A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Evolutionary Polynomial Regression Model to Predict Effective Angle of Internal Friction of Fine-Grained Soils
https://orcid.org/http://orcid.org/0000-0001-9685-5641
https://orcid.org/http://orcid.org/0000-0003-3889-2270
https://orcid.org/http://orcid.org/0000-0002-1760-9838
https://orcid.org/http://orcid.org/0000-0001-7693-6866
This study evaluates correlations between the effective angle of internal friction in fine-grained soils and clay content, sand content, liquid limit, plastic limit, and plasticity index. Using a comprehensive database, linear one-dimensional regression analysis and evolutionary computing (EPR-MOGA) were applied to develop explicit mathematical models. The EPR-MOGA model was also compared with empirical models by Bjerrum and Simons, Castellanos and Brandon, Sorensen and Okkels, and Dutta et al. Model accuracy was assessed via mean absolute error (MAE), root mean square error (RMSE), mean, and coefficient of determination (R). Results indicate that one-dimensional regression provided models with low to moderate accuracy, with R values between 0.16 and 0.67, suggesting multiple factors influence the internal friction angle. However, the EPR-MOGA model demonstrated high accuracy, with R values of 0.85 and 0.92 for training and testing datasets, respectively. Additionally, the EPR-MOGA model outperformed existing empirical models, showing a superior coefficient of determination and lower error metrics. Among the empirical models, Bjerrum and Simons model ranked second due to its higher R and lower errors. In contrast, the Dutta et al. model ranked lowest in accuracy, with the highest MAE and RMSE values, reinforcing EPR-MOGA’s robustness in predictive modeling.
Evolutionary Polynomial Regression Model to Predict Effective Angle of Internal Friction of Fine-Grained Soils
https://orcid.org/http://orcid.org/0000-0001-9685-5641
https://orcid.org/http://orcid.org/0000-0003-3889-2270
https://orcid.org/http://orcid.org/0000-0002-1760-9838
https://orcid.org/http://orcid.org/0000-0001-7693-6866
This study evaluates correlations between the effective angle of internal friction in fine-grained soils and clay content, sand content, liquid limit, plastic limit, and plasticity index. Using a comprehensive database, linear one-dimensional regression analysis and evolutionary computing (EPR-MOGA) were applied to develop explicit mathematical models. The EPR-MOGA model was also compared with empirical models by Bjerrum and Simons, Castellanos and Brandon, Sorensen and Okkels, and Dutta et al. Model accuracy was assessed via mean absolute error (MAE), root mean square error (RMSE), mean, and coefficient of determination (R). Results indicate that one-dimensional regression provided models with low to moderate accuracy, with R values between 0.16 and 0.67, suggesting multiple factors influence the internal friction angle. However, the EPR-MOGA model demonstrated high accuracy, with R values of 0.85 and 0.92 for training and testing datasets, respectively. Additionally, the EPR-MOGA model outperformed existing empirical models, showing a superior coefficient of determination and lower error metrics. Among the empirical models, Bjerrum and Simons model ranked second due to its higher R and lower errors. In contrast, the Dutta et al. model ranked lowest in accuracy, with the highest MAE and RMSE values, reinforcing EPR-MOGA’s robustness in predictive modeling.
Evolutionary Polynomial Regression Model to Predict Effective Angle of Internal Friction of Fine-Grained Soils
https://orcid.org/http://orcid.org/0000-0001-9685-5641
https://orcid.org/http://orcid.org/0000-0003-3889-2270
https://orcid.org/http://orcid.org/0000-0002-1760-9838
https://orcid.org/http://orcid.org/0000-0001-7693-6866
This study evaluates correlations between the effective angle of internal friction in fine-grained soils and clay content, sand content, liquid limit, plastic limit, and plasticity index. Using a comprehensive database, linear one-dimensional regression analysis and evolutionary computing (EPR-MOGA) were applied to develop explicit mathematical models. The EPR-MOGA model was also compared with empirical models by Bjerrum and Simons, Castellanos and Brandon, Sorensen and Okkels, and Dutta et al. Model accuracy was assessed via mean absolute error (MAE), root mean square error (RMSE), mean, and coefficient of determination (R). Results indicate that one-dimensional regression provided models with low to moderate accuracy, with R values between 0.16 and 0.67, suggesting multiple factors influence the internal friction angle. However, the EPR-MOGA model demonstrated high accuracy, with R values of 0.85 and 0.92 for training and testing datasets, respectively. Additionally, the EPR-MOGA model outperformed existing empirical models, showing a superior coefficient of determination and lower error metrics. Among the empirical models, Bjerrum and Simons model ranked second due to its higher R and lower errors. In contrast, the Dutta et al. model ranked lowest in accuracy, with the highest MAE and RMSE values, reinforcing EPR-MOGA’s robustness in predictive modeling.
Evolutionary Polynomial Regression Model to Predict Effective Angle of Internal Friction of Fine-Grained Soils
Transp. Infrastruct. Geotech.
Alzabeebee, Saif (author) / Ismael, Bashar (author) / Keawsawasvong, Suraparb (author) / Alshami, Abeer W. (author)
2025-02-01
Article (Journal)
Electronic Resource
English
Experimental Study of Internal Erosion of Fine Grained Soils
| British Library Conference Proceedings | 2011
Flexibility of Fine-Grained Soils
| British Library Conference Proceedings | 1995
Optimization-Based Evolutionary Polynomial Regression
| Springer Verlag | 2019