Scholarlink

<p>The application of Glass Fiber Reinforced Polymer (GFRP) jackets has emerged as a vital technique for enhancing the structural integrity of circular concrete columns. Despite significant research on GFRP confinement, a critical gap persists in determining the optimal GFRP thickness necessary for achieving maximum structural performance. This study develops a predictive model using machine learning techniques—specifically Decision Tree, Random Forest, Gradient Boosting, and Stacking Regressors—to estimate the ideal GFRP thickness for concrete confinement. The analysis, based on 417 specimens, reveals that column diameter, unconfined concrete strength, and fiber tensile strength are the most influential factors, with the Stacking Regressor achieving an R<sup>2</sup> of 0.98 and a Root Mean Square Error (RMSE) of 0.21&#xa0;mm in validation tests. These findings underscore the effectiveness of machine learning in optimizing GFRP thickness, contributing to safer and more efficient structural designs.</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

Machine learning-based prediction of optimal GFRP thickness for enhanced circular concrete column confinement

  • 期刊论文

  • 发表日期：

    2025年2月4日

查看全文
  • Imane Djafar-Henni,
  • Amina Sadouki

摘要

The application of Glass Fiber Reinforced Polymer (GFRP) jackets has emerged as a vital technique for enhancing the structural integrity of circular concrete columns. Despite significant research on GFRP confinement, a critical gap persists in determining the optimal GFRP thickness necessary for achieving maximum structural performance. This study develops a predictive model using machine learning techniques—specifically Decision Tree, Random Forest, Gradient Boosting, and Stacking Regressors—to estimate the ideal GFRP thickness for concrete confinement. The analysis, based on 417 specimens, reveals that column diameter, unconfined concrete strength, and fiber tensile strength are the most influential factors, with the Stacking Regressor achieving an R2 of 0.98 and a Root Mean Square Error (RMSE) of 0.21 mm in validation tests. These findings underscore the effectiveness of machine learning in optimizing GFRP thickness, contributing to safer and more efficient structural designs.