Machine-learning-based prediction of oil recovery factor for experimental CO2-Foam chemical EOR: Implications for carbon utilization projects

Hung Vo Thanh; Danial Sheini Dashtgoli; Hemeng Zhang; Baehyun Min

doi:10.1016/j.energy.2023.127860

Machine-learning-based prediction of oil recovery factor for experimental CO₂-Foam chemical EOR: Implications for carbon utilization projects

Hung Vo Thanh, Danial Sheini Dashtgoli, Hemeng Zhang, Baehyun Min

Climate and Energy Systems Engineering

Research output: Contribution to journal › Article › peer-review

4 Scopus citations

Abstract

Enhanced oil recovery (EOR) using CO₂ injection is promising with economic and environmental benefits as an active climate-change mitigation approach. Nevertheless, the low sweep efficiency of CO₂ injection remains a challenge. CO₂-foam injection has been proposed as a remedy, but its laboratory screening for specific reservoirs is costly and time-consuming. In this study, machine-learning models are employed to predict oil recovery factor (ORF) during CO₂-foam flooding cost-effectively and accurately. Four models, including general regression neural network (GRNN), cascade forward neural network with Levenberg–Marquardt optimization (CFNN-LM), cascade forward neural network with Bayesian regularization (CFNN-BR), and extreme gradient boosting (XGBoost), are evaluated based on experimental data from previous studies. Results demonstrate that the GRNN model outperforms the others, with an overall mean absolute error of 0.059 and an R² of 0.9999. The GRNN model's applicability domain is verified using a Williams plot, and an uncertainty analysis for CO₂-foam flooding projects is conducted. The novelty of this study lies in developing a machine-learning-based approach that provides an accurate and cost-effective prediction of ORF in CO₂-foam experiments. This approach has the potential to significantly reduce screening costs and time required for CO₂-foam injection, making it a more viable carbon utilization and EOR strategy.

Original language	English
Article number	127860
Journal	Energy
Volume	278
DOIs	https://doi.org/10.1016/j.energy.2023.127860
State	Published - 1 Sep 2023

Bibliographical note

Funding Information:
This research was supported by the National Research Foundation of Korea (NRF) grants (No. 2022H1D3A2A01096314 ) and the Korea Institute of Energy Technology Evaluation and Planning (KETEP) funded by the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20214710100060 and No. 20212010200010 ).

Funding Information:
In the petroleum industry and underground gas storage, ML-based models have been utilized for a variety of purposes, such as reserve appraisal in both traditional and non-traditional reservoirs [ 28–35], assessment of natural gas compressibility [36], prediction of reservoir quality [37], history matching of simulation models for oil production forecasts in fluvial channels [38], lithofacies and petrophysical predictions in carbonate reservoirs [39,40], prediction of cumulative oil production in shale formations [41], microbial enhanced oil recovery [42], pore pressure estimation using petrophysical well log data [43], and distribution 3D geostatistical models [44]. ML models have been used to predict oil recovery factors in several studies. Van Si et al. [45] built an ANN model for predicting the oil recovery factor (ORF) for CO2-EOR. Esene et al. [46] performed the ORF prediction using ANN, least-squares support vector machine (LSSVM), and gene expression programming (GEP) for a carbonate water-injection process. In their work, the ANN yielded the most accurate prediction performance with an R2 of 0.99. Recently, Larestani et al. [47] developed a series of ANN models and decision trees to predict the ORF and the net present value of chemical flooding projects, with the CFNN-LM model generating the highest predictive performance. These studies demonstrate the potential of ML models to predict recovery factors and optimize oil recovery processes. Table 1 highlights the employed machine learning approaches for prediction oil recovery performance in EOR projects.This research was supported by the National Research Foundation of Korea (NRF) grants (No. 2022H1D3A2A01096314) and the Korea Institute of Energy Technology Evaluation and Planning (KETEP) funded by the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20214710100060 and No. 20212010200010).

Publisher Copyright:
© 2023 Elsevier Ltd

Keywords

CFNN-BR
CFNN-LM
CO-EOR
CO-Foam experiments
GRNN
XGBoost

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.energy.2023.127860

Cite this

@article{870659bc0da54ff1a899035d9901b197,

title = "Machine-learning-based prediction of oil recovery factor for experimental CO2-Foam chemical EOR: Implications for carbon utilization projects",

abstract = "Enhanced oil recovery (EOR) using CO2 injection is promising with economic and environmental benefits as an active climate-change mitigation approach. Nevertheless, the low sweep efficiency of CO2 injection remains a challenge. CO2-foam injection has been proposed as a remedy, but its laboratory screening for specific reservoirs is costly and time-consuming. In this study, machine-learning models are employed to predict oil recovery factor (ORF) during CO2-foam flooding cost-effectively and accurately. Four models, including general regression neural network (GRNN), cascade forward neural network with Levenberg–Marquardt optimization (CFNN-LM), cascade forward neural network with Bayesian regularization (CFNN-BR), and extreme gradient boosting (XGBoost), are evaluated based on experimental data from previous studies. Results demonstrate that the GRNN model outperforms the others, with an overall mean absolute error of 0.059 and an R2 of 0.9999. The GRNN model's applicability domain is verified using a Williams plot, and an uncertainty analysis for CO2-foam flooding projects is conducted. The novelty of this study lies in developing a machine-learning-based approach that provides an accurate and cost-effective prediction of ORF in CO2-foam experiments. This approach has the potential to significantly reduce screening costs and time required for CO2-foam injection, making it a more viable carbon utilization and EOR strategy.",

keywords = "CFNN-BR, CFNN-LM, CO-EOR, CO-Foam experiments, GRNN, XGBoost",

author = "{Vo Thanh}, Hung and {Sheini Dashtgoli}, Danial and Hemeng Zhang and Baehyun Min",

note = "Funding Information: This research was supported by the National Research Foundation of Korea (NRF) grants (No. 2022H1D3A2A01096314 ) and the Korea Institute of Energy Technology Evaluation and Planning (KETEP) funded by the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20214710100060 and No. 20212010200010 ). Funding Information: In the petroleum industry and underground gas storage, ML-based models have been utilized for a variety of purposes, such as reserve appraisal in both traditional and non-traditional reservoirs [ 28–35], assessment of natural gas compressibility [36], prediction of reservoir quality [37], history matching of simulation models for oil production forecasts in fluvial channels [38], lithofacies and petrophysical predictions in carbonate reservoirs [39,40], prediction of cumulative oil production in shale formations [41], microbial enhanced oil recovery [42], pore pressure estimation using petrophysical well log data [43], and distribution 3D geostatistical models [44]. ML models have been used to predict oil recovery factors in several studies. Van Si et al. [45] built an ANN model for predicting the oil recovery factor (ORF) for CO2-EOR. Esene et al. [46] performed the ORF prediction using ANN, least-squares support vector machine (LSSVM), and gene expression programming (GEP) for a carbonate water-injection process. In their work, the ANN yielded the most accurate prediction performance with an R2 of 0.99. Recently, Larestani et al. [47] developed a series of ANN models and decision trees to predict the ORF and the net present value of chemical flooding projects, with the CFNN-LM model generating the highest predictive performance. These studies demonstrate the potential of ML models to predict recovery factors and optimize oil recovery processes. Table 1 highlights the employed machine learning approaches for prediction oil recovery performance in EOR projects.This research was supported by the National Research Foundation of Korea (NRF) grants (No. 2022H1D3A2A01096314) and the Korea Institute of Energy Technology Evaluation and Planning (KETEP) funded by the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20214710100060 and No. 20212010200010). Publisher Copyright: {\textcopyright} 2023 Elsevier Ltd",

year = "2023",

month = sep,

day = "1",

doi = "10.1016/j.energy.2023.127860",

language = "English",

volume = "278",

journal = "Energy",

issn = "0360-5442",

publisher = "Elsevier Ltd.",

}

TY - JOUR

T1 - Machine-learning-based prediction of oil recovery factor for experimental CO2-Foam chemical EOR

T2 - Implications for carbon utilization projects

AU - Vo Thanh, Hung

AU - Sheini Dashtgoli, Danial

AU - Zhang, Hemeng

AU - Min, Baehyun

N1 - Funding Information: This research was supported by the National Research Foundation of Korea (NRF) grants (No. 2022H1D3A2A01096314 ) and the Korea Institute of Energy Technology Evaluation and Planning (KETEP) funded by the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20214710100060 and No. 20212010200010 ). Funding Information: In the petroleum industry and underground gas storage, ML-based models have been utilized for a variety of purposes, such as reserve appraisal in both traditional and non-traditional reservoirs [ 28–35], assessment of natural gas compressibility [36], prediction of reservoir quality [37], history matching of simulation models for oil production forecasts in fluvial channels [38], lithofacies and petrophysical predictions in carbonate reservoirs [39,40], prediction of cumulative oil production in shale formations [41], microbial enhanced oil recovery [42], pore pressure estimation using petrophysical well log data [43], and distribution 3D geostatistical models [44]. ML models have been used to predict oil recovery factors in several studies. Van Si et al. [45] built an ANN model for predicting the oil recovery factor (ORF) for CO2-EOR. Esene et al. [46] performed the ORF prediction using ANN, least-squares support vector machine (LSSVM), and gene expression programming (GEP) for a carbonate water-injection process. In their work, the ANN yielded the most accurate prediction performance with an R2 of 0.99. Recently, Larestani et al. [47] developed a series of ANN models and decision trees to predict the ORF and the net present value of chemical flooding projects, with the CFNN-LM model generating the highest predictive performance. These studies demonstrate the potential of ML models to predict recovery factors and optimize oil recovery processes. Table 1 highlights the employed machine learning approaches for prediction oil recovery performance in EOR projects.This research was supported by the National Research Foundation of Korea (NRF) grants (No. 2022H1D3A2A01096314) and the Korea Institute of Energy Technology Evaluation and Planning (KETEP) funded by the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20214710100060 and No. 20212010200010). Publisher Copyright: © 2023 Elsevier Ltd

PY - 2023/9/1

Y1 - 2023/9/1

N2 - Enhanced oil recovery (EOR) using CO2 injection is promising with economic and environmental benefits as an active climate-change mitigation approach. Nevertheless, the low sweep efficiency of CO2 injection remains a challenge. CO2-foam injection has been proposed as a remedy, but its laboratory screening for specific reservoirs is costly and time-consuming. In this study, machine-learning models are employed to predict oil recovery factor (ORF) during CO2-foam flooding cost-effectively and accurately. Four models, including general regression neural network (GRNN), cascade forward neural network with Levenberg–Marquardt optimization (CFNN-LM), cascade forward neural network with Bayesian regularization (CFNN-BR), and extreme gradient boosting (XGBoost), are evaluated based on experimental data from previous studies. Results demonstrate that the GRNN model outperforms the others, with an overall mean absolute error of 0.059 and an R2 of 0.9999. The GRNN model's applicability domain is verified using a Williams plot, and an uncertainty analysis for CO2-foam flooding projects is conducted. The novelty of this study lies in developing a machine-learning-based approach that provides an accurate and cost-effective prediction of ORF in CO2-foam experiments. This approach has the potential to significantly reduce screening costs and time required for CO2-foam injection, making it a more viable carbon utilization and EOR strategy.

AB - Enhanced oil recovery (EOR) using CO2 injection is promising with economic and environmental benefits as an active climate-change mitigation approach. Nevertheless, the low sweep efficiency of CO2 injection remains a challenge. CO2-foam injection has been proposed as a remedy, but its laboratory screening for specific reservoirs is costly and time-consuming. In this study, machine-learning models are employed to predict oil recovery factor (ORF) during CO2-foam flooding cost-effectively and accurately. Four models, including general regression neural network (GRNN), cascade forward neural network with Levenberg–Marquardt optimization (CFNN-LM), cascade forward neural network with Bayesian regularization (CFNN-BR), and extreme gradient boosting (XGBoost), are evaluated based on experimental data from previous studies. Results demonstrate that the GRNN model outperforms the others, with an overall mean absolute error of 0.059 and an R2 of 0.9999. The GRNN model's applicability domain is verified using a Williams plot, and an uncertainty analysis for CO2-foam flooding projects is conducted. The novelty of this study lies in developing a machine-learning-based approach that provides an accurate and cost-effective prediction of ORF in CO2-foam experiments. This approach has the potential to significantly reduce screening costs and time required for CO2-foam injection, making it a more viable carbon utilization and EOR strategy.

KW - CFNN-BR

KW - CFNN-LM

KW - CO-EOR

KW - CO-Foam experiments

KW - GRNN

KW - XGBoost

UR - http://www.scopus.com/inward/record.url?scp=85160451417&partnerID=8YFLogxK

U2 - 10.1016/j.energy.2023.127860

DO - 10.1016/j.energy.2023.127860

M3 - Article

AN - SCOPUS:85160451417

SN - 0360-5442

VL - 278

JO - Energy

JF - Energy

M1 - 127860

ER -