Discussion on influencing factors of dominant frequency of blasting seismic waves based on model of dimensional analysis and neural network

Because of short duration, quick mutation, large randomness and many influencing factors of blasting seismic waves, variation characteristics for dominant frequency of blasting seismic signals have no obvious law to follow. Commonly experts only take two factors including maximal section explosive and blasting center distance into consideration for convenience. Prediction model for dominant frequency of blasting seismic waves was respectively built based on theory of dimensional analysis and neural network, only considering two factors of maximal section explosive and blasting center distance. With field-acquired data, feasibility of the two models was analyzed. It is suggested that more factors be considered in the course of theoretical analysis and collecting of filed data to obtain more reasonable multifactor model for dominant frequency.