Interrelation analysis method of acoustic emission for failure prediction of brittle failure based on vibration mechanics

Brittle failure of rock masses poses a great threat to safety and property in underground engineering and slope engineering. It is necessary to develop effective methods to evaluate the damage degree of rock masses and make predictions of sudden failure. In this paper, a new method is proposed to extract damage information about rocks from acoustic emission (AE) data, and three indicators are obtained: the average local maxima of the frequency response along the propagation path (ALM), the average correlation value (ACV), and the standard deviation of the main frequencies of different signals (SD). Lead break tests and uniaxial compression tests are conducted to verify the applicability and effectiveness of the indicators. The results show that the frequency of the elastic wave will change after the elastic wave propagates along a path. When final failure state is approached, the ACV often decreases, while the ALM and SD increase. The p value based on a comprehensive analysis of the three indicators is convenient for estimating the damage degree, and it performs better than the traditional b value in predicting final failure when experimental conditions are suitable. The method proposed in this paper provides a new way to analyze AE data by focusing on the differences in waves from different sensors instead of the rupture sources. The indicators can provide precursor information for the failure and damage degree of rock masses.