Evolutionary Polynomial Regression-Based Models for the One-Dimensional Compression of Chamkhaleh Sand Mixed with EPS and Tire Derived Aggregate

Abstract

The use of expanded polystyrene (EPS) and tire derived aggregate (TDA) as soil stabilization materials has received considerable attention in the last few decades. With the increased use of these products for stabilization purposes, investigating their stress-strain behavior emerges as a crucial endeavor. One of the most widely-used methods employed for evaluating the stress-strain behavior of such mixtures under varying overburden pressures is oedometer testing. In this study, the results of oedometer tests on mixtures of sand and EPS beads, as well as sand and TDA, are used to develop Evolutionary Polynomial Regression-based models for both sand-EPS and sand-TDA mixtures. The models presented are developed based on the results of laboratory oedometer tests on the aforesaid mixtures, under different overburden pressures, different relative densities and varying contents of EPS or TDA. Subsequently, by performing multivariate parametric studies on the presented models, the effect of the simultaneous variations of the influencing parameters on the resulting settlement has been investigated. In addition, by carrying out sensitivity analyses, it was found that among the studied parameters, the relative density had the least effect on the resulting settlement for both sand-EPS and sand-TDA mixtures, while the EPS content is the most effective parameter in the settlement of sand-EPS mixtures and for the sand-TDA mixtures, both the TDA content and the overburden pressure had the most effect.