人造砂岩样品多尺度微观结构表征和流体流动模拟  

Multi-scale Microstructure Characterization and Fluid Flow Simulation for an Artificial Sandstone Sample

作　　者：王晋 王海鹏[2,3,4] 刘锐[2,3] 穆欣盛 李建莉 李琨杰 杨玉双 WANG Jinxiu;WNAG Haipeng;LIU Rui;MU Xinsheng;LI Jianli;LI Kunjie;YANG Yushuang(Institute of Theoretical Physics,Shanxi University,Taiyuan 030006,China;College of Physics and Electronics Engineering,Shanxi University,Taiyuan 030006,China;Institute of Coal-derived Gas Reservoir Physical Properties and Resource Development,Shanxi University,Taiyuan 030006,China;Institute of Advanced Functional Materials and Devices,Shanxi University,Taiyuan 030006,China;Shanxi Huaxin Gas Energy Research Institute Co.LTD,Taiyuan 030032,China;Commonwealth Scientific and Industrial Research Organisation,VIC 3168,Australia)

机构地区：[1]山西大学理论物理研究所,山西太原030006 [2]山西大学物理电子工程学院,山西太原030006 [3]山西大学煤成气储层物性与资源开发研究院,山西太原030006 [4]山西大学先进功能材料与器件研究院,山西太原030006 [5]山西华新气体能源研究院有限公司,山西太原030032 [6]澳大利亚联邦科学与工业研究组织,澳大利亚维多利亚州墨尔本3168

出　　处：《山西大学学报（自然科学版）》2023年第6期1378-1385,共8页Journal of Shanxi University(Natural Science Edition)

基　　金：山西省揭榜挂帅项目(202101030201022)。

摘　　要：油气储层砂岩中的微观孔隙结构对其中流体的储集和运移具有重要影响。本文对一人造砂岩样品进行了微焦点X射线CT成像。分别采用最小二乘图像阈值分割法和数据约束模型(Data-constrained modeling,DCM)计算得到了样品组分三维分布。基于两种算法生成的样品微观结构用部分渗透格子玻尔兹曼法模拟了水在样品中的单相流动,分析了样品中孔隙分布与流体流速分布之间的关系。研究结果显示,最小二乘图像阈值分割算法可以有效提取人造砂岩样品中石英颗粒之间的大孔,但是会丢失小于CT分辨率尺度的孔隙信息。DCM模型的计算结果不仅包含了石英颗粒之间的大孔,同时包含了样品中通常与碳酸钙共存的小尺度孔隙。流动模拟的数值结果表明本文所研究样品中小于CT体元尺寸的孔隙对样品渗透率具有显著的贡献,流体进入端面处的孔隙分布对流体流速分布影响较大,流体流速分布与孔隙率分布不成正比。Microscopic porosity distributions in oil and gas reservoir sandstone have significant impacts on fluid storage and migration.In this paper,an artificial sandstone sample was imaged with X-ray computed tomography(X-ray CT).Three-dimensional microscopic distributions of porosity and mineral phases were obtained using the least-squares image threshold segmentation and the data-constrained modeling(DCM),respectively.Based on the microstructures generated with these two algorithms,single-phase flow of water in the sample was simulated using a partially percolating lattice Boltzmann model.The correlation between pore and flow velocity distributions was analyzed.The DCM simulated porosity distribution includes both the large pores between the quartz particles and the fine pores that usually coexist with calcium carbonate in the sample.Numerical results demonstrated that the pores which are smaller than the CT voxel size,have a significant impact on sample permeability.It is observed that pore distribution near the fluid inlet surface has a noticeable influence on flow velocity distribution in the sample.Evident positive correlation was not observed between fluid velocity and porosity.

关 键 词：数据约束模型 X射线CT成像 部分渗透 格子玻尔兹曼法 

分 类 号：O242[理学—计算数学]