Scholarlink

<p>Tyre trash and massive volumes of construction and demolition debris (C&amp;D) are becoming major environmental concerns on a worldwide scale. It is suggested that recycled coarse aggregate (RCA) from C&amp;D and crumb rubber (CR) made from used tyres and used instead of natural aggregates in newly manufactured building materials to address this problem. Assessing and determining the best tree-based and machine-learning methods for flexural strength (f<sub>s</sub>) predictions in fiber-reinforced rubberized recycled aggregate concrete (FRRAC) is the aim of this study. 102 empirical findings from available articles and nineteen input variables were used in this endeavor to build and analyze machine-learning approaches for evaluating f<sub>s</sub>. This goal led to the development of the Random Forests (RF). The Jellyfish search algorithm (JS) and the Tasmanian devil algorithm (TD) processes are coupled to RF to&#xa0;discover the appropriate set; hyperparameters are essential in this simulation. Combining the TD and JSO approach (also known as RF<sub>JS</sub> and RF<sub>TD</sub>) with the <InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="41939_2025_811_Article_IEq1.gif" Format="GIF" Height="14" Rendition="HTML" Resolution="72" Type="Linedraw" Width="29" /> </InlineMediaObject> <EquationSource Format="TEX">\(RF\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi mathvariant="italic">RF</mi> </mrow> </math></EquationSource> </InlineEquation> technique allowed for the discovery of the f<sub>s</sub> of FRRAC. The variance percentages of the two measurement models are at least 45%; in some instances, they are reduced by 75%, demonstrating the predictive power of the RF<sub>TD</sub> for competence and dependability. According to uncertainty findings, RF<sub>TD</sub> outperformed RF<sub>JS</sub> by 50% in learning and 75% in the assessment stages.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Hybrid random forest systems on flexural strength of fiber-reinforced rubberized recycled aggregate concrete

  • 期刊论文

  • 发表日期：

    2025年4月9日

查看全文
  • Chao Wang

摘要

Tyre trash and massive volumes of construction and demolition debris (C&D) are becoming major environmental concerns on a worldwide scale. It is suggested that recycled coarse aggregate (RCA) from C&D and crumb rubber (CR) made from used tyres and used instead of natural aggregates in newly manufactured building materials to address this problem. Assessing and determining the best tree-based and machine-learning methods for flexural strength (fs) predictions in fiber-reinforced rubberized recycled aggregate concrete (FRRAC) is the aim of this study. 102 empirical findings from available articles and nineteen input variables were used in this endeavor to build and analyze machine-learning approaches for evaluating fs. This goal led to the development of the Random Forests (RF). The Jellyfish search algorithm (JS) and the Tasmanian devil algorithm (TD) processes are coupled to RF to discover the appropriate set; hyperparameters are essential in this simulation. Combining the TD and JSO approach (also known as RFJS and RFTD) with the \(RF\) RF technique allowed for the discovery of the fs of FRRAC. The variance percentages of the two measurement models are at least 45%; in some instances, they are reduced by 75%, demonstrating the predictive power of the RFTD for competence and dependability. According to uncertainty findings, RFTD outperformed RFJS by 50% in learning and 75% in the assessment stages.