Scholarlink

<p>This study employed a finite-element and regression-based integrated framework to estimate California Bearing Ratio values from fundamental soil parameters: cohesion and angle of internal friction. Multiple penetration tests were simulated over a wide range of parameters, and the resulting dataset was used to establish correlations between soil properties and the bearing ratio for 2.5&#xa0;mm and 5&#xa0;mm plunger penetrations. Polynomial regression models of varying degrees were evaluated to identify optimal fits. For the 2.5&#xa0;mm penetration case, a fifth‑degree polynomial provided the most accurate representation, yielding a coefficient of determination of 0.99 for both the training and testing datasets, with root mean square errors of 1.97 and 1.93, respectively. For the 5&#xa0;mm penetration case, a third‑degree polynomial was identified as the best model, achieving coefficients of determination of 0.98 for the training set and 0.99 for the testing set, with root mean square errors of 24.03 and 13.87, respectively. The developed correlations demonstrated strong predictive capability and highlighted the nonlinear dependence of the bearing ratio on cohesion and the angle of internal friction, thereby providing a validated computational approach for CBR evaluation.</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

Estimation of California Bearing Ratio Using Soil Parameters: A Finite Element Coupled Regression Modelling Approach

  • 期刊论文

  • 发表日期：

    2026年7月22日

查看全文
  • Kuldeep Joshi,
  • Archana Bohra Gupta

摘要

This study employed a finite-element and regression-based integrated framework to estimate California Bearing Ratio values from fundamental soil parameters: cohesion and angle of internal friction. Multiple penetration tests were simulated over a wide range of parameters, and the resulting dataset was used to establish correlations between soil properties and the bearing ratio for 2.5 mm and 5 mm plunger penetrations. Polynomial regression models of varying degrees were evaluated to identify optimal fits. For the 2.5 mm penetration case, a fifth‑degree polynomial provided the most accurate representation, yielding a coefficient of determination of 0.99 for both the training and testing datasets, with root mean square errors of 1.97 and 1.93, respectively. For the 5 mm penetration case, a third‑degree polynomial was identified as the best model, achieving coefficients of determination of 0.98 for the training set and 0.99 for the testing set, with root mean square errors of 24.03 and 13.87, respectively. The developed correlations demonstrated strong predictive capability and highlighted the nonlinear dependence of the bearing ratio on cohesion and the angle of internal friction, thereby providing a validated computational approach for CBR evaluation.