Musti, Meichy Amanda (2023) Analisis Parameter Yang Berpengaruh Terhadap Recovery Factor Pada Expanding Solvent Steam Assisted Gravity Drainage Menggunakan Extreme Gradient Boosting. Other thesis, Universitas Islam Riau.

Text Meichy Amanda Musti.pdf - Submitted Version Restricted to Registered users only Download (3MB) | Request a copy

Abstract

Steam Assisted Gravity Drainage (SAGD) is a thermal recovery method that relies on gravity separation of steam, using a pair of parallel horizontal wells placed five meters apart. This method is mainly used in sandy oil fields (tar sands). One variation of SAGD is Expanding Solvent Steam Assisted Gravity Drainage (ES-SAGD), which involves the injection of a solvent in the process. ESSAGD promises better efficiency in terms of lower Steam Oil Ratio (SOR) and higher Recovery Factor (RF) compared to conventional SAGD methods. This study aims to investigate the effects of five ES-SAGD parameters, namely steam temperature, injection pressure, steam quality, injection rate, and solvent mole fraction, on oil recovery. The research began with 1000 experiments using the Design of Experiment (DoE) method in Computer Modeling Group (CMG) software. The response parameter measured is oil recovery. The aim of this research is to increase understanding of the parameters that influence the efficiency of ES-SAGD and optimize its implementation in sandy oil fields. The research background is based on the need for more efficient thermal recovery methods, and ES-SAGD was chosen because of the guarantee of increasing efficiency with additional solvent injection. The research methodology includes variations of ES-SAGD parameters in 1000 experiments, with feature analysis using the XGBoost algorithm to determine the relative influence of each parameter on oil recovery. This algorithm was chosen because of its ability to handle dependencies and interactions between parameters. The results of investigations using the XGBoost algorithm show that the parameters Pressure injection, Rate Injection, and Solvent Fraction have the most influence on oil recovery during the ES-SAGD process. Variations of the XGBoost hyperparameters were also tested, and the running results showed that the smallest MSE was achieved with 194 n-estimators, with R2 values for the training data and test data of 0.993 and 0.993 respectively. This research is expected to provide valuable information in the oil industry, by revealing the level of significance of each ES-SAGD parameter. These findings can help in improving the efficiency and optimization of the implementation of the ES-SAGD method in sandy oil fields.

Actions (login required)

View Item