maximization algorithm are used to predict reservoir parameters. This method mainly uses Bayesian theory to deduce the objective function to be retrieved. Based on the statistical analysis of logging data, the prior distribution of physical parameters and the statistical petrophysical model which can characterize the relationship between elastic impedance and reservoir physical parameters are established. The joint distribution of elastic impedance and physical parameters is obtained by Monte Carlo random sampling technique. Finally, the physical parameters corresponding to the maximum position in the posterior probability distribution are obtained by the expectation maximization algorithm, which is the final inversion result. In the process of physical parameter inversion, the stability of elastic impedance inversion is fully utilized to achieve the purpose of stable inversion, and the precision of inversion is improved. However, the elastic impedance is used as the intermediate parameter to relate seismic data and physical parameters. It is still an indirect inversion, not directly inversion of physical parameters from seismic data, so there are still accumulative errors in the inversion process. The inversion method is used to analyze and test the actual data, and the result of the test is satisfactory, and then the inversion method is applied to the clastic rock reservoir in the actual working area and the reservoir prediction is carried out, and the more ideal prediction results are also obtained. It is proved that this inversion method is of practical value.
【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

【相似文献】