Scholarlink

<p>The demand for raw materials in concrete production generates substantial waste annually, highlighting the potential of recycled aggregate concrete (RAC). However, challenges such as cracking and structural integrity issues arise due to adhered hardened mortar. This study evaluates six advanced models for predicting the splitting tensile strength (STS) of RAC using 605 experimental results. The models include interaction, full quadratic (FQ), artificial neural network (ANN), M5P-tree, gradient boosting (GB), and random forest (RF). Key influencing parameters were identified: water-to-cement ratio (0.25–0.97), cement (185–864&#xa0;kg/m<sup>3</sup>), gravel (0–1393.3&#xa0;kg/m<sup>3</sup>, size 4.75–37.5&#xa0;mm), sand (353–1105&#xa0;kg/m<sup>3</sup>), superplasticizer (0–2.50%), curing time (3–365&#xa0;days), and recycled coarse aggregate (RCA, 0–1393.3&#xa0;kg/m<sup>3</sup>). Among the models, the RF model excelled, achieving an R<sup>2</sup> of 0.99 and RMSE of 0.17&#xa0;MPa. Sensitivity analysis revealed curing time as the most critical factor influencing STS. These findings provide an accurate, cost-effective framework for developing eco-friendly RAC mixes. Additionally, a quadratic regression model was formulated to predict the relationship between compressive and splitting tensile strengths of RAC.</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

Innovative multi-model techniques to predict the tensile and compressive strength of recycled aggregate concrete

  • 期刊论文

  • 发表日期：

    2026年3月23日

查看全文
  • Yousif J. Bas,
  • Jamal I. Kakrasul,
  • Kamaran S. Ismail,
  • Samir M. Hamad

摘要

The demand for raw materials in concrete production generates substantial waste annually, highlighting the potential of recycled aggregate concrete (RAC). However, challenges such as cracking and structural integrity issues arise due to adhered hardened mortar. This study evaluates six advanced models for predicting the splitting tensile strength (STS) of RAC using 605 experimental results. The models include interaction, full quadratic (FQ), artificial neural network (ANN), M5P-tree, gradient boosting (GB), and random forest (RF). Key influencing parameters were identified: water-to-cement ratio (0.25–0.97), cement (185–864 kg/m3), gravel (0–1393.3 kg/m3, size 4.75–37.5 mm), sand (353–1105 kg/m3), superplasticizer (0–2.50%), curing time (3–365 days), and recycled coarse aggregate (RCA, 0–1393.3 kg/m3). Among the models, the RF model excelled, achieving an R2 of 0.99 and RMSE of 0.17 MPa. Sensitivity analysis revealed curing time as the most critical factor influencing STS. These findings provide an accurate, cost-effective framework for developing eco-friendly RAC mixes. Additionally, a quadratic regression model was formulated to predict the relationship between compressive and splitting tensile strengths of RAC.