Prediction of Elastic Modulus of Intact Rocks Using Artificial Neural Networks and non-Linear Regression Methods

The purpose of this study is developing of a model for estimation of elastic modulus of intact rocks. Mechanical rock excavation projects require static modulus of elasticity (E) of the intact rock material. High-quality core specimens of proper geometry are nedded for the direct determination of this parameter. However, it is not always possible to obtain suitable specimens from highly fractured and/or weathered rocks for this purpose. Therefore, models prediciting E based on rock index tests and intact rock properties have become alternative methods. For this reason, in this study, the advantage of artificial neural network (spacifically multilayer perceptron and radial basis function networks) and non-linear regression were examined. In addition to correlation coefficient (R), root mean square error (RMSE), mean absolute error (MAE) and mean bias error (MBE) were also used for evaluation of predition accuracy of both ANNs and non-linear regression methods between the measured and predicted parameter values. Finally, the results of this study indicated that MLP-ANN had better performance in prediction of elastic modulus of rocks rather than RBF-ANN and non-linear regression models.