Yonas Tilahun, Xiao Qinghua, Argaw Asha Ashongo, Xiangyu Han
{"title":"Determination of Compaction Parameters of Cement‐Lime Soils: Boosting‐Based Ensemble Models","authors":"Yonas Tilahun, Xiao Qinghua, Argaw Asha Ashongo, Xiangyu Han","doi":"10.1002/nag.3846","DOIUrl":null,"url":null,"abstract":"This study investigates the application of artificial intelligence (AI) models to predict soil compaction characteristics, specifically maximum dry density (<jats:italic>M</jats:italic><jats:sub>DD</jats:sub>) and optimum moisture content (<jats:italic>O</jats:italic><jats:sub>MC</jats:sub>), which are critical for stabilizing construction foundations. Traditional methods for determining <jats:italic>M</jats:italic><jats:sub>DD</jats:sub> and <jats:italic>O</jats:italic><jats:sub>MC</jats:sub> are labor‐intensive and often influenced by factors such as soil type, plasticity, and compaction energy (<jats:italic>E</jats:italic>). The research employed AI models, including random forest regression (RF‐R), gradient boosting regression (GB‐R), XGBoosting regressor (XGB‐R), and multilinear regression (ML‐R), trained on a comprehensive dataset of soil properties. For the first time, compaction energy has been used as an input variable to predict soil cement lime stabilized compaction parameters. Among the models, GB‐R demonstrated the highest prediction accuracy for <jats:italic>M</jats:italic><jats:sub>DD</jats:sub> and <jats:italic>O</jats:italic><jats:sub>MC</jats:sub>, outperforming RF‐R, XGB‐R, and ML‐R. The performance of built‐in models has been measured by three new index performance metrics: the a20‐index, the index of scatter (IS), and the index of agreement (IA), in addition to four common metrics. Taylor diagrams confirmed the robustness of these predictions during lab testing. A sensitivity analysis revealed that <jats:italic>M</jats:italic><jats:sub>DD</jats:sub> and <jats:italic>O</jats:italic><jats:sub>MC</jats:sub> were most influenced by plastic limit (PL), compaction energy (<jats:italic>E</jats:italic>), liquid limit (LL), and plasticity index (PI). Additionally, curve‐fitting techniques were applied to model the relationship between <jats:italic>M</jats:italic><jats:sub>DD</jats:sub>, <jats:italic>O</jats:italic><jats:sub>MC</jats:sub>, and these key factors. The results indicated that the GB‐R model, particularly when focused on essential features, provided superior accuracy compared to traditional regression methods, offering a reliable tool for soil stabilization assessments in construction.","PeriodicalId":13786,"journal":{"name":"International Journal for Numerical and Analytical Methods in Geomechanics","volume":null,"pages":null},"PeriodicalIF":3.4000,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal for Numerical and Analytical Methods in Geomechanics","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1002/nag.3846","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, GEOLOGICAL","Score":null,"Total":0}
引用次数: 0
Abstract
This study investigates the application of artificial intelligence (AI) models to predict soil compaction characteristics, specifically maximum dry density (MDD) and optimum moisture content (OMC), which are critical for stabilizing construction foundations. Traditional methods for determining MDD and OMC are labor‐intensive and often influenced by factors such as soil type, plasticity, and compaction energy (E). The research employed AI models, including random forest regression (RF‐R), gradient boosting regression (GB‐R), XGBoosting regressor (XGB‐R), and multilinear regression (ML‐R), trained on a comprehensive dataset of soil properties. For the first time, compaction energy has been used as an input variable to predict soil cement lime stabilized compaction parameters. Among the models, GB‐R demonstrated the highest prediction accuracy for MDD and OMC, outperforming RF‐R, XGB‐R, and ML‐R. The performance of built‐in models has been measured by three new index performance metrics: the a20‐index, the index of scatter (IS), and the index of agreement (IA), in addition to four common metrics. Taylor diagrams confirmed the robustness of these predictions during lab testing. A sensitivity analysis revealed that MDD and OMC were most influenced by plastic limit (PL), compaction energy (E), liquid limit (LL), and plasticity index (PI). Additionally, curve‐fitting techniques were applied to model the relationship between MDD, OMC, and these key factors. The results indicated that the GB‐R model, particularly when focused on essential features, provided superior accuracy compared to traditional regression methods, offering a reliable tool for soil stabilization assessments in construction.
期刊介绍:
The journal welcomes manuscripts that substantially contribute to the understanding of the complex mechanical behaviour of geomaterials (soils, rocks, concrete, ice, snow, and powders), through innovative experimental techniques, and/or through the development of novel numerical or hybrid experimental/numerical modelling concepts in geomechanics. Topics of interest include instabilities and localization, interface and surface phenomena, fracture and failure, multi-physics and other time-dependent phenomena, micromechanics and multi-scale methods, and inverse analysis and stochastic methods. Papers related to energy and environmental issues are particularly welcome. The illustration of the proposed methods and techniques to engineering problems is encouraged. However, manuscripts dealing with applications of existing methods, or proposing incremental improvements to existing methods – in particular marginal extensions of existing analytical solutions or numerical methods – will not be considered for review.