In the process of rock failure

a large number of low energy weak acoustic emission signals are easy to conceal the characteristics of a few high energy key signals

which is not conducive to the analysis of precursory characteristics and disaster warning. From the rock crack and destabilization warning point of view

this paper is aimed at optimum seeking of precursory key characteristic signals of acoustic emission in rock crack and destabilization

and puts forward the identification criterion and definition of key acoustic emission precursory signals of rock fracture. A new optimum seeking method for the key feature signal is proposed based on signal energy contribution rate

and the method is verified by experiments with granite

marble

coal and siltite. Experimental results show that the proposed method can effectively extract key acoustic emission signals which affect the overall stability of rock in the process of rock fracture

and this method can be applied in a wide range of applications. We also focus on the consistency test of the characteristic signals of the selected critical events from the precursory characteristics of the acoustic emission event rate

the relative quiet period

and the cumulative energy jump. The consistency test results show that the key feature signals after optimization are highly consistent with the precursory characteristics

and can effectively improve the calculation efficiency of acoustic emission characteristic parameters in late stage

and the method is feasible and effective.