A study on time-frequency analysis and automatic classification of microseismic and blasting signals underground mine

Abstract

Microseismic monitoring technology is an important means to realize the stability evaluation of ground stress, while the rapid warning of a microseismic monitoring system for rock mass stability depends greatly on the efficiency of effective microseismic signal identification. So far, the classification between blasting events and microseismic events is manually completed by data processors. This means the experience and subjectivity of the data processor would have a great influence on the effective microseismic data and the process is inefficient. Aimed at this problem and based on the monitoring data from a mine, this paper uses many blasting and microseismic signals for the study. The results show that the energy of the blasting signal is mainly concentrated in the high-frequency band (300−1000 Hz), while the energy of the microseismic signal is mainly concentrated in the low-frequency band (0−300 Hz). By analyzing the energy proportion of blasting and microseismic signal in different frequency bands, this paper provides a way to realize the automatic identification and classification of two kinds of signals, which promotes the actual realization of automatic classification of collected signals underground mine.