Estimation of ground vibration produced by blasting operations through intelligent and empirical models

The aim of this paper is to propose three predictive models namely empirical, artificial neural network (ANN), and adaptive neuro-fuzzy inference system (ANFIS) for prediction of ground vibration produced by blasting operations conducted in Gol-E-Gohar Iron mine, Iran. In this way, 115 operations were precisely monitored and related parameters of blasting were measured. Furthermore, maximum charge per delay and the distance from the blast-face were set and applied to construct the ground vibration predictive models. By assigning all data sets into training and testing, many ANFIS and ANN models were constructed. The results revealed that the proposed ANFIS model can estimate ground vibrations more accurately than other developed models. Root-mean-square error value of 4.644, for testing data set, shows superiority of the ANFIS predictive system in predicting ground vibration while they were achieved as 7.522 and 10.689 for ANN and empirical models, respectively.