孔隙尺度下岩性分类及参数粗化

发布时间：2018-04-02 06:06

  本文选题：岩性分类　切入点：参数粗化　出处：《科学通报》2017年16期

【摘要】：随着孔隙尺度流动模拟研究深入,基于数字图像可对岩石性质进行模拟计算,但计算结果并不能反映宏观岩样性质,因此急需合理参数粗化方法.本文建立了孔隙尺度下处理岩性分类以及参数粗化的一整套方法.对于岩相学岩性分类,计算闵可夫斯基形态学函数值,采用K均值聚类得到水平层理,并通过空间插值和K均值聚类确定出岩石不规则层理.对于水动力学岩性分类,计算逾渗几何参数,使用有监督无监督高斯混合模型进行分类,分类结果与连通孔隙结构性质匹配良好.基于层状层理、不规则层理、水动力学岩性分类结果提出相应参数粗化公式,计算局部岩样渗透率、电导率反推整体岩样渗透率、电导率.研究结果发现基于水动力学岩性分类结果建立的参数粗化公式获得的整体岩样渗透率、电导率最为准确.
[Abstract]:With the further study of pore scale flow simulation, the rock properties can be simulated and calculated based on digital image, but the calculated results can not reflect the macroscopic rock sample properties. Therefore, reasonable parameter coarsening method is urgently needed. In this paper, a set of methods for dealing with lithologic classification and parameter coarsening at pore scale are established. For lithological classification, Minkowski morphological function is calculated. Horizontal stratification is obtained by K-means clustering, and rock irregular bedding is determined by spatial interpolation and K-means clustering. For hydrodynamic lithology classification, geometric parameters of percolation are calculated, and supervised and unsupervised Gao Si mixed model is used to classify. Based on stratified stratification, irregular bedding and hydrodynamic lithology classification, the coarsening formula of corresponding parameters is proposed to calculate the permeability of local rock samples and the conductivity back to the permeability of whole rock samples. Conductivity. The results show that the permeability of whole rock sample obtained by coarsening formula based on hydrodynamic lithology classification is the most accurate.
【作者单位】： 中国石油大学石油工程学院;School
【基金】：国家自然科学基金重点项目(51234007,51490654,51504276) 山东省自然科学基金(ZR2014EEP018)资助
【分类号】：P58;P618.13
，

本文编号：1699094