Stress–strain Behavior of Modified Expansive Clay Soil: Experimental Measurements and Prediction Models

Building on expansive soil is risky due to its high compressibility, low shear strength, and differential settlement. This study evaluates the potential of the use of sodium hydroxide (NaOH) as an additive to improve the stability of clay. For this research, the compaction properties namely Dry Density, Optimum Moisture Content, Atterberg Limits, and Unconfined Compressive Strength were performed on soil samples prepared with different percentages of NaOH (0%, 5%, 10%, 15%, and 20%). In this work, the clay sample was obtained from a highway project site near the Haspolat village in the eastern suburb of North Nicosia, Cyprus. No works of such have been done in the area. The results indicated that NaOH can be added to improve the engineering properties of the soil. Moreover, this paper presents a comparative study between an empirical equation (Quadratic model (QM)), Multilayer Feed-Forward Neural Network (MFFNN), Cascade Feed-forward Neural Network (CFNN), Radial Basis Neural Network (RBNN), and Elman neural network (ENN)) and multiple linear regression (MLR) for modeling the stress–strain behavior of CS15 (contains 85% of soil and 15% of NaOH). The coefficient of determination, root mean squared error, mean absolute error, Nash–Sutcliffe efficiency, and Willmott's index of the agreement were used to select the best predictive model. The results indicate that all the developed models are expedient in predicting the stress–strain behavior of treated soil. Furthermore, the findings demonstrated that the QM model performed well and presented high accuracy in modeling the stress–strain behavior.