Scholarlink

<p>This study presents the <b>first AI-assisted design and optimization framework</b> integrating <b>green chemical modification of banana fibers using Ca(OH)₂</b> for the development of <b>high-performance</b>,<b> eco-friendly epoxy composites</b>. Unlike conventional NaOH treatments, the Ca(OH)₂-based modification provides a milder and more sustainable approach that minimizes cellulose degradation while enhancing interfacial bonding. The integration of <b>Artificial Intelligence (AI)</b>,<b> Response Surface Methodology (RSM)</b>, and <b>Pareto-based multi-objective optimization (NSGA-II)</b> constitutes a novel methodological advance that unites experimental data, statistical design, and computational modeling within one framework—an approach rarely explored in natural fiber composites. At the optimized condition (20 wt% banana fibers and 5 wt% Ca(OH)₂), the composite exhibited <b>an ≈ 80% increase in tensile strength</b>, <b>a ≈ 70% improvement in flexural and compressive strength</b>, and <b>a twofold rise in impact resistance</b> relative to neat epoxy. Fire resistance improved substantially, with the limiting oxygen index increasing by <b>around 20%</b> and the burning rate reduced by <b>more than 35%</b>. Thermal stability was enhanced, showing <b>≈ 15% higher degradation onset temperature</b> and <b>≈ 45% greater char yield</b>. AI-assisted predictions were highly consistent with experimental and RSM results (R² &gt; 0.95), and Pareto analysis identified a balanced “knee-point” composition achieving optimal synergy between strength, flame retardancy, and thermal performance. Finite Element Analysis (FEA) further verified uniform stress distribution and structural integrity at the optimal formulation. Collectively, this research introduces a <b>new sustainable route for valorizing agricultural waste fibers</b> through the synergy of green chemistry and artificial intelligence. The proposed <b>AI–green optimization framework</b> not only advances the design of multifunctional natural-fiber composites but also establishes a scalable model for developing next-generation lightweight, fire-safe, and environmentally friendly structural materials.</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

Artificial intelligence-assisted design of green-chemically treated banana fiber/epoxy composites for enhanced mechanical and fire-resistant performance

  • 期刊论文

  • 发表日期：

    2025年12月10日

查看全文
  • Tuan Anh Nguyen,
  • Thi Thu Trang Bui,
  • Thi Huong Nguyen,
  • Xuan Thang Dam,
  • Huu Trung Dang

摘要

This study presents the first AI-assisted design and optimization framework integrating green chemical modification of banana fibers using Ca(OH)₂ for the development of high-performance, eco-friendly epoxy composites. Unlike conventional NaOH treatments, the Ca(OH)₂-based modification provides a milder and more sustainable approach that minimizes cellulose degradation while enhancing interfacial bonding. The integration of Artificial Intelligence (AI), Response Surface Methodology (RSM), and Pareto-based multi-objective optimization (NSGA-II) constitutes a novel methodological advance that unites experimental data, statistical design, and computational modeling within one framework—an approach rarely explored in natural fiber composites. At the optimized condition (20 wt% banana fibers and 5 wt% Ca(OH)₂), the composite exhibited an ≈ 80% increase in tensile strength, a ≈ 70% improvement in flexural and compressive strength, and a twofold rise in impact resistance relative to neat epoxy. Fire resistance improved substantially, with the limiting oxygen index increasing by around 20% and the burning rate reduced by more than 35%. Thermal stability was enhanced, showing ≈ 15% higher degradation onset temperature and ≈ 45% greater char yield. AI-assisted predictions were highly consistent with experimental and RSM results (R² > 0.95), and Pareto analysis identified a balanced “knee-point” composition achieving optimal synergy between strength, flame retardancy, and thermal performance. Finite Element Analysis (FEA) further verified uniform stress distribution and structural integrity at the optimal formulation. Collectively, this research introduces a new sustainable route for valorizing agricultural waste fibers through the synergy of green chemistry and artificial intelligence. The proposed AI–green optimization framework not only advances the design of multifunctional natural-fiber composites but also establishes a scalable model for developing next-generation lightweight, fire-safe, and environmentally friendly structural materials.