Application of Artificial Intelligence in Forecasting Geothermal Production

Abstract

Abstract A reservoir is the main asset of a geothermal business. Decreasing reservoir performance affects the sustainability of production in the future. In planning future production strategies, forecasting production projections are used with the assumption of particular parameter values. A reservoir is a porous media with a heterogeneous nature with a high degree of uncertainty, so using a specific production forecasting method’s assumption parameter becomes inaccurate. This study aims to develop alternative methods for optimizing and estimating geothermal production by eliminating assumptions. Artificial intelligence (AI) is an alternative method that can be used to predict reservoir productions that has properties with a high degree of uncertainty and can be used to optimize production. The AI model was created using measurement data of production parameters at the steam dominated geothermal field in Patuha, Indonesia. AI models in estimating geothermal reservoir production are more accurate in calculating production projections because they use real field data and eliminate assumptions.