Scholarlink

<p>The biodegradability of synthetic‑filled 3D‑printed products remains a concern, posing environmental challenges and prompting growing interest in natural fiber‑reinforced composites as more sustainable alternatives to conventional filaments. In this work, an AI‑based predictive framework is presented to support the use of paper‑based and lignocellulosic composite materials in additive manufacturing by identifying trends in key mechanical and surface‑related properties. A hybrid modelling approach combining Artificial Neural Networks (ANN), Gene Expression Programming (GEP), and Support Vector Machines (SVM) is used to predict tensile strength, Young’s modulus, and surface roughness based on material and processing parameters. The results indicate that surface roughness can be predicted more consistently than tensile strength and Young’s modulus, as it is more strongly influenced by controllable printing parameters, while stiffness‑related properties show greater variation due to inherent material non‑uniformity. Overall, the proposed framework is intended as a decision‑support tool to guide material selection and process-parameter studies for natural fiber‑based composites in sustainable additive manufacturing.</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

AI-guided structural optimization of natural material-based additive manufacturing for packaging applications

  • 期刊论文

  • 发表日期：

    2026年4月29日

查看全文
  • Darsan Rajeevan Sheela,
  • Hari Narayanan A G

摘要

The biodegradability of synthetic‑filled 3D‑printed products remains a concern, posing environmental challenges and prompting growing interest in natural fiber‑reinforced composites as more sustainable alternatives to conventional filaments. In this work, an AI‑based predictive framework is presented to support the use of paper‑based and lignocellulosic composite materials in additive manufacturing by identifying trends in key mechanical and surface‑related properties. A hybrid modelling approach combining Artificial Neural Networks (ANN), Gene Expression Programming (GEP), and Support Vector Machines (SVM) is used to predict tensile strength, Young’s modulus, and surface roughness based on material and processing parameters. The results indicate that surface roughness can be predicted more consistently than tensile strength and Young’s modulus, as it is more strongly influenced by controllable printing parameters, while stiffness‑related properties show greater variation due to inherent material non‑uniformity. Overall, the proposed framework is intended as a decision‑support tool to guide material selection and process-parameter studies for natural fiber‑based composites in sustainable additive manufacturing.