检索结果-南通市图书馆

Modeli.g of Total Di.solved Soli.s (TDS) and Sodi.m Absorpti.n Rati. (SAR) i. the Edwards-Tri.i.y Plateau and Ogallala Aqui.ers i. the Mi.land-Odessa Regi.n Usi.g Random Forest Regressi.n and eXtreme Gradi.nt Boosti.g

引用

Journal of Geosci.nce and Envi.onment Protecti.n 2024年第5期12卷 218-241页

作者： azuka i. udeh Osayamen J. i.arhi.gbe Erepamo J. Omi.ti.i Department of Geosci.nce University of Texas Permian Basin Odessa Texas USA Department of Geology University of Pretoria Pretoria South Africa

Effi.i.nt water quali.y moni.ori.g and ensuri.g the safety of dri.ki.g water by government agenci.s i. areas where the resource i. constantly depleted due to anthropogeni. or natural factors cannot be overemphasi.ed. The above statement holds for West Texas, Mi.land, and Odessa Preci.ely. Two machi.e learni.g regressi.n algori.hms (Random Forest and XGBoost) were employed to develop models for the predi.ti.n of total di.solved soli.s (TDS) and sodi.m absorpti.n rati. (SAR) for effi.i.nt water quali.y moni.ori.g of two vi.al aqui.ers: Edward-Tri.i.y (plateau), and Ogallala aqui.ers. These two aqui.ers have contri.uted i.mensely to provi.i.g water for di.ferent uses rangi.g from domesti., agri.ultural, i.dustri.l, etc. The data was obtai.ed from the Texas Water Development Board (TWDB). The XGBoost and Random Forest models used i. thi. study gave an accurate predi.ti.n of observed data (TDS and SAR) for both the Edward-Tri.i.y (plateau) and Ogallala aqui.ers wi.h the R2 values consi.tently greater than 0.83. The Random Forest model gave a better predi.ti.n of TDS and SAR concentrati.n wi.h an average R, MAE, RMSE and MSE of 0.977, 0.015, 0.029 and 0.00, respecti.ely. For the XGBoost, an average R, MAE, RMSE, and MSE of 0.953, 0.016, 0.037 and 0.00, respecti.ely, were achi.ved. The overall performance of the models produced was i.pressi.e. From thi. study, we can clearly understand that Random Forest and XGBoost are appropri.te for water quali.y predi.ti.n and moni.ori.g i. an area of hi.h hydrocarbon acti.i.i.s li.e Mi.land and Odessa and West Texas at large.

关键词： Water Quali.y Predi.ti.n Predi.ti.e Modeli.g Aqui.ers Machi.e Learni.g Regressi.n eXtreme Gradi.nt Boosti.g

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Machi.e Learni.g and Stati.ti.al Analysi. i. Groundwater Moni.ori.g for Total Di.solved Soli.s Assessment i. Wi.kler County, Texas

引用

Journal of Geosci.nce and Envi.onment Protecti.n 2024年第6期12卷 1-29页

作者： azuka i. udeh Osayamen J. i.arhi.gbe Erepamo J. Omi.ti.i.Abdulqudus O. Mohammed Oluwatomi.ola Andre-Obayanju Department of Geosci.nce University of Texas Permian Basin Odessa USA Department of Geology University of Pretoria Pretoria South Africa Department of Geology University of Benin Benin Nigeria

Thi. research ai.s to develop reli.ble models usi.g machi.e learni.g algori.hms to preci.ely predi.t Total Di.solved Soli.s (TDS) i. wells of the Permi.n basi., Wi.kler County, Texas. The data for thi. contri.uti.n was obtai.ed from the Texas Water Development Board websi.e (TWDB). Fi.e hundred and ni.ety-three samples were obtai.ed from two hundred and ni.ety-ei.ht wells i. the study area. The wells were dri.led at di.ferent county locati.ns i.to fi.e aqui.ers, i.cludi.g Pecos Valley, Dockum, Capi.an Reef, Edward Tri.i.y, and Rustler aqui.ers. A total of fourteen di.ferent water quali.y parameters were used, and they i.clude Potenti.l hydrogen (pH), Sodi.m, Chlori.e, Magnesi.m, Fluori.e, TDS, Speci.i. Conductance, Ni.rate, Total Hardness, Calci.m, Temperature, Well Depth, Sulphate, and Bi.arbonates. Four machi.e learni.g regressi.n algori.hms were developed to get a good model to help predi.t TDS i. thi. area: Deci.i.n Tree regressi.n, Li.ear regressi.n, Support Vector Regressi.n, and K-nearest nei.hbor. The study showed that the Deci.i.n Tree produced the best model wi.h attri.utes li.e the coeffi.i.nt of determi.ati.n R2 = 1.00 and 0.96 for the trai.i.g and testi.g, respecti.ely. i. also produced the lowest score of mean absolute error MAE = 0.00 and 0.04 for trai.i.g and testi.g, respecti.ely. Thi. study wi.l reduce the cost of obtai.i.g di.ferent water quali.y parameters i. TDS determi.ati.n by leveragi.g machi.e learni.g to use only the parameters contri.uti.g to TDS, thereby helpi.g researchers obtai. only the parameters necessary for TDS predi.ti.n. i. wi.l also help the authori.i.s enact poli.i.s that wi.l i.prove the water quali.y i. areas where dri.ki.g water avai.abi.i.y i. a challenge by provi.i.g i.portant i.formati.n for moni.ori.g and assessi.g groundwater quali.y.

关键词： Machi.e Learni.g Regressi.n Aqui.ers Wi.kler County Si.kholes

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：