Scholarlink

Abstract <p>This study introduces an ensemble of neural networks to predict the tensile strength of 3D-printed poly(lactic acid). Trained with experimental data, the ensemble uses strain, infill percentage, and infill pattern (linear, triangular, hexagonal) to predict stress. After optimizing the artificial neural network architecture via cross-validation, the final model proved highly accurate on new data, achieving a coefficient of determination of 0.99. Interpretability analyses confirmed the model learned relationships consistent with the underlying physics, with strain being the most critical factor. This validates the ensemble as an effective tool for optimizing printing parameters while reducing extensive physical testing.</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

Ensemble learning for mechanical behavior modeling of 3D-printed PLA under tension

  • 期刊论文

  • 发表日期：

    2026年2月9日

查看全文
  • Alejandro E. Rodríguez-Sánchez,
  • S. L. Rodríguez-Reyna,
  • Fidencio Tapia,
  • H. R. Acevedo-Parra,
  • J. H. Díaz-Aguilera

摘要

Abstract

This study introduces an ensemble of neural networks to predict the tensile strength of 3D-printed poly(lactic acid). Trained with experimental data, the ensemble uses strain, infill percentage, and infill pattern (linear, triangular, hexagonal) to predict stress. After optimizing the artificial neural network architecture via cross-validation, the final model proved highly accurate on new data, achieving a coefficient of determination of 0.99. Interpretability analyses confirmed the model learned relationships consistent with the underlying physics, with strain being the most critical factor. This validates the ensemble as an effective tool for optimizing printing parameters while reducing extensive physical testing.

Graphical abstract