Impact of blasting parameters on vibration signal spectrum: Determination and statistical evidence

•Blast signals were clustered into several distinct clusters in spectrum by K-mean.•Student’s t-test was used to detect the discrepancy of continuous parameters.•Random simulation was conducted to interpret the influence of discrete parameters.•Result shows three parameters have effect on spectrum while the other three do not.

Research on the impact of blasting parameters on vibration signals is of significant value for guiding blast-resistant design. Previous research was primarily aimed at the impact on vibration amplitude in the time domain but rarely focused on energy distribution in the frequency domain (i.e., spectrum). Based on large amounts of blast signals from a series of events, in this study, the primary parameters that affect the vibration spectrum were determined. First, the K-means method was used to cluster all of the signals into six distinct spectrum clusters. The T test was then utilized among different clusters, to detect discrepancies in continuous parameters, including total charge, maximum charge, concrete age and distance between the explosion source and measuring point. Meanwhile, a random clustering simulation was conducted to determine whether the other two discrete parameters, i.e., the number of detonator relays and the explosion source location, influence the spectrums of vibration signals. The results show that three of the six parameters studied have a close link to the vibration spectrum, whereas the other three do not. This study also discusses how the parameters impact the occurrence and evolution of vibration signals.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slide