Scholarlink

<p>This study investigates the potential of incorporating silicon carbide waste (SiCW) as a partial replacement for fine aggregate in M25-grade concrete to promote sustainability in construction materials. SiCW, recognized for its superior thermal resistance, hardness, and durability, was used in varying proportions of 5%, 10%, 15%, 20%, and 25% to evaluate its effects on mechanical and durability properties. Experimental results revealed that the optimum replacement level was 15%, achieving a maximum compressive strength of approximately 50&#xa0;MPa, flexural strength of 9.5&#xa0;MPa, and improved split tensile strength at 10% replacement, with a slight reduction in workability due to the angular nature of SiC particles. Beyond 15%, the strength gradually declined. To improve predictive understanding, five machine learning (ML) models, Linear Regression (LR), Decision Tree (DT), Random Forest (RF), Support Vector Regression (SVR), and K-Nearest Neighbors (KNN) were employed to estimate compressive strength. Among these, the Decision Tree model exhibited superior performance, achieving the lowest Mean Squared Error (MSE = 80.26) and Mean Absolute Error (MAE), outperforming LR (133.13), RF (85.46), KNN (108.72), and SVR (117.09). The integration of experimental and AI-driven prediction approaches confirms that 15% SiCW provides optimal mechanical performance and creates a clear path for developing future eco-friendly concretes.</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

Utilization of silicon carbide waste in concrete: experimental assessment and AI-driven strength prediction

  • 期刊论文

  • 发表日期：

    2025年12月27日

查看全文
  • R. Prithvi,
  • P. Sangeetha,
  • P. Kaythry,
  • Ninu Praseetha Nirmala Sreedharan

摘要

This study investigates the potential of incorporating silicon carbide waste (SiCW) as a partial replacement for fine aggregate in M25-grade concrete to promote sustainability in construction materials. SiCW, recognized for its superior thermal resistance, hardness, and durability, was used in varying proportions of 5%, 10%, 15%, 20%, and 25% to evaluate its effects on mechanical and durability properties. Experimental results revealed that the optimum replacement level was 15%, achieving a maximum compressive strength of approximately 50 MPa, flexural strength of 9.5 MPa, and improved split tensile strength at 10% replacement, with a slight reduction in workability due to the angular nature of SiC particles. Beyond 15%, the strength gradually declined. To improve predictive understanding, five machine learning (ML) models, Linear Regression (LR), Decision Tree (DT), Random Forest (RF), Support Vector Regression (SVR), and K-Nearest Neighbors (KNN) were employed to estimate compressive strength. Among these, the Decision Tree model exhibited superior performance, achieving the lowest Mean Squared Error (MSE = 80.26) and Mean Absolute Error (MAE), outperforming LR (133.13), RF (85.46), KNN (108.72), and SVR (117.09). The integration of experimental and AI-driven prediction approaches confirms that 15% SiCW provides optimal mechanical performance and creates a clear path for developing future eco-friendly concretes.