Design of blasting pattern in proportion to the peak particle velocity (PPV): Artificial neural networks approach

Ground vibration is a side effect of blasting and causes the destruction of buildings and other surrounding facilities. Different damage mitigation standards have been presented in this connection. Ground vibration is affected by parameters of blasting pattern design, distance from blasting site and explosive weight. In this research, ground vibrations data generated by 20 blasts in Sarcheshmeh copper mine, Kerman, at 47 locations have been recorded. The artificial neural network (ANN) has been trained using these peak particle velocity (PPV) data and other parameters such as block volume and explosive type employed. The trained network is capable of presenting appropriate specifications for the safe blasting pattern, considering the structure in question and its allowable vibration. The network outputs include burden, spacing and total weight of explosive used. To verify training corrections, network was tested and correlation coefficients of 0.651, 0.77 and 0.963 were obtained for the total explosive weight, burden and spacing, respectively. The effects of explosive type were studied with due regards to recorded data.

► ANN is used to design blasting pattern using PPV in Sarcheshmeh copper mine.
► Empirical data from 20 blast yielded burden, spacing and charge weight as outputs.
► CC’s of 0.651, 0.77 and 0.963 were obtained for charge wt., burden and spacing.