Ramadhani, Gema (2022) Prediksi Permeabilitas Efektif Minyak (KO) Pada Buildup Test Dengan Menggunakan Extreme Gradient Boosting Algorithm Pada Sumur X. Other thesis, Universitas Islam Riau.
Text
173210233.pdf - Submitted Version Download (2MB) |
Abstract
Well Test is one way to determine the behaviour of reservoirs and wells, and can also determine the ability of a reservoir to produce. The parameter is the effective oil permeability (Ko). One method that can be done is testing to close the well (buildup test). This study uses the Computer Modeling Group (CMG)-IMEX reservoir simulation software for reservoir modelling. Then built 500 dataset experiments with five inputs well test parameters, namely Initial Pressure, Production Rate, Thickness, Volume Formation Factor, and Viscosity using the Design of Experiment (DoE) method with effective oil permeability as the response parameter. To make it easier to determine the effective oil permeability (Ko), it is done using the Extreme Gradient Boosting Algorithm method by building predictive modelling. With a ratio of 80% for the training model to determine the effective oil permeability (Ko) and 20%, the model was tested to obtain an excellent XGBoost Algorithm prediction model that can predict the value of the effective oil permeability (Ko). From the simulation results carried out by the Extreme Gradient Boosting Algorithm, it is found that the predictive model with Mean Square Error (MSE) and Mean Absolute Error (MAE) values is close to 0 and for R2 training and testing with values of 0.898 and 0.922, respectively. This study describes the application of machine learning in determining reservoir parameters and predictive models of the Extreme Gradient Boosting Algorithm that can be used as a reference and evaluate the buildup test performance in predicting the value of the effective oil permeability (Ko) with an accuracy level of 0.922 and taking a shorter time without using reservoir simulation which can take a lot of time. Keywords: Well Testing, Buildup Test, Effective Oil Permeability, Machine Learning, Extreme Gradient Boosting Algorithm
Item Type: | Thesis (Other) | ||||||
---|---|---|---|---|---|---|---|
Contributors: |
|
||||||
Uncontrolled Keywords: | Well Testing, Buildup Test, Effective Oil Permeability, Machine Learning, Extreme Gradient Boosting Algorithm | ||||||
Subjects: | T Technology > TN Mining engineering. Metallurgy | ||||||
Divisions: | > Teknik Perminyakan | ||||||
Depositing User: | Saputra Yogi UNILAK | ||||||
Date Deposited: | 30 Aug 2022 07:54 | ||||||
Last Modified: | 30 Aug 2022 07:54 | ||||||
URI: | http://repository.uir.ac.id/id/eprint/14530 |
Actions (login required)
View Item |