Analisis Parameter Yang Berpengaruh Terhadap Recovery Factor & Sor Pada Steam Assisted Gravity Drainage Menggunakan Artificial Neural Network

Dandy, Dandy (2022) Analisis Parameter Yang Berpengaruh Terhadap Recovery Factor & Sor Pada Steam Assisted Gravity Drainage Menggunakan Artificial Neural Network. Other thesis, Universitas Islam Riau.

Text 173210023.pdf - Submitted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Heavy oil is not easy to exploit, so more effort is needed to obtain it by using Enhanced Oil Recovery (EOR). Steam Assisted Gravity Drainage (SAGD) is an Enhanced Oil Recovery (EOR) thermal recovery process based on steam injection combined with horizontal well technology. In the Steam Assisted Gravity Drainage (SAGD) process, high uncertainty of reservoir properties and operational parameters affect the recovery production, therefore it is necessary to analyze the performance of the sensitivity level of operating parameters from influential to not influential in order to maximize the SAGD process. This is the basis for the analysis of the influence of the SAGD parameter on this research. The purpose of this research is to analyze the parameters that affect the value of recovery factor and SOR on Steam Assisted Gravity Drainage using Artificial Neural Network (ANN). The method used in this study is a simulation research by modeling the SAGD basecase using CMG-STARS software and sensitivity data using CMG-CMOST with input parameters of steam temperature, injection pressure, Steam quality, injection volume, injection rate and Pre-heating period with output in the form of recovery factor and steam oil ratio using Artificial Neural Network (ANN) with backpropagation method so that accurate prediction results can be obtained. The analysis model uses a neural network with the backpropagation algorithm on the value of recovery factor and steam oil ratio with 1008 sample data. The results are quite good with nodes hidden layer 12 by getting R2 results that are close to 1, namely training 0.999 and testing 0.999, and also getting an RMSE value. and MAPE is close to 0 i.e. RMSE training value is 0.032 and the testing data is 0.034, while the MAPE training value is 0.065 and the testing data is 0.068. From the simulation results, it is also found that the level of test parameters that affect the recovery factor and steam oil ratio in the order of injection volume, injection rate, steam quality, preheating period, injection temperature and injection pressure.

Actions (login required)

View Item