Time Frequency Feature Extraction And Recognition Method Of Rock Burst And Blasting Vibration

Microseismic signal contains abundant rock burst information, by the monitoring and data processing we can get the location of the broken rock mass and energy release. At present, it has been widely used in the field of dynamic disasters monitoring and early warning of rock burst, coal-gas outburst. But the environment is complex and changeable, and often has rock blasting operations. The microseismic signals which are picked up by the pick up device are often mixed with the unknown blasting interference signals. So it is very important that how to extract the characteristic parameter information to identify the rock burst signal and the blasting vibration signal.Based on the time-varying non-stationary characteristics of rock burst microseismic signal and blasting vibration signals, this paper compares several kinds of time-frequency analysis method performance--short time Fourier transform, wavelet transform and Hilbert huang transform, and presents the time-frequency energy feature extraction and recognition method of rock burst microseismic signal and blasting vibration signal based on ensemble empirical mode decomposition. First of all, through wavelet threshold denoising, eliminating the measured noise signals in as far as possible, to restore the true signal wave; secondly the measured signal is decomposed by EEMD into several intrinsic mode function (IMF); finally obtain each IMF energy ratio of the total signal energy as a time-frequency energy distribution of the signal. Due to the different frequency distribution of rock burst microseismic signal and blasting vibration signals, intrinsic mode function of the distribution of the energy ratio will be obtained as the characteristic parameter to identify the rock burst microseismic signal and blasting vibration signal.Based on the 80 group coal and rock burst microtremor signals and blasting vibration signal in experiment, results show that the two signals IMF energy distribution has a larger difference, the microseismic signals of coal and rock burst mainly focus on 20Hz-100Hz in IMF2, IMF3 and IMF4 frequency, the blasting vibration signal in high frequency IMF1(225Hz-375Hz) is more concentrated. For maximum differences between the two signal, to form a distinguish effective characteristic parameters.Combining IMF2, IMF3, and IMF4 band energy into a new band. The energy proportion of blasting vibration signal in IMF1 and coal and rock burst microtremor signals in IMF(2+3+4) is above 80%. In this case the difference between the two signals is most obvious. Therefore, the IMF1 and IMF (2+3+4) energy characteristic ratio is used as the characteristic index to distinguish the microseismic signal and the blasting vibration signal.