PEMODELAN NUMERIK SISTEM HIDROTERMAL DI LAPANGAN PANASBUMI KAMOJANG JAWA BARAT

• 主要作者: SUMARDI
• 格式: Dissertations
• 語言: Indonesia
• 在線閱讀: https://digilib.itb.ac.id/gdl/view/5543

<br>Abstrac: </b><p align=\"justify\"> <br /> This objective of research is to explore the most appropriate model of hydrothermal system in the geothermal field of Kamojang. The model is analyzed by using finite difference and finite element simulators that are made in this research and are validated by using TOUGH2 simulator. The finite element simulator is used as a comparator to analyze the pressure distributions. The finite difference simulator is used to analyze the pressure distributions caused by fluctuated infiltration rate, discharge rates which are constant, exponentially decline, and chaotic change. It is also used to analyze the temperature distributions. <br /> <p align=\"justify\"> <br /> The hydrothermal system in the geothermal field of Kamojang can be modeled as a weak causality system because it subjects to the chaotic pertubation from the atmosphere such as fluctuated rainfall and the turbulent mass and heat flows from the lower fluid so that the discharge rates change in chaotic manner. The chaotic perturbation of rainfall has a little effect to the hydrothermal model of Kamojang. The chaotic perturbation of lower fluid has a large effect to it. By including this chaotic perturbation, it is found the hydrothermal model that has short term prediction ability with relative error 17.5%. The other model of hydrothermal system that performs long term prediction with constant and exponentially decline discharge rates, the variations of rock heterogeneity and grid size, result in the relative error greater than 21%. <br /> <p align=\"justify\"> <br /> The finite difference and TOUGH2 simulators are also used to analyze the temperature distributions in a layer reservoir model with homogeneous permeability and constant discharge rates. The finite difference simulator results in the relative error of 4.4%, while the TOUGH2 simulator results in the relative error of 3.2%.