![](https://digilib.itb.ac.id/assets/icon/pdf.gif)
2024 TA PP HAFIZH MUALIK 1-ABSTRAK
Terbatas  Suharsiyah
» Gedung UPT Perpustakaan
Terbatas  Suharsiyah
» Gedung UPT Perpustakaan
The global demand for energy continues to rise, highlighting the need for sustainable energy solutions, especially given the decline in oil and natural gas production. This study focuses on enhancing the recovery of heavy oil using the Steam Assisted Gravity Drainage (SAGD) method, a thermal recovery technique that effectively reduces oil viscosity and improves recovery rates. SAGD works by injecting steam into the reservoir through horizontal wells, creating a steam chamber that heats the heavy oil, thus reducing its viscosity. The heated oil then flows down to the production well below due to gravity. This method has proven effective in various locations, particularly in areas with significant heavy oil reserves.
A synthetic reservoir model was developed using CMG Builder and STARS, and various parameters were assessed through sensitivity analysis and optimization using the CMG DECE engine. The research employed polynomial regression and artificial neural networks (ANN) to create predictive models for cumulative oil production and steam chamber volume. The ANN configuration demonstrated superior performance, explaining 99.8% of the variability in the training data and 91.8% in the verification data. The study's findings confirmed that steam injection significantly enhances oil production rates by reducing oil viscosity and forming an extensive steam chamber. Over a ten-year period, the base model achieved a cumulative production recovery factor of approximately 38.9%.
This study provides a comprehensive analysis of the critical parameters influencing SAGD performance and offers practical insights into optimizing thermal recovery methods for heavy oil reservoirs. The results underscore the potential of advanced modelling and optimization techniques in improving reservoir management and production strategies.