论文详情
针对济阳坳陷地区致密页岩储层微孔缝中以吸附态和游离态存在的页岩油预测问题,在测井资料的基础上,通过岩石物理分析,发展了一种陆相页岩油总有机碳含量(TOC)叠前地震反演预测方法。该方法首先对济阳坳陷已有测井数据的纵横波速度、密度、孔渗性、TOC和泥质含量等信息进行分析,研究对TOC敏感的弹性参数,根据纵、横波速度和密度的标准化弹性阻抗方程,建立TOC和弹性阻抗之间的关系,即确定性岩石物理模型,并将误差项加入这种关系中,从而获取统计性岩石物理模型,建立弹性参数与物性参数的桥梁。再进一步假设TOC先验分布服从混合高斯分布,并假设噪声也是高斯的,则TOC的后验概率密度是混合高斯分布。利用最大期望化(EM)算法计算混合高斯和高斯分布参数,解析地计算出TOC的后验概率分布,取最大后验估计(MAP)为最终反演结果。将该方法应用于济阳坳陷典型陆相页岩油工区,得到了和测井数据较为一致的TOC预测结果,为该地区页岩油地质甜点预测提供了参考依据。
A total organic carbon (TOC) prestack seismic inversion prediction method for continental shale oil was developed using well-logging data and the analysis of rock physics,to solve the prediction problem of adsorbed and free oil in tight shale reservoirs in the Jiyang depression.First,the velocity,density,porosity and permeability,TOC,and shale content from logging data in the Jiyang depression were analyzed,and the elastic parameters sensitive to TOC were studied.The relationship between TOC and elastic impedance,which is the deterministic petrophysical model,was established using the standardized elastic impedance equation.After adding an error term in this relationship,the statistical petrophysical model was obtained to establish the bridge between elastic parameters and physical parameters.Furthermore,it was assumed that the prior distribution of TOC follows a Gaussian mixture distribution and the noise is also Gaussian.The posterior probability density of TOC also exhibits a Gaussian mixture distribution.The maximum expectation (EM) algorithm was used to calculate the parameters of Gaussian mixture and Gaussian distribution,and then the posterior probability distribution of TOC was analytically calculated.The result of maximum posterior estimation was taken as the final inversion result.This method was applied to the typical continental shale oil area of the Jiyang depression and the results demonstrated that the TOC prediction results matched the logging data,providing a valuable reference for future shale oil prediction in this area.
国家自然科学基金石油化工基金联合项目(U1762103)资助。