Predicting blast-induced ground vibration using various types of neural networks

Prediction of vibration is very important in mining operations as well as civil engineering projects. In this paper, multi layer perceptron neural network (MLPNN), radial basis function neural network (RBFNN) and general regression neural network (GRNN) were utilized to predict ground vibration level in a Sarcheshmeh copper mine, Iran. It was observed that the MLPNN gives the best results. For this technique root mean square error and coefficient of correlation were found 0.03 and 0.954, respectively. Sensitivity analysis showed that distance from the blast, number of holes per delay and maximum charge per delay are the most effective parameters in making ground vibration in the blasting operation.