|
Abstract Because of the influence of charge parameters and rock properties，it is difficult to accurately predicate the vibration characteristics in the engineering blasting. Based on the 20 meters platform and nuclear island blasting excavation monitoring in the second phase of Ling′ao nuclear power station，Guangdong Province，the artificial neural network is adopted to predict the peak velocity of blasting vibration. In the analysis，the charge hole diameter，distance，and depth，column distance between charge holes，line of least resistance，maximum charge of single hole，maximum charge weight per delay interval，clogging depth of hole，total charge，magnitude of relative altitude and explosive distance are considered to establish the back-propagation neural network model. The prediction results through artificial neural network are more accurate than those of Sadaovsk formula.
|
|
Received: 28 April 2007
|
|
|
|
