Recognition of frequency information in acoustic emission monitoring of rock fracture

The mathematical methods of FFT and wavelet transform are used to deal with the acoustic emission (AE) frequency information of rock fracture firstly. Combined with the spectral distribution of AE signal, the main response spectrum of this lithology in specific form of fracture is selected. The AE precursor advantage spectrum and its calculation method, the most suitable short-term and optimal impending prediction parameters should be optimized, which gives a new way to monitor the disasters of rock mass engineering. In order to prove the validity of this method, the circular tunnel model is used in horizontal unloading tests in laboratory. The results show that the model edge wall around the tunnel is fractured at the moment of horizontal unloading. The energy achieves 73% between 31.25 and 62 kHz, which reflects the properties of rock materials. When it is near the breakdown (TP=2425 s), the changing coefficient of becomes from 0.5 to 15, and it has characteristics of typical disaster response. (7.8125 ~ 15.625 kHz) is the precursor advantage spectrum, and can be used as the macro rupture of short-term and impending prediction parameters combined with the sudden emergence in (90 ~ 105 kHz) and (13 ~ 20kHz) of AE main-frequency. The proposed method may provide new characteristic parameters for disasters monitoring and early warning in rock mass engineering.