Scholarlink

<p>The development of High-Performance Fiber-Reinforced Cementitious Composites (HPFRCC) with adequate compressive strength and high ductility is essential for constructing safer and more resilient infrastructure, particularly under seismic or dynamic loading conditions. This study develops a hybrid framework that explicitly integrates Artificial Neural Network (ANN) predictive modeling with the Fast-Convergence Multi-Objective Equilibrium Optimizer with Adaptive Elitist Preservation (FC-MOEO/AEP) to achieve simultaneous optimization of compressive strength and ductility in HPFRCC mixtures. The ANN was trained on experimental data to capture the nonlinear influence of mix proportions and then embedded within the optimization procedure to generate a Pareto front of non-dominated solutions. The results demonstrate that this integration can effectively predict the impact of varying proportions of cement (C), metakaolin (MTK), silica fume (SF), slag (SL), and limestone powder (LP) on HPFRCC performance while systematically identifying balanced design alternatives. Experimental validation on selected mixtures confirmed the framework’s effectiveness within the scope of this study. Overall, the integration of ANN with FC-MOEO/AEP provides a novel and efficient computational–experimental pathway for reducing the extent of trial-and-error testing and supporting informed decision-making in HPFRCC mix design.</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

Optimization-Based Prediction of Ductility and Compressive Strength in High-Performance Fiber-Reinforced Concrete: A Multi-Objective Framework with Experimental Insights

  • 期刊论文

  • 发表日期：

    2025年10月13日

查看全文
  • Peyman Farhadiyeganeh,
  • Maryam Firoozinezamabadi,
  • Ata Hojatkashani,
  • Abbas Akbarpour NikghalbRashti,
  • Hassan Abbasi

摘要

The development of High-Performance Fiber-Reinforced Cementitious Composites (HPFRCC) with adequate compressive strength and high ductility is essential for constructing safer and more resilient infrastructure, particularly under seismic or dynamic loading conditions. This study develops a hybrid framework that explicitly integrates Artificial Neural Network (ANN) predictive modeling with the Fast-Convergence Multi-Objective Equilibrium Optimizer with Adaptive Elitist Preservation (FC-MOEO/AEP) to achieve simultaneous optimization of compressive strength and ductility in HPFRCC mixtures. The ANN was trained on experimental data to capture the nonlinear influence of mix proportions and then embedded within the optimization procedure to generate a Pareto front of non-dominated solutions. The results demonstrate that this integration can effectively predict the impact of varying proportions of cement (C), metakaolin (MTK), silica fume (SF), slag (SL), and limestone powder (LP) on HPFRCC performance while systematically identifying balanced design alternatives. Experimental validation on selected mixtures confirmed the framework’s effectiveness within the scope of this study. Overall, the integration of ANN with FC-MOEO/AEP provides a novel and efficient computational–experimental pathway for reducing the extent of trial-and-error testing and supporting informed decision-making in HPFRCC mix design.