Some Investigations on the Influence of Blast Design Parameters on the Prediction of Ground Vibrations--an Artificial Neural Network Approach

摘要

The present investigation deals with the effect of blast design parameters, such as, hole diameter, hole depth burden, spacing, charge length, explosive per hole, velocity of detonation of explosive and density of explosive on the prediction of ground vibrations. Artificial neural network (ANN) model has been emploved to analyze and predict the output parameters. A back propagation neural network is used to predict the ground vibrations. Predictions from ANN have been compared with actual values observed from the field and are very close with the field results. Three cases have been investigated by reducing the number of input parameters and in each case, the ground vibrations are predicted. In the second case the obtained results matched very closely with the measured values from the field data. Thus, the developed ANN model can be applied for analyzing the influence of various blast design parameters on the prediction of ground vibrations.