SECONDARY POROSITY IDENTIFICATION: VUGS IN CARBONATE ROCK FORMATIONS USING MACHINE LEARNING BASED ON CONVENTIONAL LOGS DATA

• Main Author: Jati Syahrul Alim, Lukman
• Format: Theses
• Language: Indonesia
• Subjects: Pertambangan dan operasi berkaitan
• Online Access: https://digilib.itb.ac.id/gdl/view/85076
• Institution: Institut Teknologi Bandung

In most cases, conventional methods that use well log and image analysis to determine and delineate secondary porosity are time-consuming and costly. Consequently, this research will use machine learning methods to process and analyze log data such as gamma ray, neutron, density and resistivity in wells that do not have image log data, in order to improve the accuracy and rate of identification of secondary porosity types in carbonate rock formations using datasets that have been annotated with secondary porosity information from Formation Micro Imager (FMI). This research can improve the efficiency and accuracy of the secondary porosity identification process in carbonate rock formations, in order to open up better reservoir discovery opportunities at minimal cost. In this research, additional features such as shale volume, porosity density, total porosity, and effective porosity were created to improve the performance of the model. This research succeeded in finding an accurate and efficient machine learning algorithm to determine the type of secondary porosity, which is expected to reduce the time and cost for reservoir analysis and characterization.