ANALISIS CITRA DIGITAL DAN PEMODELAN FISIKA BATUAN UNTUK ESTIMASI INDEKS KERAPUHAN PADA BATUAN SEDIMEN DAN METAMARF

• Main Author: Kautsar Rahmareza, Muhammad
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/83413
• Institution: Institut Teknologi Bandung

The physical properties of rocks around us are influenced by their structure and mineral composition. These parameters affect the degree of deformation of a rock when subjected to load or stress. To observe the degree of rock deformation, calculations of the rock brittleness index can be performed. Several approaches to the value of the rock brittleness index, such as elastic parameters, mineral composition, and well data, have been extensively studied previously. However, there are some drawbacks, such as long processing time, higher costs, and potential damage to the rock samples. With the advancement of science, we can mitigate these drawbacks by using a digital rock physics approach. Using digital images of rock samples, segmentation is performed. Physical parameters can be estimated using a numerical approach. These estimated results can then be correlated with the rock brittleness index values. In this study, samples of carbonate sedimentary and metamorphic rock images with a size of 3003 pixels were used. The segmentation process using the k-means clustering algorithm yielded a porosity of 10.73%, calcite 31.26%, ankerite 47.13%, and quartz 10.88% for the sedimentary sample. For the metamorphic sample, it resulted in a porosity of 3.42%, calcite 8.24%, quartz 36.86%, magnetite 17.87%, and pyrrhotite 33.61%. The segmentation process using the k-means clustering method for both samples had a structural similarity value of more than 0.95. Estimation of physical parameters using a numerical approach yielded good results because the data fell within the Voigt-Reuss-Hill and Hashin-Shtrikman boundary models. Variations in quartz and porosity played an important role in the variation of estimated elastic modulus values in both samples. The calculation results of the brittleness index indicate that the metamorphic sample has a brittleness index ranging from 333 to 6262 (0.26 to 0.49 after normalization) with an average of 478 (0.37). Meanwhile, the sedimentary sample has a brittleness index ranging from 228 to 332.5 (0.18 to 0.26 after normalization) with an average of 280.4 (0.22). The classification results show that the metamorphic sample is relatively more brittle compared to the sedimentary sample. The analysis results indicate that the higher the porosity and calcite mineral content in the sample, the lower its brittleness index. Conversely, the higher the quartz mineral content in the sample, the higher the brittleness index. Additionally, if the pores are filled with fluid, water makes the rock softer compared to other fluids, thereby reducing its brittleness index. Overall, based on this study, the physical parameters of the rock can be estimated accurately without damaging the sample. Moreover, rock composition information can be effectively obtained from the segmentation process using the k-means clustering algorithm. Information on variations in mineral composition, porosity, and the presence of fluid in the pores influences the elastic modulus of the rock and also its brittleness index