{"id":60084,"date":"2023-04-24T05:00:49","date_gmt":"2023-04-23T21:00:49","guid":{"rendered":"http:\/\/learncmg.cn\/?p=60084"},"modified":"2023-04-24T05:00:49","modified_gmt":"2023-04-23T21:00:49","slug":"273-%e5%bc%80%e5%8f%91%e6%9c%ba%e5%99%a8%e5%ad%a6%e4%b9%a0%e6%a8%a1%e5%9e%8b%e9%a2%84%e6%b5%8b%e6%b7%b1%e5%b1%82%e5%90%ab%e6%b0%b4%e5%b1%82%e4%ba%8c%e6%b0%a7%e5%8c%96%e7%a2%b3%e5%b0%81%e5%ad%98","status":"publish","type":"post","link":"http:\/\/learncmg.cn\/?p=60084","title":{"rendered":"273. \u5f00\u53d1\u673a\u5668\u5b66\u4e60\u6a21\u578b\u9884\u6d4b\u6df1\u5c42\u542b\u6c34\u5c42\u4e8c\u6c27\u5316\u78b3\u5c01\u5b58\u6548\u679c"},"content":{"rendered":"<h3><\/h3>\n<h3 style=\"text-align: center;\"><a href=\"https:\/\/assets.researchsquare.com\/files\/rs-587644\/v1\/623cae14-a6c1-42d8-9c80-7e2eafdaea47.pdf?c=1637244892\">Developing machine learning models to predict CO2 trapping performance in deep saline aquifers<\/a><\/h3>\n<p><span style=\"color: #000000;\">\u6df1\u5c42\u76d0\u6c34\u5c42\u88ab\u89c6\u4e3a\u5730\u8d28\u50a8\u5b58\u4e8c\u6c27\u5316\u78b3\uff08GCS\uff09\u7684\u6f5c\u5728\u5730\u70b9\u3002\u4e3a\u4e86\u66f4\u6df1\u5165\u5730\u4e86\u89e3\u76d0\u6c34\u5c42\u4e2dCO\u2082\u6355\u96c6\u7684\u673a\u5236\uff0c\u6709\u5fc5\u8981\u5f00\u53d1\u53ef\u9760\u7684\u5de5\u5177\u6765\u8bc4\u4f30CO\u2082\u56fa\u5b9a\u6548\u7387\u3002\u672c\u6587\u4ecb\u7ecd\u4e86\u9ad8\u65af\u8fc7\u7a0b\u56de\u5f52\uff08GPR\uff09\u3001\u652f\u6301\u5411\u91cf\u673a\uff08SVM\uff09\u548c\u968f\u673a\u68ee\u6797\uff08RF\uff09\u5728\u76d0\u6c34\u5c42\u4e2d\u9884\u6d4bCO\u2082\u6355\u96c6\u6548\u7387\u7684\u5e94\u7528\u3002<\/span><br \/>\n<span style=\"color: #000000;\">\u9996\u5148\uff0c\u901a\u8fc7\u4f7f\u7528\u5730\u8d28\u53c2\u6570\u3001\u5ca9\u77f3\u7269\u7406\u7279\u6027\u548c\u5176\u4ed6\u7269\u7406\u7279\u6027\u6570\u636e\u7b49\u4e0d\u786e\u5b9a\u53d8\u91cf\uff0c\u521b\u5efa\u4e86\u4e00\u4e2a\u8bad\u7ec3\u6570\u636e\u96c6\u3002\u968f\u540e\u8fdb\u884c\u4e86\u603b\u5171101\u6b21\u6cb9\u85cf\u6a21\u62df\u6837\u672c\uff0c\u5e76\u6536\u96c6\u4e86\u675f\u7f1a\u7a7a\u95f4\u6355\u96c6\u3001\u6eb6\u89e3\u6355\u96c6\u548c\u7d2f\u8ba1\u6ce8\u5165CO\u2082\u7684\u6570\u636e\u3002<\/span><br \/>\n<span style=\"color: #000000;\">\u9884\u6d4b\u7ed3\u679c\u663e\u793a\uff0c\u4e09\u4e2a\u673a\u5668\u5b66\u4e60\uff08ML\uff09\u6a21\u578b\u7684\u6027\u80fd\u6392\u5e8f\u4e3a\uff1aGPR\u3001SVM\u548cRF\u3002\u8fd9\u4e9b\u6a21\u578b\u53ef\u7528\u4e8e\u9884\u6d4b\u6df1\u5c42\u76d0\u6c34\u5c42\u4e2d\u7684CO\u2082\u6355\u96c6\u6548\u7387\u3002\u5176\u4e2d\uff0cGPR\u6a21\u578b\u8868\u73b0\u51fa\u8272\uff0c\u5177\u6709\u6700\u9ad8\u7684\u76f8\u5173\u7cfb\u6570\uff08R\u00b2 = 0.992\uff09\u548c\u6700\u4f4e\u7684\u5747\u65b9\u6839\u8bef\u5dee\uff08RMSE = 0.00491\uff09\u3002<\/span><br \/>\n<span style=\"color: #000000;\">\u901a\u8fc7\u5728\u8d8a\u5357\u8fd1\u6d77\u7684\u4e00\u4e2a\u5b9e\u9645\u6cb9\u85cf\u4e2d\u9a8c\u8bc1\u4e86GPR\u6a21\u578b\u7684\u51c6\u786e\u6027\u548c\u7a33\u5b9a\u6027\u3002\u9884\u6d4b\u6a21\u578b\u5728\u6a21\u62df\u573a\u666f\u4e0e\u9884\u6d4b\u56fa\u5b9a\u6307\u6570\u4e4b\u95f4\u53d6\u5f97\u4e86\u5f88\u597d\u7684\u4e00\u81f4\u6027\u3002\u8fd9\u4e9b\u7814\u7a76\u7ed3\u679c\u8868\u660e\uff0cGPR\u673a\u5668\u5b66\u4e60\u6a21\u578b\u53ef\u4ee5\u4f5c\u4e3a\u4e00\u4e2a\u53ef\u9760\u7684\u9884\u6d4b\u5de5\u5177\uff0c\u652f\u6301\u6570\u503c\u6a21\u62df\u6765\u4f30\u8ba1\u5730\u4e0bCO\u2082\u56fa\u5b9a\u7684\u6027\u80fd\u3002<\/span><\/p>\n<h5><span style=\"color: #000000;\">Abstract and Figures<\/span><\/h5>\n<p><span style=\"color: #000000;\">Deep saline formations are considered as potential sites for geological carbon storage (GCS). To better understand the CO 2 trapping mechanism in saline aquifers, it is necessary to develop robust tools to evaluate CO 2 trapping efficiency. This paper introduces the application of Gaussian process regression (GPR), support vector machine (SVM), and random forest (RF) to predict CO 2 trapping efficiency in saline formations. First, the uncertainty variables, including geologic parameters, petrophysical properties, and other physical characteristics data were utilized to create a training dataset. A total of 101 reservoir simulation samples were then performed, and the residual trapping, solubility trapping, and cumulative CO 2 injection were collected.<\/span><br \/>\n<span style=\"color: #000000;\">The predicted results indicate that three machine learning (ML) models that evaluate performance from high to low: GPR, SVM, and RF can be selected to predict the CO 2 trapping efficiency in deep saline formations. The GPR model has an excellent CO 2 trapping prediction efficiency with the highest correlation factor (R \u00b2 = 0.992) and lowest root mean square error (RMSE = 0.00491). The accuracy and stability of the GPR models were verified for an actual reservoir in offshore Vietnam. The predictive models obtained a good agreement between the simulated field and the predicted trapping index. These findings indicate that the GPR ML models can support the numerical simulation as a robust predictive tool for estimating the performance of CO 2 trapping in the subsurface.<\/span><br \/>\n<img class=\"wp-image-60085 aligncenter\" src=\"http:\/\/learncmg.cn\/wp-content\/uploads\/2023\/04\/unnamed-file-38.png\" alt=\"\u6587\u672c, \u65e5\u7a0b\u8868\n\u63cf\u8ff0\u5df2\u81ea\u52a8\u751f\u6210\" \/><br \/>\n<img class=\"wp-image-60086 aligncenter\" src=\"http:\/\/learncmg.cn\/wp-content\/uploads\/2023\/04\/unnamed-file-39.png\" alt=\"\u8868\u683c\n\u63cf\u8ff0\u5df2\u81ea\u52a8\u751f\u6210\" \/><br \/>\n<img class=\"wp-image-60087 alignright\" src=\"http:\/\/learncmg.cn\/wp-content\/uploads\/2023\/04\/unnamed-file-40.png\" alt=\"\u56fe\u793a\n\u63cf\u8ff0\u5df2\u81ea\u52a8\u751f\u6210\" \/><br \/>\n<img class=\"wp-image-60088\" src=\"http:\/\/learncmg.cn\/wp-content\/uploads\/2023\/04\/unnamed-file-41.png\" alt=\"\u56fe\u793a, \u6587\u672c\n\u63cf\u8ff0\u5df2\u81ea\u52a8\u751f\u6210\" \/><br \/>\n<img class=\"wp-image-60089 aligncenter\" src=\"http:\/\/learncmg.cn\/wp-content\/uploads\/2023\/04\/unnamed-file-42.png\" alt=\"\u56fe\u8868\n\u63cf\u8ff0\u5df2\u81ea\u52a8\u751f\u6210\" \/><br \/>\n<img class=\"wp-image-60090 aligncenter\" src=\"http:\/\/learncmg.cn\/wp-content\/uploads\/2023\/04\/unnamed-file-43.png\" alt=\"\u56fe\u793a\n\u63cf\u8ff0\u5df2\u81ea\u52a8\u751f\u6210\" \/><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Developing machine learning models to predict CO2 <span class=\"more-link\"><a href=\"http:\/\/learncmg.cn\/?p=60084\">Continue Reading<\/a><\/span><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[159],"tags":[251,387,457],"_links":{"self":[{"href":"http:\/\/learncmg.cn\/index.php?rest_route=\/wp\/v2\/posts\/60084"}],"collection":[{"href":"http:\/\/learncmg.cn\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/learncmg.cn\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/learncmg.cn\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/learncmg.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=60084"}],"version-history":[{"count":0,"href":"http:\/\/learncmg.cn\/index.php?rest_route=\/wp\/v2\/posts\/60084\/revisions"}],"wp:attachment":[{"href":"http:\/\/learncmg.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=60084"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/learncmg.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=60084"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/learncmg.cn\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=60084"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}