Predicting damping ratio of fine-grained soils using soft computing methodology

Abstract

Accurate prediction of dynamic soil properties is very important to basic understanding of soil behavior and also practical soil modeling. Shear modulus and damping ratio play a vital role in the design of geotechnical structures subjected to dynamic loads. In this study, artificial neural network (ANN), and adaptive neuro-fuzzy inference system (ANFIS) were employed for prediction of damping ratio of fine-grained soils. Most effective factors that affect this parameter include shear strain, plasticity index, and effective confining pressure. A wide-ranging database of soil element tests was used to develop an advanced model, capable of predicting soil damping ratio accurately. Results of geotechnical centrifuge tests were also involved during the training process for adequate generalization of the algorithm for future predictions. Contributions of the effective variables were evaluated through a parametric study. It was found that the ANN model developed with feed-forward back propagation (FFBF) algorithm exhibits higher performance in prediction of soil damping ratio than those developed by radial basis function (RBF) and ANFIS. The results indicate that the soft computing-based model could provide accurate and reasonable predictions, compared with the available practical charts.