Intelligent Modeling of Unconfined Compressive Strength of Stabilized Clay Soil using Gene Expression Programming

Articles in Press

Document Type : Original Article

Authors

1 Associate Professor, Department of Civil Engineering, Sirjan University of Technology, Sirjan, Iran

2 M.Sc. Department of Civil Engineering Sirjan University of Technology, Sirjan. Iran

10.22034/road.2023.399936.2171

Abstract

Application of different stabilizers and additives such as cement, lime, and fly ash is one of the solutions for improving weak and problematic subgrade soils. In this study, a clay soil was stabilized using different percentages of cement and lime (0, 3, 5, 7 and 9%) and with different moisture content (optimum humidity level, wet side and dry side) and unconfined compressive strength (UCS) test was conducted at different curing times (7, 14, 21, 28 and 60 days). Then, the gene expression programming (GEP) method was employed to model the UCS of cement-stabilized and lime-stabilized clay soil. In the developed models, three variables of additive percentage, curing time and moisture content are used as the input variables to predict the UCS. The results of this study showed that the coefficient of determination (R2) for the model developed to predict the UCS of cement-stabilized clay soil is 0.935 and 0.926 for training and testing data, respectively. The R2 value for the model developed to predict the UCS of lime-stabilized clay soil is 0.911 and 0.884 for training and testing data, respectively. The results of the parametric study showed that the UCS increases with the increase in the percentage of cement and lime and the curing time and decreases with the increase in the moisture content.

Keywords

Files

Share

How to cite

Statistics