SANDSTONE MICROSTRUCTURE MODELLING USING RANDOM POLYHEDRON PARTIALLY PENETRABLE

• Main Author: Sri Nurhayati, Kiki
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/53651
• Institution: Institut Teknologi Bandung

Quantitative analysis of porous rock can be carried out conventionally and numerically. Digital Rock Physics is a method of obtaining quantitative physical quantities of rock using numerical predictions on digital images obtained from microtomographic scans of the original sample. Modeling is carried out using the forward modeling technique which produces an image in discrete space with a partially concentrated random deposition algorithm. The resulting image is varied with overlap degree values, grain radius, and dimension size. The image characterization process is carried out using CT-An software to obtain the value of the image physical parameters. Then, a boxplot curve for the porosity value of the curve is made. The porosity curve shows that the smaller the sample dimensions, the wider the variance of the porosity value. Whereas in actual rocks, the porosity value does not depend on dimensions. Microscopic structural modeling of rocks combines micro and macro scale analysis. The main problem in modeling with a multi-scale approach is the determination of representative elementary volume, the minimum volume of rock samples when certain parameters become independent (independent) of the sample size. REV needs to be determined to determine the minimum volume that can be used as a representation of the actual sandstone. Based on the porosity curve, qualitatively the REV of the sandstone image is 2563 voxel. The grain shape of the 2563 voxel image is modified from round to polyhedron. The Berea sandstone image whose size has been adjusted to the size of the model, namely 2563 voxel, is analyzed using CTan and its physical parameter values, in the form of fractal dimensions, specific surface area, and degree of anisotropy, are compared with the physical parameters of the modeled image. Comparison of the fractal dimension and specific surface area parameters shows that the degree of overlap of the Berea samples is 0.2 and the radius is 20 pixel, while the comparison of the degree of anisotropy parameter shows that the degree of overlap of the Berea samples is 0.2 with different radii. These results indicate that this modeling have not represented sandstone perfectly so that other parameters are needed in the modeling.