Experimental Study On Acoustic Emission Characteristics Of Red Sandstone With Different Water Content Under Uniaxial Compression

In order to obtain the criteria and precursor characteristics of critical rock fracture,laboratory experiments on the characteristics of acoustic emission(AE)of red sandstone specimens with different water content under uniaxial compression are carried out.The stress threshold of red sandstone in different failure stages was determined by crack volumetric strain method.The variation of AE characteristic parameters with axial strain in high and low frequency channels is analyzed.The difference between the dominant frequency and the amplitude of the AE in the high frequency and low frequency channels is compared.Then,the precursor criteria of failure under the AE signal mode of red sandstone with different water content are established.The main conclusions are as follows:(1)The AE hit rate and count rate of the high-frequency channel at different water contents have the same trend,and the number decreases only as the water content increases.The hit rate of low frequency channel has the opposite trend with the ring counting rate,and water plays a certain “delayed” effect on this law.At the same time,it is indicated that the low-frequency AE signal is dominant during the red sandstone destruction process.(2)In the high and low frequency channels,the r(cumulative hit rate / cumulative energy rate)values have experienced the process of “rise-down-maintain low value”,and the increase of water content delays the appearance of large energy hit.(3)The analysis of RA(AE signal rise time/amplitude)value shows that low frequency shear fracture is dominant in the early stage,and low frequency tension fracture is dominant in the later stage of rock compression process.The whole process is accompanied by high frequency shear fracture.High frequency tension fracture and low frequency shear fracture increase before the peak stress.(4)Prediction of red sandstone failure and instability by AE signal(dominant frequency-amplitude)model in high and low frequency channels is helpful to improve prediction accuracy.