Modeling unsaturated hydraulic conductivity of compacted bentonite using a constrained CatBoost with bootstrap analysis

IF 5.3 2区地球科学 Q2 CHEMISTRY, PHYSICAL

Applied Clay Science Pub Date : 2024-08-21 DOI:10.1016/j.clay.2024.107530

Reza Taherdangkoo, Thomas Nagel, Chaofan Chen, Mostafa Mollaali, Mehran Ghasabeh, Olivier Cuisinier, Adel Abdallah, Christoph Butscher

{"title":"Modeling unsaturated hydraulic conductivity of compacted bentonite using a constrained CatBoost with bootstrap analysis","authors":"Reza Taherdangkoo, Thomas Nagel, Chaofan Chen, Mostafa Mollaali, Mehran Ghasabeh, Olivier Cuisinier, Adel Abdallah, Christoph Butscher","doi":"10.1016/j.clay.2024.107530","DOIUrl":null,"url":null,"abstract":"Accurately determining the hydraulic conductivity of unsaturated bentonite is important for modeling subsurface thermo-hydro-mechanical and chemical processes. This study introduced a new hybrid model that employs a constrained categorial boosting (CatBoost) algorithm, combined with a genetic algorithm for hyperparameter tuning, to estimate the hydraulic conductivity of unsaturated compacted bentonite The performance of the constrained CatBoost model was benchmarked against a diverse set of data-driven baseline regression models, including lasso, elastic net, polynomial, k-nearest neighbors, decision tree, bagging tree, random forest, and CatBoost. The results indicated that the constrained CatBoost model offers a superior balance between model robustness and predictive accuracy in estimating the hydraulic conductivity of compacted bentonite-based materials during the wetting phase. The model effectively captured the U-shape relationship between hydraulic conductivity and suction, a key characteristic of bentonite behavior. Additionally, bootstrapping analyses confirmed the model's reliability under data variability, further validating its applicability in environmental and geotechnical applications.","PeriodicalId":245,"journal":{"name":"Applied Clay Science","volume":"49 1","pages":""},"PeriodicalIF":5.3000,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Applied Clay Science","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.1016/j.clay.2024.107530","RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"CHEMISTRY, PHYSICAL","Score":null,"Total":0}

引用次数: 0

Abstract

Accurately determining the hydraulic conductivity of unsaturated bentonite is important for modeling subsurface thermo-hydro-mechanical and chemical processes. This study introduced a new hybrid model that employs a constrained categorial boosting (CatBoost) algorithm, combined with a genetic algorithm for hyperparameter tuning, to estimate the hydraulic conductivity of unsaturated compacted bentonite The performance of the constrained CatBoost model was benchmarked against a diverse set of data-driven baseline regression models, including lasso, elastic net, polynomial, k-nearest neighbors, decision tree, bagging tree, random forest, and CatBoost. The results indicated that the constrained CatBoost model offers a superior balance between model robustness and predictive accuracy in estimating the hydraulic conductivity of compacted bentonite-based materials during the wetting phase. The model effectively captured the U-shape relationship between hydraulic conductivity and suction, a key characteristic of bentonite behavior. Additionally, bootstrapping analyses confirmed the model's reliability under data variability, further validating its applicability in environmental and geotechnical applications.

查看原文本刊更多论文

利用带引导分析的约束 CatBoost 对压实膨润土的非饱和导水率进行建模

准确测定非饱和膨润土的导水率对于地下热-水-机械和化学过程建模非常重要。本研究引入了一种新的混合模型，该模型采用受约束分类提升（CatBoost）算法，并结合遗传算法进行超参数调整，以估算非饱和压实膨润土的水力传导性。受约束 CatBoost 模型的性能以一系列数据驱动的基线回归模型为基准，包括套索、弹性网、多项式、k 近邻、决策树、袋装树、随机森林和 CatBoost。结果表明，约束 CatBoost 模型在估算湿润阶段压实膨润土基材料的水力传导性时，在模型稳健性和预测准确性之间实现了出色的平衡。该模型有效地捕捉到了水导率与吸力之间的 U 型关系，这是膨润土行为的一个关键特征。此外，引导分析证实了该模型在数据变化情况下的可靠性，进一步验证了其在环境和岩土工程应用中的适用性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Applied Clay Science 地学-矿物学

CiteScore

10.30

自引率

10.70%

发文量

289

审稿时长

39 days

期刊介绍： Applied Clay Science aims to be an international journal attracting high quality scientific papers on clays and clay minerals, including research papers, reviews, and technical notes. The journal covers typical subjects of Fundamental and Applied Clay Science such as: • Synthesis and purification • Structural, crystallographic and mineralogical properties of clays and clay minerals • Thermal properties of clays and clay minerals • Physico-chemical properties including i) surface and interface properties; ii) thermodynamic properties; iii) mechanical properties • Interaction with water, with polar and apolar molecules • Colloidal properties and rheology • Adsorption, Intercalation, Ionic exchange • Genesis and deposits of clay minerals • Geology and geochemistry of clays • Modification of clays and clay minerals properties by thermal and physical treatments • Modification by chemical treatments with organic and inorganic molecules(organoclays, pillared clays) • Modification by biological microorganisms. etc...