Optimization of Machine Learning Models for Jiwa Garuda in Predicting Geothermal Well Flow Rates

Abstract

The accurate prediction of geothermal well flow rates is critical for optimizing resource utilization and ensuring sustainable energy production. This study focuses on the optimization of machine learning models, termed "Jiwa Garuda," specifically designed for geothermal applications. The research aims to develop a robust predictive framework by leveraging advanced machine learning techniques to model complex thermodynamic and fluid dynamic behaviors within geothermal reservoirs. The outcomes of this research provide actionable insights for geothermal field operators, including predictive capabilities for well flow rates under varying operational scenarios. Furthermore, the Jiwa Garuda model offers potential scalability to other geothermal sites, contributing to the broader adoption of machine learning in sustainable energy development.