Abstract: Microseismic monitoring is an important auxiliary tool for safety and high efficiency mining at depth. However, the real-time data collected by the underground monitoring station network includes microseismic signals,blasting signals,drilling rigs,mine car machinery and equipment noise,and so on. Manual classification of different signals is not only time-consuming,labor-intensive,inefficient and subjective,but also causes great interference to the subsequent microseismic signal pickup and microseismic event location analysis. The signals collected by the microseismic monitoring system in a copper mine were used as the research object,firstly,the signal energy ratio curve within the moving time window was calculated based on the energy extremum method (EEV),and then the number of extremum points on energy ratio curve of different signals,the duration,and the duration between the extremum points and each other were selected as the main three indexes,finally,an automatic identification method of microseismic signals and blasting signals based on the above indexes was developed. MATLAB software was used to analyze and process the data collected at the copper mine,and the results showed that the method can accurately identify microseismic signals,quarry blast signals,excavation blast signals,the recognition accuracy of the signals collected for three consecutive months was 90. 7%,91. 8% and 95. 2%,respectively. The method proposed in this paper could be considered as an alternative tool for microseismic signal processing and has good engineering application value.

Key words: microseismic monitoring,microseismic signals,blast signals,signal classification,MATLAB