Perpustakaan Digital ITB

This study aims to analyze suitable Water Alternating Gas (WAG) parameters to optimize Enhanced Oil Recovery (EOR) techniques. The integration of machine learning with reservoir simulation techniques provides a robust framework for predicting and enhancing the effectiveness of Water Alternating Gas (WAG) injections. Key parameters such as injection rates, cycle times, and fluid properties are fine-tuned to maximize oil recovery while minimizing CO2 emissions. The research involves the application of advanced gradient boosting models to evaluate the performance of WAG injections under various scenarios. The results demonstrate that the optimized WAG parameters significantly improve the incremental oil recovery factor compared to conventional methods. Additionally, the study provides insights into the economic and environmental benefits of implementing optimized WAG strategies in mature oil fields. The findings underscore the potential of machine learning as a powerful tool in optimizing EOR processes, contributing to sustainable and efficient hydrocarbon production.