Scholarlink

<p>The untreated natural fiber reinforced polymer composites limited their broader structural applications due to their limited mechanical performance and poor interfacial adhesion. To counteract this, the tensile strength of modified banana fiber reinforced epoxy composites incorporated with nano-silica (SiO<sub>2</sub>) at different concentrations of 0.5 to 3 wt% was predicted by utilizing an integrated Box Behnken Design (BBD) and Artificial Neural Network (ANN) model NaOH was used to treat banana fibers during 30to 120&#xa0;min and nano-silica (SiO<sub>2</sub>) was added in 0.5 to 3 wt%. The implementation of a three-factor, three-level BBD, which has 20 experimental runs was carried out and a multilayer perceptron ANN (351 architecture) was trained during the tensile strength prediction with the help of the Levenberg–Marquardt algorithm. Tensile strength ranged between 15 and 47Mpa and numerical optimization estimated tensile strength of 35.88 Mpa at 3.97 wt% chemical treatment parameter, 50.15&#xa0;min is NaOH treatment, and 0.96 wt% nano-silica content. ANN model demonstrated very high predictive power; <i>R</i> = 0.99766 and minimum MSE of 0.0020576 which is high accuracy and rapid convergence. The ANN BBD framework developed is an effective and inexpensive tool in predicting and optimization of mechanical performance of natural fibre hybrid nanocomposites. These composites can be used in the lightweight structural elements of automotive panels, interior structures and in sustainable engineering.</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

Integrated Box–Behnken Design and Artificial Neural Network modeling for tensile performance enhancement of banana fiber/nano-silica epoxy nanocomposites

  • 期刊论文

  • 发表日期：

    2026年5月19日

查看全文
  • B. N. Arathi,
  • Panjagari Kavitha,
  • Magesh Babu D.,
  • M. S. Girija,
  • T. Mythilipriya,
  • R. G. Purnima

摘要

The untreated natural fiber reinforced polymer composites limited their broader structural applications due to their limited mechanical performance and poor interfacial adhesion. To counteract this, the tensile strength of modified banana fiber reinforced epoxy composites incorporated with nano-silica (SiO2) at different concentrations of 0.5 to 3 wt% was predicted by utilizing an integrated Box Behnken Design (BBD) and Artificial Neural Network (ANN) model NaOH was used to treat banana fibers during 30to 120 min and nano-silica (SiO2) was added in 0.5 to 3 wt%. The implementation of a three-factor, three-level BBD, which has 20 experimental runs was carried out and a multilayer perceptron ANN (351 architecture) was trained during the tensile strength prediction with the help of the Levenberg–Marquardt algorithm. Tensile strength ranged between 15 and 47Mpa and numerical optimization estimated tensile strength of 35.88 Mpa at 3.97 wt% chemical treatment parameter, 50.15 min is NaOH treatment, and 0.96 wt% nano-silica content. ANN model demonstrated very high predictive power; R = 0.99766 and minimum MSE of 0.0020576 which is high accuracy and rapid convergence. The ANN BBD framework developed is an effective and inexpensive tool in predicting and optimization of mechanical performance of natural fibre hybrid nanocomposites. These composites can be used in the lightweight structural elements of automotive panels, interior structures and in sustainable engineering.