Application of artificial neural networks to predict blast-induced ground vibration in a diamond mine

Abstract

In this paper, ground vibration that is induced by blasting was predicted using data gathered from a diamond mine. Artificial neural network (ANN) is used to train the model from the 94 blast dataset using Levenberg–Marquardt algorithm, and we tested and verified the built model. Different ANN models were compared using Root mean square error (RMSE) and coefficient of determinant (R2), and the optimum ANN model was selected. Blasthole depth, blasthole diameter, burden, spacing, stemming length, powder factor, maximum charge and distance from the blast face to the monitoring point were used as input parameters. Ground vibration was the output parameter that was predicted using the built model. Processes in building the machine learning model are presented, together with prediction results and are compared against each other.