Acoustic Emission Behavior Of Granite In Different Characteristic Strength Stages And Recognition Of Damage Strength

With the gradual shift of resource mining to deep rock,rock engineering disasters occur frequently and rock mechanics problems become more and more complex.Among them,rock characteristic strength and damage evolution are important topics in deep rock mechanics.In order to study the crack evolution law and acoustic emission behavior of rocks with different strength ranges,three granites with different particle sizes were selected to conduct acoustic emission tests under uniaxial compression,and the surface displacement field was measured by digital image correlation method.The main research contents and conclusions are as follows:(1)The characteristic strength is measured by volume strain method.The results show that the damage strength increases with the decrease of particle size,and the ratio of damage strength to peak strength increases with the increase of particle size,indicating that the bearing capacity of rock decreases significantly with the increase of particle size after the damage strength.(2)Digital image correlation method(DIC)was used to obtain the surface displacement field of the sample in the whole process.The analysis showed that: The larger the particle size of granite sample is,the more significant the distribution of primary fracture defects in the granite sample is,and the crack growth scale is larger than that of the sample with smaller particle size near the peak strength.(3)The AE of samples with different particle sizes has a certain "relative calm period",and the AE event rate is used to characterize the rock damage variable D.When the load exceeds the damage strength,the damage variable D of samples with smaller particle size is at a relatively low level,while the damage variable D of samples with larger particle size is at a very high level.It is confirmed that the bearing capacity decreases with the increase of particle size after damage strength.(4)The acoustic emission b value showed a rising and rapidly declining trend before and after the crack initiation strength,respectively.This phenomenon indicated that the microcracks before the crack initiation strength were mainly small scale cracks,and after the crack initiation strength was mainly large scale cracks.Meanwhile the large-scale cracks are easier to form in the samples with larger particle size.(5)The sensitive identification parameters were optimized according to Spearman correlation coefficient,and the stage identification model before and after damage intensity was constructed based on the classification principle of support vector machine.The results show that the RBF kernel and PSO algorithm have the best recognition performance,and the recognition accuracy increases with the decrease of test rock particle size.