Scholarlink

<p>This study aims to develop a reliable artificial neural network (ANN) framework for predicting the tensile strength of polypropylene/carbon nanotube (PP/CNT) and low-density polyethylene/carbon nanotube (LDPE/CNT) nanocomposites fabricated using microwave-assisted processing. Four key input parameters, i.e., CNT concentration, microwave power, applied pressure, and exposure time, were used to train and validate the ANN model. The predictive performance of the models was evaluated using Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and coefficient of determination (R<sup>2</sup>). The results demonstrate that both ANN models successfully captured the nonlinear relationships between processing variables and tensile strength. While the PP/CNT model exhibited stable prediction capability, the LDPE/CNT model achieved superior fitting accuracy, lower prediction errors, and improved generalization, with MSE, RMSE, MAE, and R<sup>2</sup> deviations of 7.05%, 0.64%, 1.5%, and 0.02%, respectively. Furthermore, LDPE/CNT composites showed consistently lower percentage errors (0.8–1.2%) compared to PP/CNT composites (1.5–2.0%), indicating enhanced reliability of the ANN predictions. This work provides a predictive design tool for optimizing CNT-reinforced thermoplastic nanocomposites in lightweight structural components, automotive parts, and electronic packaging applications where rapid material screening and performance-driven processing optimization are required.</p> Graphical Abstract <p></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

Computational analysis of polymer carbon nanotube composites using experiments and ANN modeling for manufacturing applications

  • 期刊论文

  • 发表日期：

    2026年7月29日

查看全文
  • Gaurav Arora,
  • Papiya Bhowmik,
  • Mohit Kumar,
  • Vinod Ayyappan,
  • Harshit Sharma,
  • Anuj Kumar Sehgal,
  • Sanjay Mavinkere Rangappa,
  • Suchart Siengchin

摘要

This study aims to develop a reliable artificial neural network (ANN) framework for predicting the tensile strength of polypropylene/carbon nanotube (PP/CNT) and low-density polyethylene/carbon nanotube (LDPE/CNT) nanocomposites fabricated using microwave-assisted processing. Four key input parameters, i.e., CNT concentration, microwave power, applied pressure, and exposure time, were used to train and validate the ANN model. The predictive performance of the models was evaluated using Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and coefficient of determination (R2). The results demonstrate that both ANN models successfully captured the nonlinear relationships between processing variables and tensile strength. While the PP/CNT model exhibited stable prediction capability, the LDPE/CNT model achieved superior fitting accuracy, lower prediction errors, and improved generalization, with MSE, RMSE, MAE, and R2 deviations of 7.05%, 0.64%, 1.5%, and 0.02%, respectively. Furthermore, LDPE/CNT composites showed consistently lower percentage errors (0.8–1.2%) compared to PP/CNT composites (1.5–2.0%), indicating enhanced reliability of the ANN predictions. This work provides a predictive design tool for optimizing CNT-reinforced thermoplastic nanocomposites in lightweight structural components, automotive parts, and electronic packaging applications where rapid material screening and performance-driven processing optimization are required.

Graphical Abstract