Machine Learning-Assisted Prediction of Oil Production and CO<sub>2</sub> Storage Effect in CO<sub>2</sub>-Water-Alternating-Gas Injection (CO<sub>2</sub>-WAG)
In recent years, CO<sub>2</sub> flooding has emerged as an efficient method for improving oil recovery. It also has the advantage of storing CO<sub>2</sub> underground. As one of the promising types of CO<sub>2</sub> enhanced oil recovery (CO<sub>2</sub&g...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-10-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/21/10958 |
| _version_ | 1827647403905253376 |
|---|---|
| author | Hangyu Li Changping Gong Shuyang Liu Jianchun Xu Gloire Imani |
| author_facet | Hangyu Li Changping Gong Shuyang Liu Jianchun Xu Gloire Imani |
| author_sort | Hangyu Li |
| collection | DOAJ |
| description | In recent years, CO<sub>2</sub> flooding has emerged as an efficient method for improving oil recovery. It also has the advantage of storing CO<sub>2</sub> underground. As one of the promising types of CO<sub>2</sub> enhanced oil recovery (CO<sub>2</sub>-EOR), CO<sub>2</sub> water-alternating-gas injection (CO<sub>2</sub>-WAG) can suppress CO<sub>2</sub> fingering and early breakthrough problems that occur during oil recovery by CO<sub>2</sub> flooding. However, the evaluation of CO<sub>2</sub>-WAG is strongly dependent on the injection parameters, which in turn renders numerical simulations computationally expensive. So, in this work, machine learning is used to help predict how well CO<sub>2</sub>-WAG will work when different injection parameters are used. A total of 216 models were built by using CMG numerical simulation software to represent CO<sub>2</sub>-WAG development scenarios of various injection parameters where 70% of them were used as training sets and 30% as testing sets. A random forest regression algorithm was used to predict CO<sub>2</sub>-WAG performance in terms of oil production, CO<sub>2</sub> storage amount, and CO<sub>2</sub> storage efficiency. The CO<sub>2</sub>-WAG period, CO<sub>2</sub> injection rate, and water–gas ratio were chosen as the three main characteristics of injection parameters. The prediction results showed that the predicted value of the test set was very close to the true value. The average absolute prediction deviations of cumulative oil production, CO<sub>2</sub> storage amount, and CO<sub>2</sub> storage efficiency were 1.10%, 3.04%, and 2.24%, respectively. Furthermore, it only takes about 10 s to predict the results of all 216 scenarios by using machine learning methods, while the CMG simulation method spends about 108 min. It demonstrated that the proposed machine-learning method can rapidly predict CO<sub>2</sub>-WAG performance with high accuracy and high computational efficiency under conditions of various injection parameters. This work gives more insights into the optimization of the injection parameters for CO<sub>2</sub>-EOR. |
| first_indexed | 2024-03-09T19:18:38Z |
| format | Article |
| id | doaj.art-fad322ba03114f2c84cb5ef951dbeea1 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T19:18:38Z |
| publishDate | 2022-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-fad322ba03114f2c84cb5ef951dbeea12023-11-24T03:35:39ZengMDPI AGApplied Sciences2076-34172022-10-0112211095810.3390/app122110958Machine Learning-Assisted Prediction of Oil Production and CO<sub>2</sub> Storage Effect in CO<sub>2</sub>-Water-Alternating-Gas Injection (CO<sub>2</sub>-WAG)Hangyu Li0Changping Gong1Shuyang Liu2Jianchun Xu3Gloire Imani4Key Laboratory of Unconventional Oil & Gas Development, China University of Petroleum (East China), Ministry of Education, Qingdao 266580, ChinaKey Laboratory of Unconventional Oil & Gas Development, China University of Petroleum (East China), Ministry of Education, Qingdao 266580, ChinaKey Laboratory of Unconventional Oil & Gas Development, China University of Petroleum (East China), Ministry of Education, Qingdao 266580, ChinaKey Laboratory of Unconventional Oil & Gas Development, China University of Petroleum (East China), Ministry of Education, Qingdao 266580, ChinaKey Laboratory of Unconventional Oil & Gas Development, China University of Petroleum (East China), Ministry of Education, Qingdao 266580, ChinaIn recent years, CO<sub>2</sub> flooding has emerged as an efficient method for improving oil recovery. It also has the advantage of storing CO<sub>2</sub> underground. As one of the promising types of CO<sub>2</sub> enhanced oil recovery (CO<sub>2</sub>-EOR), CO<sub>2</sub> water-alternating-gas injection (CO<sub>2</sub>-WAG) can suppress CO<sub>2</sub> fingering and early breakthrough problems that occur during oil recovery by CO<sub>2</sub> flooding. However, the evaluation of CO<sub>2</sub>-WAG is strongly dependent on the injection parameters, which in turn renders numerical simulations computationally expensive. So, in this work, machine learning is used to help predict how well CO<sub>2</sub>-WAG will work when different injection parameters are used. A total of 216 models were built by using CMG numerical simulation software to represent CO<sub>2</sub>-WAG development scenarios of various injection parameters where 70% of them were used as training sets and 30% as testing sets. A random forest regression algorithm was used to predict CO<sub>2</sub>-WAG performance in terms of oil production, CO<sub>2</sub> storage amount, and CO<sub>2</sub> storage efficiency. The CO<sub>2</sub>-WAG period, CO<sub>2</sub> injection rate, and water–gas ratio were chosen as the three main characteristics of injection parameters. The prediction results showed that the predicted value of the test set was very close to the true value. The average absolute prediction deviations of cumulative oil production, CO<sub>2</sub> storage amount, and CO<sub>2</sub> storage efficiency were 1.10%, 3.04%, and 2.24%, respectively. Furthermore, it only takes about 10 s to predict the results of all 216 scenarios by using machine learning methods, while the CMG simulation method spends about 108 min. It demonstrated that the proposed machine-learning method can rapidly predict CO<sub>2</sub>-WAG performance with high accuracy and high computational efficiency under conditions of various injection parameters. This work gives more insights into the optimization of the injection parameters for CO<sub>2</sub>-EOR.https://www.mdpi.com/2076-3417/12/21/10958CO<sub>2</sub>-WAGmachine learninginjection parametersrandom forest regressionCCUS |
| spellingShingle | Hangyu Li Changping Gong Shuyang Liu Jianchun Xu Gloire Imani Machine Learning-Assisted Prediction of Oil Production and CO<sub>2</sub> Storage Effect in CO<sub>2</sub>-Water-Alternating-Gas Injection (CO<sub>2</sub>-WAG) Applied Sciences CO<sub>2</sub>-WAG machine learning injection parameters random forest regression CCUS |
| title | Machine Learning-Assisted Prediction of Oil Production and CO<sub>2</sub> Storage Effect in CO<sub>2</sub>-Water-Alternating-Gas Injection (CO<sub>2</sub>-WAG) |
| title_full | Machine Learning-Assisted Prediction of Oil Production and CO<sub>2</sub> Storage Effect in CO<sub>2</sub>-Water-Alternating-Gas Injection (CO<sub>2</sub>-WAG) |
| title_fullStr | Machine Learning-Assisted Prediction of Oil Production and CO<sub>2</sub> Storage Effect in CO<sub>2</sub>-Water-Alternating-Gas Injection (CO<sub>2</sub>-WAG) |
| title_full_unstemmed | Machine Learning-Assisted Prediction of Oil Production and CO<sub>2</sub> Storage Effect in CO<sub>2</sub>-Water-Alternating-Gas Injection (CO<sub>2</sub>-WAG) |
| title_short | Machine Learning-Assisted Prediction of Oil Production and CO<sub>2</sub> Storage Effect in CO<sub>2</sub>-Water-Alternating-Gas Injection (CO<sub>2</sub>-WAG) |
| title_sort | machine learning assisted prediction of oil production and co sub 2 sub storage effect in co sub 2 sub water alternating gas injection co sub 2 sub wag |
| topic | CO<sub>2</sub>-WAG machine learning injection parameters random forest regression CCUS |
| url | https://www.mdpi.com/2076-3417/12/21/10958 |
| work_keys_str_mv | AT hangyuli machinelearningassistedpredictionofoilproductionandcosub2substorageeffectincosub2subwateralternatinggasinjectioncosub2subwag AT changpinggong machinelearningassistedpredictionofoilproductionandcosub2substorageeffectincosub2subwateralternatinggasinjectioncosub2subwag AT shuyangliu machinelearningassistedpredictionofoilproductionandcosub2substorageeffectincosub2subwateralternatinggasinjectioncosub2subwag AT jianchunxu machinelearningassistedpredictionofoilproductionandcosub2substorageeffectincosub2subwateralternatinggasinjectioncosub2subwag AT gloireimani machinelearningassistedpredictionofoilproductionandcosub2substorageeffectincosub2subwateralternatinggasinjectioncosub2subwag |