Scholarlink

<p>Empirical models for Fiber-Reinforced Polymer (FRP)-confined concrete cylinders often show limited accuracy due to complex confinement behavior and small-sample constraints. This study develops a Physics-Informed Machine Learning (PIML) framework that embeds axial constitutive equations into ensemble learning models to enhance prediction accuracy and interpretability. Physical stress–strain laws are integrated into extreme gradient boosting and random forest loss functions through Lagrangian multipliers, while Bayesian optimization with <i>K</i>-fold validation improves model generalization. Using 310 experimental data sets, the optimized PIML model achieved a 47.7% lower root mean square error and 32% less uncertainty compared with eight empirical formulations, maintaining below 10% error for 85% of test cases. Feature analysis identified <i>ε</i><sub>h,rup</sub>, <i>s</i><sub>f</sub>, and <i>f</i><sub>1</sub> as dominant parameters consistent with FRP confinement mechanisms. The proposed PIML framework effectively bridges data-driven learning and physical theory, offering a robust and interpretable approach for intelligent design and robotic construction of FRP-confined concrete structures.</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

Physics-Informed Machine Learning framework enhancing predictive accuracy and interpretability of fiber-reinforced polymer-confined concrete cylinders

  • 期刊论文

  • 发表日期：

    2026年4月1日

查看全文
  • Wenyu Wang,
  • Kui Hu,
  • Xiaotong Du,
  • Syed Tafheem Abbas Gillani,
  • Giuseppe Carlo Marano

摘要

Empirical models for Fiber-Reinforced Polymer (FRP)-confined concrete cylinders often show limited accuracy due to complex confinement behavior and small-sample constraints. This study develops a Physics-Informed Machine Learning (PIML) framework that embeds axial constitutive equations into ensemble learning models to enhance prediction accuracy and interpretability. Physical stress–strain laws are integrated into extreme gradient boosting and random forest loss functions through Lagrangian multipliers, while Bayesian optimization with K-fold validation improves model generalization. Using 310 experimental data sets, the optimized PIML model achieved a 47.7% lower root mean square error and 32% less uncertainty compared with eight empirical formulations, maintaining below 10% error for 85% of test cases. Feature analysis identified εh,rup, sf, and f1 as dominant parameters consistent with FRP confinement mechanisms. The proposed PIML framework effectively bridges data-driven learning and physical theory, offering a robust and interpretable approach for intelligent design and robotic construction of FRP-confined concrete structures.