Predicting the Limit Void Ratios of Coarse-Grained Soils Using Artificial Neural Networks

Abstract

In this study, the prediction performance of the artificial neural network (ANN) and multiple regression (MR) models in predicting the limit void ratios of coarse-grained soils was investigated and compared. The data available in the literature were collected and used to construct both two distinct ANN-1 and ANN-2 models and two distinct MR-1 and MR-2 models: ANN-1 and MR-1 for the prediction of minimum void ratio (emin) and ANN-2 and MR-2 for the prediction of maximum void ratio (emax) of coarse-grained soils. Two basic soil graining properties such as coefficient of uniformity (Cu) and mean grain size (D50) are utilized in the simulation of the feed forward ANN models with back propagation algorithm and the MR models.The emax and emin values predicted from both ANN and MR models were compared with the experimental values taken from the literature. Moreover, five performance indices i.e. the determination coefficient, variance account for, mean absolute error, root mean square error, and the scaled percent error were calculated to examine the prediction capacity of the ANN and MR models developed in this study. The performance indices calculated indicated that both ANN models showed better performance than both MR models. It has been demonstrated that both ANN models can be used satisfactorily to predict limit void ratio values of coarse-grained soils as a rapid inexpensive substitute for laboratory techniques.