Relationship between the Intrinsic Properties of Sands and the Parameters of Mathematical Particle Size Distribution Models for Predicting Geotechnical Quantities

This work characterizes the relationships between the intrinsic properties of sands and the parameters of four mathematical models that best simulate the experimental curves and geotechnical properties of sands used in construction. Origin.Pro.2019" software was used to smooth the grading curves, define the parameters of the mathematical models and link them to the geotechnical data. To achieve this objective, the correlations between the intrinsic properties of the sands are developed using mathematical models with the highest coefficient of determination (R2) and the lowest statistical coefficient (χ²). The correlations used are those with a coefficient of determination greater than or equal to 0.9. The results obtained show that the models used provide a good description of the experimental curves. The model parameters are correlated with the granulometric fractions and the geotechnical parameters. The evolution of the points expressing the parameters of the Gaussian and exponential models (A1, Xc, A, W, Yo) and the parameter (t1) as a function of seven randomly chosen geotechnical quantities, are polylinear and linear fits, respectively. This study is important for predicting a geotechnical quantity from a modelled grading curve, by solving the mathematical expressions of the models used.