Shale porosity prediction based on random forest algorithm
Precise and fast acquisition of shale porosity is important for the prediction of the spatial distribution of shale oil and the exploration target. To address the problem of low accuracy of porosity prediction using logging response equation， a porosity prediction model based on random forest algori...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Petroleum Geology and Recovery Efficiency
2023-11-01
|
| Series: | Youqi dizhi yu caishoulu |
| Subjects: | |
| Online Access: | http://yqdzycsl.cnjournals.com/pgre/article/abstract/202306002?st=search |
| _version_ | 1797272295118471168 |
|---|---|
| author | CUI Junfeng YANG Jinlu WANG Min WANG Xin WU Yan YU Changqi |
| author_facet | CUI Junfeng YANG Jinlu WANG Min WANG Xin WU Yan YU Changqi |
| author_sort | CUI Junfeng |
| collection | DOAJ |
| description | Precise and fast acquisition of shale porosity is important for the prediction of the spatial distribution of shale oil and the exploration target. To address the problem of low accuracy of porosity prediction using logging response equation， a porosity prediction model based on random forest algorithm is established， and the prediction accuracy is compared with those of BP neural network， support vector machine， and XGBoost algorithm， and the importance and influence range of logging parameters are analyzed by SHAP method. The results show that the random forest algorithm can better predict shale porosity， and the prediction effect is better than BP neural network， support vector machine， and XGBoost algorithm； the application of shale porosity prediction based on random forest algorithm in a depression in Bohai Bay Basin finds that the top three most important logging parameters for model prediction of porosity are compensation neutron， natural gamma， and ordinary apparent resistivity； the shale porosity prediction model based on random forest algorithm can quickly identify the porosity of a single well， which can not only compensate for the difficulty of obtaining the complete porosity distribution characteristics due to the inability of continuous coring but also significantly improve the efficiency and accuracy of porosity prediction. |
| first_indexed | 2024-03-07T14:26:17Z |
| format | Article |
| id | doaj.art-8800a2bbe12d4e729e8293aaead198e0 |
| institution | Directory Open Access Journal |
| issn | 1009-9603 |
| language | zho |
| last_indexed | 2024-03-07T14:26:17Z |
| publishDate | 2023-11-01 |
| publisher | Editorial Office of Petroleum Geology and Recovery Efficiency |
| record_format | Article |
| series | Youqi dizhi yu caishoulu |
| spelling | doaj.art-8800a2bbe12d4e729e8293aaead198e02024-03-06T08:02:59ZzhoEditorial Office of Petroleum Geology and Recovery EfficiencyYouqi dizhi yu caishoulu1009-96032023-11-01306132110.13673/j.cnki.cn37-1359/TE.2023060021009-9603（2023）06-0013-09Shale porosity prediction based on random forest algorithmCUI Junfeng0YANG Jinlu1WANG Min2WANG Xin3WU Yan4YU Changqi5Research Institute of Petroleum Exploration & Development， PetroChina， Beijing City， 100083， ChinaLaboratory of Deep Oil and Gas， China University of Petroleum （East China）， Qingdao City， Shandong Province， 266580， ChinaLaboratory of Deep Oil and Gas， China University of Petroleum （East China）， Qingdao City， Shandong Province， 266580， ChinaLaboratory of Deep Oil and Gas， China University of Petroleum （East China）， Qingdao City， Shandong Province， 266580， ChinaLaboratory of Deep Oil and Gas， China University of Petroleum （East China）， Qingdao City， Shandong Province， 266580， ChinaLaboratory of Deep Oil and Gas， China University of Petroleum （East China）， Qingdao City， Shandong Province， 266580， ChinaPrecise and fast acquisition of shale porosity is important for the prediction of the spatial distribution of shale oil and the exploration target. To address the problem of low accuracy of porosity prediction using logging response equation， a porosity prediction model based on random forest algorithm is established， and the prediction accuracy is compared with those of BP neural network， support vector machine， and XGBoost algorithm， and the importance and influence range of logging parameters are analyzed by SHAP method. The results show that the random forest algorithm can better predict shale porosity， and the prediction effect is better than BP neural network， support vector machine， and XGBoost algorithm； the application of shale porosity prediction based on random forest algorithm in a depression in Bohai Bay Basin finds that the top three most important logging parameters for model prediction of porosity are compensation neutron， natural gamma， and ordinary apparent resistivity； the shale porosity prediction model based on random forest algorithm can quickly identify the porosity of a single well， which can not only compensate for the difficulty of obtaining the complete porosity distribution characteristics due to the inability of continuous coring but also significantly improve the efficiency and accuracy of porosity prediction.http://yqdzycsl.cnjournals.com/pgre/article/abstract/202306002?st=searchrandom forestmachine learningloggingporosity predictionshale |
| spellingShingle | CUI Junfeng YANG Jinlu WANG Min WANG Xin WU Yan YU Changqi Shale porosity prediction based on random forest algorithm Youqi dizhi yu caishoulu random forest machine learning logging porosity prediction shale |
| title | Shale porosity prediction based on random forest algorithm |
| title_full | Shale porosity prediction based on random forest algorithm |
| title_fullStr | Shale porosity prediction based on random forest algorithm |
| title_full_unstemmed | Shale porosity prediction based on random forest algorithm |
| title_short | Shale porosity prediction based on random forest algorithm |
| title_sort | shale porosity prediction based on random forest algorithm |
| topic | random forest machine learning logging porosity prediction shale |
| url | http://yqdzycsl.cnjournals.com/pgre/article/abstract/202306002?st=search |
| work_keys_str_mv | AT cuijunfeng shaleporositypredictionbasedonrandomforestalgorithm AT yangjinlu shaleporositypredictionbasedonrandomforestalgorithm AT wangmin shaleporositypredictionbasedonrandomforestalgorithm AT wangxin shaleporositypredictionbasedonrandomforestalgorithm AT wuyan shaleporositypredictionbasedonrandomforestalgorithm AT yuchangqi shaleporositypredictionbasedonrandomforestalgorithm |