论文部分内容阅读
孔隙结构是反映土体物理、力学性质变化的本征指标,传统医用CT分辨率尺度不足以辨识表征岩土体细观结构的单个颗粒或孔隙。使用了14μm高空间分辨率的工业显微CT,对砂雨法生成的南京粉砂试样进行了扫描。对CT图像预处理后所得二值数字图像序列进行了三维重构,实现了任意表征单元体(REV)提取,并可对REV采用三维二值矩阵加以表征与定量分析。通过分析断层图像序列,计算出土样整体的体积孔隙率,与实验孔隙率误差仅3.93%。沿试样高度计算表观孔隙率在20.97%~46.77%范围内波动。从试样底部一REV中提取出水平方向、与水平面成60°角以及两正交垂直方向共4个典型切面,对其孔隙定向性进行统计分析,结果表明,水平切面具有最小主定向角,两正交垂直切面中,一个具有最大主定向角,另一具有最小各向异性率,斜切面孔隙呈近似等向分布,无明显定向性。采用孔隙网络模型进行空间孔径定量分析,结果发现,所提取出3个REV的最大孔径分布在629~696μm,最小孔径分布在54~77μm,平均孔径分布在166~185μm,孔径分布在100~200μm范围内最为集中。方法可用于热、水、力等各类因素作用下土体空间孔隙结构的无损定量分析。
Pore structure is an intrinsic index that reflects the change of physical and mechanical properties of soil, and the resolution scale of traditional medical CT is not enough to identify single particles or pores that characterize the meso-structure of rock and soil. Using industrial micro-CT with 14μm high spatial resolution, the samples of Nanjing silt produced by the sand-rain method were scanned. Three-dimensional reconstruction of the binary digital image sequence obtained after CT image preprocessing has realized the extraction of arbitrary characterization unit body (REV) and the characterization and quantitative analysis of REV using three-dimensional binary matrix. By analyzing the sequence of tomographic images, the volume porosity of the whole sample is calculated, which is only 3.93% of the experimental porosity error. The apparent porosity along the sample height fluctuated within the range of 20.97% -46.77%. A REV from the bottom of the sample was extracted from the horizontal direction, horizontal angle of 60 ° and two orthogonal vertical direction of a total of four typical sections, the statistical analysis of the pore orientation, the results show that the horizontal section has the smallest main orientation angle, One of the two orthogonal vertical sections has the largest principal orientation angle and the other has the smallest anisotropy rate. The chamfered surface pores have an approximately isotropic distribution with no obvious directivity. Quantitative pore space analysis using pore network model results showed that the maximum REEs distribution of the three REVs ranged from 629 to 696μm, the minimum pore size distribution ranged from 54 to 77μm, the average pore size distribution ranged from 166 to 185μm and the pore size distribution ranged from 100 to 200μm The most concentrated area. The method can be applied to nondestructive quantitative analysis of pore structure of soil under the action of various factors such as heat, water and force.