EXPERIMENTAL DESIGN AND RESPONSE SURFACE METHODOLOGY IN PROBABILISTIC GEOTHERMAL RESOURCE ASSESSMENT: A STUDY CASE FOR ATADEI GEOTHERMAL FIELD, INDONESIA

• Main Author: Christian Supijo, Marchel
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/39473
• Institution: Institut Teknologi Bandung

Energy reserve is the most important aspect in geothermal development. The amount of energy reserve in geothermal will affect the development strategy. Therefore, reserve calculation is an important thing to do. The best method to calculate the reserves and commonly used at this time was the numerical simulation method. However, data input for numerical simulation in geothermal has a high uncertainty with a low level of confidence. The estimation of reserve potential needs to be done using experimental design (ED) and response surface method (RSM) combined with a probabilistic approach that starts with numerical simulation output to increase the level of confidence. This approach using five types of ED and RSM with a total number of 179 running models. The calculation of energy potential in numerical model using all types of ED dan RSM which aim to find out which type has the most effective and efficient regression equation by comparing the R-square value (R2). The numerical model built using TOUGH2 V.2. The numerical model using Atadei geothermal field, where all the data obtained from the published paper. The numerical model that has reached natural state condition is then integrated into the Leapfrog Geothermal application to facilitate slicing in various directions in order to update the conceptual model of the Atadei geothermal field. The regression equation based on energy potential calculations for all types of ED and RSM then integrated into Monte Carlo simulations to obtain probabilistic geothermal energy reserves for P10, P50, and P90 respectively. The comparison results for FFD 2 levels, FFD 3 levels, PBD, BBD, and CCD show that BBD has the greatest R-square (R2) value (99.72%). This indicate that BBD is the most effective and efficient type in calculating probabilistic energy potential using numerical models.