摘要
储层物性参数是储层描述的重要参数,常规的基于贝叶斯理论的储层物性参数反演方法大多是通过反演获得的弹性参数进一步转换而获得物性参数,本文提出一种基于弹性阻抗数据预测储层物性参数的反演方法.该方法主要通过建立可以表征弹性阻抗与储层物性参数之间关系的统计岩石物理模型,联合蒙特卡罗仿真模拟技术,在贝叶斯理论框架的指导下,应用期望最大化算法估计物性参数的后验概率分布,最终实现储层物性参数反演.经过模型测试和实际资料的处理,其结果表明本文提出的方法具有预测精度高,稳定性强,横向连续性好等优点.
Reservoir parameters are important to petrophysical parameter inversion method based of elastic parameters and their transform. This reservoir description. The conventional reservoir on Bayesian theory are mostly based on inversion study proposes an inversion method of reservoir petrophysical parameters based on elastic impedance data. This method establishes the statistical rock physics model to characterize the relationship between elastic impedance and reservoir petrophysical parameters, combines the Monte Carlo simulation techniques and applies the expectation maximization algorithm to estimate petrophysical parameter posterior probability distribution in the framework of Bayesian theory, and then ultimately finishes reservoir parameter inversion. The feasibility of the method is verified by the model test and actual seismic data. The results indicate that the proposed method can improve the prediction accuracy and stability, meanwhile has a better lateral continuity.
出处
《地球物理学报》
SCIE
EI
CAS
CSCD
北大核心
2014年第12期4132-4140,共9页
Chinese Journal of Geophysics
基金
国家重点基础研究发展计划(973计划)项目(2013CB228604)
国家油气重大专项(2011ZX05030-004-002)
中国博士后科学基金
山东省博士后创新基金
中石化重点实验室基金联合资助
关键词
弹性阻抗
储层物性参数
贝叶斯理论
EM算法
统计性岩石物理模型
Elastic impedance
Reservoir petrophysical Statistical rock physics model parameters
Bayesian theory
EM algorithm
