Loading…

Modeling Viscosity of CO2–N2 Gaseous Mixtures Using Robust Tree-Based Techniques: Extra Tree, Random Forest, GBoost, and LightGBM

Carbon dioxide (CO2) has an essential role in most enhanced oil recovery (EOR) methods in the oil industry. Oil swelling and viscosity reduction are the dominant mechanisms in an immiscible CO2-EOR process. Besides numerous CO2 applications in EOR, most oil reservoirs do not have access to natural C...

Full description

Saved in:
Bibliographic Details
Published in:ACS omega 2023-04, Vol.8 (15), p.13863-13875
Main Authors: Zheng, Haimin, Mahmoudzadeh, Atena, Amiri-Ramsheh, Behnam, Hemmati-Sarapardeh, Abdolhossein
Format: Article
Language:English
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 13875
container_issue 15
container_start_page 13863
container_title ACS omega
container_volume 8
creator Zheng, Haimin
Mahmoudzadeh, Atena
Amiri-Ramsheh, Behnam
Hemmati-Sarapardeh, Abdolhossein
description Carbon dioxide (CO2) has an essential role in most enhanced oil recovery (EOR) methods in the oil industry. Oil swelling and viscosity reduction are the dominant mechanisms in an immiscible CO2-EOR process. Besides numerous CO2 applications in EOR, most oil reservoirs do not have access to natural CO2, and capturing it from flue gas and other sources is costly. Flue gases are available in huge quantities at a significantly lower price and can be considered economically viable agents for EOR operations. In this work, four powerful machine learning algorithms, namely, extra tree (ET), random forest (RF), gradient boosting (GBoost), and light gradient boosted machine (LightGBM) were utilized to accurately estimate the viscosity of CO2–N2 mixtures. To this aim, a databank was employed, containing 3036 data points over wide ranges of pressures and temperatures. Temperature, pressure, and CO2 mole fraction were applied as input parameters, and the viscosity of the CO2–N2 mixture was the output. The RF smart model had the highest precision with the lowest average absolute percent relative error (AAPRE) of 1.58%, root mean square error (RMSE) of 2.221, and determination coefficient (R 2) of 0.9993. The trend analysis showed that the RF model could precisely predict the real physical behavior of the CO2–N2 viscosity variation. Finally, the outlier detection was performed using the leverage approach to demonstrate the validity of the utilized databank and the applicability area of the developed RF model. Accordingly, nearly 96% of the data points seemed to be dependable and valid, and the rest of them were located in the suspected and out-of-leverage data zones.
doi_str_mv 10.1021/acsomega.3c00228
format article
fullrecord <record><control><sourceid>proquest_doaj_</sourceid><recordid>TN_cdi_doaj_primary_oai_doaj_org_article_3fbb6927482549e5bfd1876f65d331cb</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><doaj_id>oai_doaj_org_article_3fbb6927482549e5bfd1876f65d331cb</doaj_id><sourcerecordid>2805518861</sourcerecordid><originalsourceid>FETCH-LOGICAL-a389t-5bbe82f139445bdb68e6c36e3ec98dba295636b0faf451103ec88e45e7b87bb93</originalsourceid><addsrcrecordid>eNpVkc1O3DAUhaOqSEWUPUsvu5hQ_8SO003VGcEUaaZIaGBr2clNxqMkprGDYFepj9A35ElwmKkEq2ude_xJ95wkOSP4nGBKvurSuw4afc5KjCmVH5JjmuU4JSxjH9-8PyWn3u8wxkRIKqk4Tv6uXQWt7Rt0Z33pvA1PyNVocU2f__z7RdFSe3CjR2v7GMYBPLr1k_nGmdEHtBkA0nm0VGgD5ba3v0fw39DFYxj063KGbnRfuQ5duvg5zNBy7tw0o4pWttmG5Xz9OTmqdevh9DBPktvLi83iZ7q6Xl4tfqxSzWQRUm4MSFoTVmQZN5UREkTJBDAoC1kZTQsumDC41nXGCcFRlxIyDrmRuTEFO0mu9tzK6Z26H2ynhyfltFWvghsapYdgyxYUq40RBc0zSXlWADd1RWQuasErxkhpIuv7nnU_mg6qEvp4cvsO-n7T261q3IMimBAhaB4JXw6EwU2xBdXFBqBtdT8lrqjEnBMpBYnW2d4ae1Y7Nw59jCmS1FS--l--OpTPXgBfgqWZ</addsrcrecordid><sourcetype>Open Website</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2805518861</pqid></control><display><type>article</type><title>Modeling Viscosity of CO2–N2 Gaseous Mixtures Using Robust Tree-Based Techniques: Extra Tree, Random Forest, GBoost, and LightGBM</title><source>American Chemical Society (ACS) Open Access</source><source>PubMed Central</source><creator>Zheng, Haimin ; Mahmoudzadeh, Atena ; Amiri-Ramsheh, Behnam ; Hemmati-Sarapardeh, Abdolhossein</creator><creatorcontrib>Zheng, Haimin ; Mahmoudzadeh, Atena ; Amiri-Ramsheh, Behnam ; Hemmati-Sarapardeh, Abdolhossein</creatorcontrib><description>Carbon dioxide (CO2) has an essential role in most enhanced oil recovery (EOR) methods in the oil industry. Oil swelling and viscosity reduction are the dominant mechanisms in an immiscible CO2-EOR process. Besides numerous CO2 applications in EOR, most oil reservoirs do not have access to natural CO2, and capturing it from flue gas and other sources is costly. Flue gases are available in huge quantities at a significantly lower price and can be considered economically viable agents for EOR operations. In this work, four powerful machine learning algorithms, namely, extra tree (ET), random forest (RF), gradient boosting (GBoost), and light gradient boosted machine (LightGBM) were utilized to accurately estimate the viscosity of CO2–N2 mixtures. To this aim, a databank was employed, containing 3036 data points over wide ranges of pressures and temperatures. Temperature, pressure, and CO2 mole fraction were applied as input parameters, and the viscosity of the CO2–N2 mixture was the output. The RF smart model had the highest precision with the lowest average absolute percent relative error (AAPRE) of 1.58%, root mean square error (RMSE) of 2.221, and determination coefficient (R 2) of 0.9993. The trend analysis showed that the RF model could precisely predict the real physical behavior of the CO2–N2 viscosity variation. Finally, the outlier detection was performed using the leverage approach to demonstrate the validity of the utilized databank and the applicability area of the developed RF model. Accordingly, nearly 96% of the data points seemed to be dependable and valid, and the rest of them were located in the suspected and out-of-leverage data zones.</description><identifier>ISSN: 2470-1343</identifier><identifier>EISSN: 2470-1343</identifier><identifier>DOI: 10.1021/acsomega.3c00228</identifier><language>eng</language><publisher>American Chemical Society</publisher><ispartof>ACS omega, 2023-04, Vol.8 (15), p.13863-13875</ispartof><rights>2023 The Authors. Published by American Chemical Society</rights><rights>2023 The Authors. Published by American Chemical Society 2023 The Authors</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><orcidid>0000-0002-5889-150X</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://pubs.acs.org/doi/pdf/10.1021/acsomega.3c00228$$EPDF$$P50$$Gacs$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://pubs.acs.org/doi/10.1021/acsomega.3c00228$$EHTML$$P50$$Gacs$$Hfree_for_read</linktohtml><link.rule.ids>230,314,727,780,784,885,27080,27924,27925,53791,53793,56762,56812</link.rule.ids></links><search><creatorcontrib>Zheng, Haimin</creatorcontrib><creatorcontrib>Mahmoudzadeh, Atena</creatorcontrib><creatorcontrib>Amiri-Ramsheh, Behnam</creatorcontrib><creatorcontrib>Hemmati-Sarapardeh, Abdolhossein</creatorcontrib><title>Modeling Viscosity of CO2–N2 Gaseous Mixtures Using Robust Tree-Based Techniques: Extra Tree, Random Forest, GBoost, and LightGBM</title><title>ACS omega</title><addtitle>ACS Omega</addtitle><description>Carbon dioxide (CO2) has an essential role in most enhanced oil recovery (EOR) methods in the oil industry. Oil swelling and viscosity reduction are the dominant mechanisms in an immiscible CO2-EOR process. Besides numerous CO2 applications in EOR, most oil reservoirs do not have access to natural CO2, and capturing it from flue gas and other sources is costly. Flue gases are available in huge quantities at a significantly lower price and can be considered economically viable agents for EOR operations. In this work, four powerful machine learning algorithms, namely, extra tree (ET), random forest (RF), gradient boosting (GBoost), and light gradient boosted machine (LightGBM) were utilized to accurately estimate the viscosity of CO2–N2 mixtures. To this aim, a databank was employed, containing 3036 data points over wide ranges of pressures and temperatures. Temperature, pressure, and CO2 mole fraction were applied as input parameters, and the viscosity of the CO2–N2 mixture was the output. The RF smart model had the highest precision with the lowest average absolute percent relative error (AAPRE) of 1.58%, root mean square error (RMSE) of 2.221, and determination coefficient (R 2) of 0.9993. The trend analysis showed that the RF model could precisely predict the real physical behavior of the CO2–N2 viscosity variation. Finally, the outlier detection was performed using the leverage approach to demonstrate the validity of the utilized databank and the applicability area of the developed RF model. Accordingly, nearly 96% of the data points seemed to be dependable and valid, and the rest of them were located in the suspected and out-of-leverage data zones.</description><issn>2470-1343</issn><issn>2470-1343</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>N~.</sourceid><sourceid>DOA</sourceid><recordid>eNpVkc1O3DAUhaOqSEWUPUsvu5hQ_8SO003VGcEUaaZIaGBr2clNxqMkprGDYFepj9A35ElwmKkEq2ude_xJ95wkOSP4nGBKvurSuw4afc5KjCmVH5JjmuU4JSxjH9-8PyWn3u8wxkRIKqk4Tv6uXQWt7Rt0Z33pvA1PyNVocU2f__z7RdFSe3CjR2v7GMYBPLr1k_nGmdEHtBkA0nm0VGgD5ba3v0fw39DFYxj063KGbnRfuQ5duvg5zNBy7tw0o4pWttmG5Xz9OTmqdevh9DBPktvLi83iZ7q6Xl4tfqxSzWQRUm4MSFoTVmQZN5UREkTJBDAoC1kZTQsumDC41nXGCcFRlxIyDrmRuTEFO0mu9tzK6Z26H2ynhyfltFWvghsapYdgyxYUq40RBc0zSXlWADd1RWQuasErxkhpIuv7nnU_mg6qEvp4cvsO-n7T261q3IMimBAhaB4JXw6EwU2xBdXFBqBtdT8lrqjEnBMpBYnW2d4ae1Y7Nw59jCmS1FS--l--OpTPXgBfgqWZ</recordid><startdate>20230418</startdate><enddate>20230418</enddate><creator>Zheng, Haimin</creator><creator>Mahmoudzadeh, Atena</creator><creator>Amiri-Ramsheh, Behnam</creator><creator>Hemmati-Sarapardeh, Abdolhossein</creator><general>American Chemical Society</general><scope>N~.</scope><scope>7X8</scope><scope>5PM</scope><scope>DOA</scope><orcidid>https://orcid.org/0000-0002-5889-150X</orcidid></search><sort><creationdate>20230418</creationdate><title>Modeling Viscosity of CO2–N2 Gaseous Mixtures Using Robust Tree-Based Techniques: Extra Tree, Random Forest, GBoost, and LightGBM</title><author>Zheng, Haimin ; Mahmoudzadeh, Atena ; Amiri-Ramsheh, Behnam ; Hemmati-Sarapardeh, Abdolhossein</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a389t-5bbe82f139445bdb68e6c36e3ec98dba295636b0faf451103ec88e45e7b87bb93</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Zheng, Haimin</creatorcontrib><creatorcontrib>Mahmoudzadeh, Atena</creatorcontrib><creatorcontrib>Amiri-Ramsheh, Behnam</creatorcontrib><creatorcontrib>Hemmati-Sarapardeh, Abdolhossein</creatorcontrib><collection>American Chemical Society (ACS) Open Access</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><collection>Directory of Open Access Journals</collection><jtitle>ACS omega</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Zheng, Haimin</au><au>Mahmoudzadeh, Atena</au><au>Amiri-Ramsheh, Behnam</au><au>Hemmati-Sarapardeh, Abdolhossein</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Modeling Viscosity of CO2–N2 Gaseous Mixtures Using Robust Tree-Based Techniques: Extra Tree, Random Forest, GBoost, and LightGBM</atitle><jtitle>ACS omega</jtitle><addtitle>ACS Omega</addtitle><date>2023-04-18</date><risdate>2023</risdate><volume>8</volume><issue>15</issue><spage>13863</spage><epage>13875</epage><pages>13863-13875</pages><issn>2470-1343</issn><eissn>2470-1343</eissn><abstract>Carbon dioxide (CO2) has an essential role in most enhanced oil recovery (EOR) methods in the oil industry. Oil swelling and viscosity reduction are the dominant mechanisms in an immiscible CO2-EOR process. Besides numerous CO2 applications in EOR, most oil reservoirs do not have access to natural CO2, and capturing it from flue gas and other sources is costly. Flue gases are available in huge quantities at a significantly lower price and can be considered economically viable agents for EOR operations. In this work, four powerful machine learning algorithms, namely, extra tree (ET), random forest (RF), gradient boosting (GBoost), and light gradient boosted machine (LightGBM) were utilized to accurately estimate the viscosity of CO2–N2 mixtures. To this aim, a databank was employed, containing 3036 data points over wide ranges of pressures and temperatures. Temperature, pressure, and CO2 mole fraction were applied as input parameters, and the viscosity of the CO2–N2 mixture was the output. The RF smart model had the highest precision with the lowest average absolute percent relative error (AAPRE) of 1.58%, root mean square error (RMSE) of 2.221, and determination coefficient (R 2) of 0.9993. The trend analysis showed that the RF model could precisely predict the real physical behavior of the CO2–N2 viscosity variation. Finally, the outlier detection was performed using the leverage approach to demonstrate the validity of the utilized databank and the applicability area of the developed RF model. Accordingly, nearly 96% of the data points seemed to be dependable and valid, and the rest of them were located in the suspected and out-of-leverage data zones.</abstract><pub>American Chemical Society</pub><doi>10.1021/acsomega.3c00228</doi><tpages>13</tpages><orcidid>https://orcid.org/0000-0002-5889-150X</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 2470-1343
ispartof ACS omega, 2023-04, Vol.8 (15), p.13863-13875
issn 2470-1343
2470-1343
language eng
recordid cdi_doaj_primary_oai_doaj_org_article_3fbb6927482549e5bfd1876f65d331cb
source American Chemical Society (ACS) Open Access; PubMed Central
title Modeling Viscosity of CO2–N2 Gaseous Mixtures Using Robust Tree-Based Techniques: Extra Tree, Random Forest, GBoost, and LightGBM
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T16%3A58%3A57IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_doaj_&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Modeling%20Viscosity%20of%20CO2%E2%80%93N2%20Gaseous%20Mixtures%20Using%20Robust%20Tree-Based%20Techniques:%20Extra%20Tree,%20Random%20Forest,%20GBoost,%20and%20LightGBM&rft.jtitle=ACS%20omega&rft.au=Zheng,%20Haimin&rft.date=2023-04-18&rft.volume=8&rft.issue=15&rft.spage=13863&rft.epage=13875&rft.pages=13863-13875&rft.issn=2470-1343&rft.eissn=2470-1343&rft_id=info:doi/10.1021/acsomega.3c00228&rft_dat=%3Cproquest_doaj_%3E2805518861%3C/proquest_doaj_%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-a389t-5bbe82f139445bdb68e6c36e3ec98dba295636b0faf451103ec88e45e7b87bb93%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2805518861&rft_id=info:pmid/&rfr_iscdi=true