Scholarlink

<p>The present article evaluated the compressive, flexural and split tensile strength at 28 and 90 days, of ternary blended concrete incorporating Ground Granulated Blast Furnace Slag (GGBS) as partial replacement of cement. Concrete mix were prepared with 30%, 45% and 60% GGBS, combined with 5%, 10% and 15% Metakaolin (MK)/Silica Fume (SF). Results indicated that binary GGBS blends achieve peak performance at 45% replacement, beyond which strength declines. In ternary systems, both MK and SF enhanced mechanical performance upto 10% substitution, followed by a reduction at 15% attributed to excessive micro filler demand and delayed hydration kinetics. The optimal mix comprising 45% GGBS with 10% MK/SF demonstrated superior strength across all test categories as compared to other blends. For optimization purpose, a hybrid Artificial Neural Network (ANN) - Genetic Algorithm (GA) model was developed to capture the non-linear interaction between constituent materials and strength. This proposed hybrid ANN-GA approach highlighted that 45% GGBS with 10% supplementary cementitious materials (SCMs) offers an optimal balance, which is suitable for structural and pavement applications.</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

Optimisation of GGBS-Based Ternary Concrete with Metakaolin and Silica Fume Using ANN-GA

  • 期刊论文

  • 发表日期：

    2026年6月3日

查看全文
  • Amol Sharma,
  • Siddharth Garia,
  • Sneha Das,
  • Arnab Kumar Pal

摘要

The present article evaluated the compressive, flexural and split tensile strength at 28 and 90 days, of ternary blended concrete incorporating Ground Granulated Blast Furnace Slag (GGBS) as partial replacement of cement. Concrete mix were prepared with 30%, 45% and 60% GGBS, combined with 5%, 10% and 15% Metakaolin (MK)/Silica Fume (SF). Results indicated that binary GGBS blends achieve peak performance at 45% replacement, beyond which strength declines. In ternary systems, both MK and SF enhanced mechanical performance upto 10% substitution, followed by a reduction at 15% attributed to excessive micro filler demand and delayed hydration kinetics. The optimal mix comprising 45% GGBS with 10% MK/SF demonstrated superior strength across all test categories as compared to other blends. For optimization purpose, a hybrid Artificial Neural Network (ANN) - Genetic Algorithm (GA) model was developed to capture the non-linear interaction between constituent materials and strength. This proposed hybrid ANN-GA approach highlighted that 45% GGBS with 10% supplementary cementitious materials (SCMs) offers an optimal balance, which is suitable for structural and pavement applications.