Using Intelligent Approach to Predict Blast-induced Ground Vibration

Abstract

Blast-induced ground vibration is an enormous hazardous event in the mining industry. Prediction of peak particle velocity (PPV) is very complicated due to the number of influencing parameters affecting seism wave propagation. In this paper, artificial neural network (ANN) is implemented to develop a model to predict PPV in a blasting operation. Based on the measured parameters of maximum explosive charge used per delay and distance between blast face to monitoring point, a three-layer ANN was found to be optimum with architecture 2-5-1. To investigate the suitability of this approach, ANN predictions are compared with Sodev’s predictions. Through the analysis of coefficient of determination (CoD) and mean absolute error (MAE) between monitored and predicted values of PPV, it indicates that the forecast data by the ANN model are closer to the actua1 values than Sodev’s predictor.