Research on the Characteristics of Acoustic Emission Activities of Granite and Marble under Different Loading Methods

The basic process of acoustic emission is that material under the action of external force, stress concentration acts on internal defects when the stress exceeds the limit that can be tolerated, the internal defects are damaged and deformed, the bearing capacity is lost, and the stress is relaxed; the energy released by the instability of the internal structure during the failure process is released in the form of elastic waves [1-3]. Acoustic emission detection technology is to use acoustic emission instruments to detect the acoustic emission signal generated by the change in external conditions; changes in the internal structure of materials are detected by analyzing the measured acoustic emission signals [4, 5]. Since the acoustic emission signal measured by the acoustic emission technology is emitted by the internal defects of the material, even if the defects of the same nature show different acoustic emission characteristics under different stress states [6-8], therefore, the acoustic emission signal contains extremely rich defect evolution information; using this information, the damage evolution of the material under loading can be analyzed accurately and in detail [9, 10]. In recent years, underground geotechnical engineering has appeared more and more in urban construction, but due to the shortage of urban construction land resources, most of the projects are located in building complexes, and there are many unstable factors during the construction process; it affects people’s normal life and personal safety, so it is necessary to carry out reasonable tests on the failure process of geotechnical engineering. Compared with other nondestructive testing methods, acoustic emission technology is an unconventional dynamic nondestructive testing method. Because of its unique dynamic and real-time testing performance, it can continuously monitor the internal damage and deformation of rock mass, so it has advantages that other testing methods do not have [11, 12]. With the maturity of rock acoustic emission technology, more and more acoustic emission is applied in various geotechnical engineering: to determine the stability of underground oil and gas storage [13-15] and measure the state of rock mass in-situ stress [16-19]; track and forecast the safety dynamic changes of the mine pit [20-22]; detecting internal defects of large concrete bridges and steel bridges to determine their safety [23-25]; there are successful cases in real-time monitoring of rock and soil stability of dams, slopes, large building foundations, and so on [26-28].

At present, acoustic emission detection technology has attracted extensive attention from scholars, and a series of research results show that acoustic emission detection technology has great potential in the field of engineering practice. Literature [29-31] studied the effect of loading rate on rock acoustic emission parameters and rock mechanical properties. Literature [32-34] studied the dynamic characteristics of the AE b value in the process of rock fracture and propagation and the AE Doppler effect in rock friction and sliding. Literature [35, 36] discussed the effect of cyclic loading tests on the evolution of acoustic emission signals during rock fracture. Literature [17, 37] studied the correlation between acoustic emission and electromagnetic launch monitoring methods and stress drop and successfully applied the research results to the monitoring and forecasting of rock bursts. Literature [38, 39] applied two test methods of uniaxial compression and triaxial compression; the influence of high temperature on the strength and damage state of granite was found by analyzing the acoustic emission signal of the granite damage process, and the time and deformation behavior of each stage of fracture were predicted. Literature [40-42] concluded that the deformation and fracture of the loaded coal rock mass and the acoustic emission signal were discontinuous according to the experimental study of coal rock fracture acoustic emission; rather, it is paroxysmal; during the rupture of the loaded coal rock mass, the acoustic emission signal shows a gradually increasing trend, which is of great significance for predicting the dynamic phenomenon of coal and rock disasters.

German scientist Kaiser discovered an important theory in 1950—the Kaiser effect; it reflects the phenomenon that during the repeated loading of the material, no obvious acoustic emission occurs before the original maximum load is applied. Then, with the deepening of acoustic emission research, the Felicity study found a phenomenon opposed to the Kaiser effect—the Felicity effect, that is, when the material is repeatedly loaded, the phenomenon of obvious acoustic emission occurs before the repeated load reaches the original maximum load. Subsequent studies by scholars have shown that [43-45], as the stress on the rock increases, the Kaiser effect fails, and the rock exhibits a Felicity effect, after which the rock reaches peak strength and fails with an accumulation of damage. Literature [46, 47] conducted penetration tests of different materials under different stresses, the characteristics of the transformation of the acoustic emission phenomenon from the Kaiser effect to the Felicity effect. Literature [48-50] found that the main reason for the Felicity effect of rock under compression is plastic deformation. Literature [51] studied the acoustic emission effect of anisotropic phyllite, and it was found that it is easier to determine the Kaiser effect of rocks using uniaxial compression tests. Literature [52-54] conducted uniaxial compression tests and acoustic emission tests on different rock samples under different loading stress levels and pointed out that the existence of the Caesar effect can be judged by analyzing the stress points. Literature [55-57] based on the cyclic loading and unloading test of marble and granite, and the dynamic changes of the combination of AE count rate, Felicity ratio, and D value in different stages of rock fracture were studied. Literature [58-60] carried out acoustic emission monitoring on rock samples of mine boreholes; based on fractal theory, a relatively simple method for determining the position of the Kaiser point is established.

Combined with the research status of rock acoustic emission, it is found that there is a lack of systematic understanding of various rock acoustic emission characteristics in the academic community, which makes it difficult to apply acoustic emission technology to various engineering practices. In this paper, the acoustic emission characteristics of two different rock samples of marble and granite under uniaxial compression were analyzed through the indoor acoustic emission test; during the test, the acoustic emission instrument was used to monitor the acoustic emission activity of the two kinds of rocks in the whole process of damage and failure under uniaxial compression; by analyzing the acoustic emission data recorded in the test, the fitting diagram of the acoustic emission parameters, stress and strain, was drawn, and the similarities and differences of the acoustic emission signal characteristics corresponding to the different stages of the two types of rock failure were analyzed. Based on the study of acoustic emission events, the relationship between rock damage and acoustic emission is analyzed, and the damage constitutive model is established. Using cyclic loading to study the Kaiser effect of different rock samples by analyzing the relationship between the Felicity ratio of different rocks and the stress level, the stress range of the Kaiser and Felicity effects of the rock sample is determined, and understand the fatigue failure mechanism of rock, discuss the AE precursor characteristics of rock rupture and instability in geotechnical engineering practice, and provide an experimental basis for AE early warning of geotechnical engineering disasters.

Two different rocks, marble, and granite were used in this experiment. To achieve the expected goal of the test, the lithology, strength, uniformity, and so on were mainly considered in the sampling process to make the collected rock samples as representative as possible for comparison in the test. The strength of marble is relatively low and relatively uniform, the texture of granite is hard, and the mineral grains can be seen. The samples were processed into cylindrical standard specimens of Φ50 mm × 100 mm. To ensure that the acoustic emission probe is in good contact with the sample, in this test, high-temperature vacuum silicon grease was used as the coupling between the sensor and the contact surface of the material; it has the characteristics of a small sound attenuation coefficient, moderate adhesion, not easy to dry, good stability and uniformity, and so on; it can tightly adhere to the surface of the material, and the sensor and the material cannot be separated due to the tight adhesion, improve the resolution of the sensor, reduce the loss of acoustic energy, and it can effectively reduce unnecessary errors in the test.

Rock is a kind of material with acoustic anisotropy and a small sound wave attenuation coefficient; the frequency range, amplitude range, noise signal, and the shape of rock samples of the acoustic emission signal of such materials are considered comprehensively. This test uses the SAEU2S multi-channel acoustic emission system, the loading system is a CSS-waw 2000dl electro-hydraulic servo test system, and the maximum test force of the loading system is 2000 KN. The sensor uses an SR150 resonant high-sensitivity sensor and a PAI-type broadband preamplifier. The test equipment is shown in Figure 1.

In this experiment, six cylindrical marble and granite samples with a size of 50 mm × 100 mm were used, simultaneously starting the loading system and the acoustic emission system to uniaxially compress the sample. Axial strain-controlled loading was used at a loading rate of 0.2 mm/minutes. Set the peak definition time to 50 seconds, the hit definition time is 200 microseconds, and the latch Time is 300 microseconds.

During the test, parameters such as acoustic emission signal, load, deformation, and time of the rock sample in the whole process of deformation and failure were measured. Comparing the number and rate of AE events for each rock sample, summarize the acoustic emission characteristics of various lithological rocks. A stress-time curve is plotted from the recorded stress data, fitting with the AE event rate and the number of events to analyze the AE signal characteristics corresponding to different stages of rock failure, a comparative study of acoustic emission activity characteristics in different rock failure processes.

In this experiment, five cylindrical marble and granite test blocks with a size of 50 mm × 100 mm were used, in the way of cyclic loading, stress-controlled loading is adopted, and the loading rate is 0.5 MPa/s.

The marble loading process is as follows: loading at a rate of 0.5 MPa/s to 13 MPa, K. The marble is loaded at a rate of 0.5 MPa/s, with loading levels of 13 MPa, 26 MPa, 39 MPa, 52 MPa, 65 MPa, 78 MPa, and 91 MPa…that is, 13 MPa is added each time to stabilize the pressure for 30 seconds, then unloaded at the same rate to 0 MPa, and then stabilized for 30 seconds, and the cycle is continued until the test block is broken. The loading process of granite is as follows: at a rate of 0.5 MPa/s, the loading stress levels are 12 MPa, 24 MPa, 36 MPa, 48 MPa, 60 MPa, 72 MPa, 84 MPa, 96 MPa, 108 MPa, and 120 MPa…that is, increase by 12 MPa each time, and stabilize the voltage for the 30 seconds. Unload to 0 MPa at a rate of 0.5 MPa/s, and then stabilized the voltage for 30 seconds; this cycle is repeated until the test block is destroyed. Record the number of acoustic emission events, event rate, stress, strain, and other data of marble and granite rock samples, draw a stress-strain diagram to fit the number of events and event rate, and investigate the Kaiser effect and Felicity effect of different rock samples.

The degree of irreversibility during cyclic loading can be expressed by the Felicity ratio. In a loop, the Felicity ratio of the acoustic emission process is defined as:

in the formula:

F(k)—Felicity ratio in the kth cycle;

P(k+1)—stress level when the effective acoustic emission is restored during the k+1th loading process, MPa;

P(k)—the maximum stress level reached by the kth loading, MPa.

The Felicity ratio can describe the irreversibility of the rock acoustic emission process in more detail. According to the definition of the Kaiser effect, the Kaiser effect is strictly valid only when F(k) ≥ 1; when F(k) < 1, the Kaiser effect can be considered invalid. However, due to certain errors in the experiment, when studying the Kaiser effect, it is generally believed that the Kaiser effect is still valid as long as the Felicity ratio is not less than 0.9.

In practical engineering, the phenomenon of constant load force of rock mass is common; this experiment will further analyze the relationship between the acoustic emission parameters in different rock failure processes through the study of the whole process acoustic emission test of graded loading.

Five cylindrical marble and granite test blocks with a size of 50 mm × 100 mm were used, and loading is carried out in a step-by-step manner using stress control. The marble loading process is to load at a rate of 0.5 MPa/s to 33 MPa and maintain a constant voltage for 60 seconds, then load it to 59 MPa at the same rate, keep the constant voltage for 60 seconds, continue to load to 105 MPa, hold the constant voltage for 60 seconds, and continue to load until the test block is destroyed. The loading process of granite is at a rate of 0.5 MPa/s, and the loading stress levels are 22, 40, and 71 MPa, and the pressure was stabilized for 60 seconds and finally loaded until the test block was destroyed. Record the acoustic emission time, event rate, stress, strain, and other data of each rock test block, the stress time diagram is drawn to fit the event rate and several events, and the failure mechanism of the rock is further studied.

It can be seen from Figure 2 that marble d-4 showed a very high acoustic emission rate at the beginning of loading, lasted for about 50 seconds, and then decreased rapidly; there is no acoustic emission phenomenon between 50 seconds and 100 seconds. A large number of acoustic emission phenomena appear in marble at the beginning because the internal particles are uneven, and some soft particles are destroyed first, causing the internal collapse or surface collapse of the rock sample, resulting in a certain amount of microcracks. The acoustic emission appears relatively stable from 100 seconds to 450 seconds; only two higher acoustic emission phenomena occur at 100 seconds and 300 seconds; the compacted microcracks in the rock sample are gradually closed, and the internal particle strength is high, not easy to be crushed, no new microcracks are generated, so no acoustic emission is generated; this stage is the compaction stage. After the plateau period, the rock enters the elastic stage, and as the load gradually increases, the AE event rate also increased rapidly, with several AE rate peaks at 65% of peak load. After 550 seconds, the acoustic emission rate began to decrease, and at 76% of the peak intensity, the acoustic emission rate reached a trough and then continued to rise. In the late elastic stage and the early plastic stage, the load reaches 89% of the peak load at 650 seconds, and the acoustic emission rate reaches the maximum. Then, until the rock sample breaks, the rock enters the post-peak failure stage, the load dropped sharply, and the acoustic emission activity also decreased, but the acoustic emission rate was still high. Throughout the whole loading process, the marble AE activity level is generally high. Compared with d-4, marble d-5 has a certain amount of acoustic emission events in the early stage of loading; that is, in the compaction stage, it only lasts about 25 seconds, followed by a very small amount of acoustic emission activity from 25 seconds to 360 seconds, and no acoustic emission occurs between 360 seconds and 425 seconds; this shows more clearly than d-4 that the grains inside the marble are uneven; after compression, the interior of the rock sample collapses or collapses, and a small amount of acoustic emission events occurs during the compaction stage; with the uniform compaction of the rock sample, there is no acoustic emission event. After 425 seconds, the load gradually increases after entering the elastic stage, the acoustic emission event rate also increased, and the acoustic emission activity was very active, with several large event rates occurring in the middle. However, after entering the plastic stage, when the load reaches 90% of its peak strength, the AE event rate decreases significantly, and a relatively quiet period of acoustic emission events occurs. This quiet period appears before the rock sample failure, which may be a precursor to rock failure and instability. It is worth noting that, after the peak intensity, the post-peak failure phase, there are three sudden drops in stress while the acoustic emission rate reaches the maximum. Although both d-4 and d-5 belong to marble, the overall law of acoustic emission is consistent; however, the acoustic emission characteristics of the two rock samples are different. This further shows that the acoustic emission of rocks is affected by various factors.

It can be seen from Figure 3 that a small amount of acoustic emission appears in the initial stage of granite h-4 loading, which is the same as that of marble, granite also goes through a compaction phase. Very few acoustic emissions followed for a long time. At 410 seconds, the acoustic emission rate began to increase, indicating that the rock entered the elastic stage. When the load increases sharply, the acoustic emission rate also increases, and the acoustic emission rate reaches the maximum at 66% peak load. At 75% peak load, the acoustic emission rate begins to decrease; after 83% peak load, the acoustic emission quiet period appeared in the middle and late plastic stage; after the peak load, the load decreases rapidly with very few acoustic emission events. Compared with marble, the peaks and valleys of acoustic emission events are more obvious in the whole loading process. The acoustic emission activities of granite h-5 and h-4 are similar, and the only difference is that the acoustic emission events increase faster in the elastic phase of acoustic emission.

Both rock acoustic emissions experienced an initial compaction zone, rising zone, peak zone, and descending zone. The performance of marble and granite at different stages is compared and analyzed in Figures 2 and 3. Initial compaction zone: in the initial stage of loading, the uneven particles inside the marble lead to the destruction of softer particles, a certain amount of microcracks are generated, and a high acoustic emission rate occurs. However, the granite of the same period only produces a small amount of acoustic emission. With the continuous loading, the two kinds of rocks enter a stable period, and basically, no acoustic emission phenomenon occurs. Rising zone (elastic stage): After the steady phase, the rock enters the elastic stage, many microcracks are generated inside the rock, and the AE event rate increases linearly. At peak loads of 60%–70%, granite has a peak in acoustic emission rate; in contrast, the acoustic emission rate peak of marble occurs between 80% and 90%. Peak zone (plastic stage): Marble enters the plastic stage; when the load reaches 89% of the peak load, the acoustic emission rate reaches its maximum, and subsequently, the rate of acoustic emission events decreased significantly, and a quiet period of acoustic emission appeared, which lasted for a short time. Different from marble, the acoustic emission rate of granite reaches its maximum at 66% peak load, then there is a long period of quiet period of acoustic emission in the middle and late plastic stage after 83% peak load, which is more obvious. Descending zone (destruction stage): After the marble enters the post-peak failure stage, the load drops sharply, and the acoustic emission activity decreases, but the acoustic emission rate remains high; the AE event rate trend of granite after peak intensity is similar to that of marble; however, in the whole loading process, compared with marble, the peak and trough of the acoustic emission event rate are more obvious.

Comprehensive analysis in Figures 4, 5 and 6, we can more clearly see the four stages of rock acoustic emission. A-B is the initial compaction stage; microcracks, fissures, and inhomogeneous particles in the rock generate a small amount of acoustic emission events after initial stress, and no acoustic emission event occurs after compaction; B-C is a stable stage, and the acoustic emission also increases with the continuous increase of the load; C-D is the rising stage, and new microcracks are constantly generated inside the rock; at the same time, the microcracks continue to expand and penetrate to form larger cracks, resulting in a large number of acoustic emission events; after D-E, the rock enters the destruction stage, and it can be seen that the marble and the other two kinds of rocks show different properties during the failure stage; at this stage, the acoustic emission of marble does not gradually become calm like that of granite. Instead, a large number of acoustic emission events are continuously generated until the bearing capacity is completely lost; this shows that marble has high strength and is less brittle than granite. After the peak strength, the rock still has a certain bearing capacity, and this is clearly seen from the three stress drops after the d-5 peak strength; therefore, the acoustic emission events continue to increase after the marble is destroyed.

Based on the continuum damage mechanics and Weibull distribution density function, Zhang Ming [61] established the rock damage model represented by the number of acoustic emission events.

In the formula:

σ is stress;

E is the initial nondestructive elastic modulus of rock;

N is the cumulative number of acoustic emission events when the infinitesimal area A′ is destroyed;

N_m is the cumulative number of AE events when section area A is damaged.

From this constitutive equation, the curve fitting diagram of theoretical stress-strain and actual stress-strain of different rock samples at failure stages under uniaxial compression is plotted, as shown in Figure 7.

The acoustic emission results of microcrack propagation in the rock under uniaxial axial pressure provide information on rock failure. The analysis of acoustic emission results reveals the following: In the elastic stage, crack closure and linear elastic deformation occur in brittle rocks. Microcrack propagation does not occur, and the change of damage variable can only be caused by microcrack propagation; in the nonlinear stage, cracks generate and grow steadily, and crack damage and unstable growth occur. Microcrack propagation occurs, damage begins to increase, and the growth rate is relatively fast; at the failure stage, when the rock reaches the strain fatigue state, the damage continues to increase, and the rock is finally destroyed.

On this basis, the damage characteristics of different rock samples are studied, and the damage constitutive model of rock is established. It is found that the theoretical stress-strain curve fitted by acoustic emission event number parameters can reasonably reflect the actual stress-strain characteristics of rock, which verifies the correctness of the damage constitutive model. The simulated and experimental stress-strain curves are very similar, indicating that the microcrack damage model is very suitable for the establishment of a brittle rock model under uniaxial axial pressure. This model also well reflects the strain fatigue phenomenon. The test shows that the damage to rock during elastic deformation is almost zero. Once the nonlinear deformation stage is reached, the damage increases gradually, which indicates that when the external load exceeds the elastic limit of the rock, microcrack propagation occurs. When the external load increases, the rock damage rate gradually increases. When the peak strength is reached, the microcrack propagation increases rapidly, and the damage to rock is enhanced, which is consistent with the acoustic emission results. It verifies that the acoustic emission of rock has an inevitable relationship with damage.

Combining the failures of the two rock specimens in Figure 8, it can be found that granite has high strength and dense structure; in the initial stage of loading, there is no acoustic emission activity for a long period; however, with the continuous increase of stress, the acoustic emission number suddenly increases sharply, concentration occurs before the failure, and the sample is broken with a cracking sound during failure; marble has high strength but is not as brittle as granite and exists joints; at the initial stage of loading, a certain amount of acoustic emission events are generated; during the loading process, the splitting sound caused by the caving of the rock sample can be heard, two loud bangs were heard during the destruction, two main cracks can be seen penetrating, the rock sample was split and damaged, and fragments jumped out at the same time. Compared with granite, marble has more AE events and longer duration.

Figures 9 and 10 are histograms of acoustic emission events and stress-time curves for the two rocks, Event Rate versus Number of Events Histograms and Stress-Time Graphs.

Analysis of Acoustic emission rate Parameters at Different Stages of Cyclic Loading of marble and granite, determining the stress level at which effective acoustic emission is restored for each cycle, and the Felicity effect ratios of the two rocks are obtained, as shown in Tables 1 and 2.

From the relationship between the ratio of marble and granite and the relative stress level in Figure 11, we can analytically see that the rock is in the initial stage of damage development before the stress reaches 20% of the peak strength. At this stage, the Felicity ratio is less than 0.9, showing the Felicity effect; at this time, because this stage is in the low-stress stage, the original microcracks in the rock are closed and cracked by the initial stress, resulting in more acoustic emission events. Moreover, the Felicity ratio fluctuated greatly, and the memory ability of stress was not stable enough. When the stress is between 20% and 70% of the peak strength, the rock is in a stable stage of damage development. With the increase of the stress level, after the initial adjustment and compaction stage of loading, the internal cracks of the rock expand steadily, the acoustic emission rate is small and remains stable, and the Felicity ratio is kept between 0.9 and 1.0, with a small fluctuation. This indicates that the damage of the rock at this stage is continuously accumulated when it reaches the maximum stress of the last cycle, and there is an obvious Kaiser effect, and the ability to remember the stress is very accurate. In the stable stage, the rock endured certain stress during the previous cyclic loading process, and the internal micro-elements have been ruptured, resulting in irreversible damage; when the stress has not reached the last maximum stress, there will be no cracking and no acoustic emission. After the stress exceeds 70% of the peak strength, the microcracks in the rock rapidly expand and penetrate, resulting in a large number of macrocracks until failure and instability, the Felicity ratio dropped to around 0.8, at which point the number of AE events produced a large number of AE events before reaching the maximum stress of the previous cycle, the rock has an advanced memory capacity for stress, and the Kaiser effect fails.

In summary, during the cyclic loading process of rock, the law of damage development is divided into three stages: initial stage, stable stage, and instability stage. When the stress of the two rocks is between 20% and 70% of the peak strength, the Kaiser effect is obvious, and the memory ability of the rock is accurate and stable; after the stress exceeds 80% of the peak value, the Kaiser effect fails, and an advanced memory capacity for stress occurs, resulting in the Felicity effect. The Kaiser effect does not appear before the stress is 20%, and the memory capacity of the stress is not stable enough.

Different from the large acoustic emission rate of marble at the beginning of monotonic loading, there is basically no acoustic emission event of marble at the initial stage of loading. With the increase of stress, the acoustic emission rate gradually increases. The acoustic emission rate rapidly decreases during the dead load period and maintains a very low level, but with the next cycle of loading, the acoustic emission rate returns to a higher level. During the first three loading periods, the acoustic emission rate of marble is kept at a certain level. Before the failure and instability, although the acoustic emission rate of marble during the dead load period is reduced, it is higher than the previous two times. At this time, the stress reaches a higher level, and the internal damage of rock has accumulated to a very high degree. Before the failure, the acoustic emission rate rapidly increases to the peak level until failure, and the acoustic emission rate rapidly decreases until disappearance. However, granite produces a large amount of acoustic emission in the early stage of loading. The acoustic emission law in the loading period and constant load zone is the same as that of marble; however, the acoustic emission rate of granite will gradually decrease with the increase of each level of load from the beginning of loading until the failure; before the peak load, the acoustic emission phenomenon disappears and enters a quiet period, the rock then failed, and the acoustic emission rate was low before the failure. Marble and granite both have a quiet period of acoustic emission before rock instability, and the higher the rock strength, the longer the quiet period lasts; the more obvious this shows that in practical engineering with complex forces, taking the quiet period before rock failure and instability as a precursor feature of rock mass instability has extremely important application value for prediction and prediction before disasters.

The difference in loading rate affects the rate of change of acoustic emission, and the increase in loading rate leads to an aggravation of rock damage. Comparing and analyzing Figures 2, 3 and 12, it can be seen that there is a 30% difference between the peak strengths of marble and granite when they fail under two different loading rates: monotonic loading and cyclic loading.

1. The acoustic emission of marble and granite has experienced an initial compaction zone, rising zone, peak zone, and falling zone. Compare and analyze the performance of the two rocks at different stages: (a) Initial compaction zone: in the initial stage of loading, the uneven particles inside the marble lead to the destruction of softer particles; a certain amount of microcracks are generated, and a high acoustic emission rate occurs. However, the granite of the same period only produces a small amount of acoustic emission. With the continuous loading, the two kinds of rocks enter a stable period, and basically, no acoustic emission phenomenon occurs. (b) Rising zone (elastic stage): After the plateau, the rock enters the elastic stage, many microcracks are generated inside the rock, and the AE event rate increases linearly. At peak loads of 60%–70%, granite exhibits a high-AE peak, while that of marble occurs between 80% and 90%. (c) Peak zone (plastic stage): Marble enters the plastic stage; when the load reaches 89% of the peak load, the acoustic emission rate reaches its maximum; subsequently, the rate of acoustic emission events decreased significantly, and a quiet period of acoustic emission appeared, which lasted for a short time. Different from marble, the acoustic emission rate of granite reaches its maximum at 66% peak load, then there is a long period of quiet period of acoustic emission in the middle and late plastic stage after 83% peak load, relatively obvious. (d) Drop zone (destruction stage): After the marble enters the post-peak failure stage, the load drops sharply, acoustic emission activity decreases, but the acoustic emission rate remains high; the AE event rate trend of granite after peak intensity is similar to that of marble; however, in the whole loading process, compared with marble, the peak and trough of the acoustic emission event rate are more obvious. It can be seen that the duration of each acoustic emission stage of different rocks is related to the properties of the rocks.

2. Marble and granite both have a quiet period of acoustic emission before rock instability, and the higher the rock strength, the longer the quiet period lasts; this shows that in geotechnical engineering, taking the quiet period before rock failure and instability as a precursor feature of rock mass instability has extremely important application value for prediction and prediction before disasters.

3. During the cyclic loading process of rock, the damage development law is divided into three stages: initial stage, stable stage, and instability stage. When the stress of the two rocks is between 20% and 70% of the peak strength, the Kaiser effect is obvious, and the memory ability of the rock is accurate and stable; after the stress exceeds 80% of the peak value, the Kaiser effect fails, and there is an advanced memory ability for stress, and the Felicity effect occurs. The Kaiser effect does not appear before the stress is 20%, and the memory capacity of the stress is not stable enough.

4. The difference in loading rate affects the rate of change of acoustic emission, and the increase in the loading speed resulted in increased damage to the rock. The comparative analysis shows that there is a 30% difference between the peak strengths of marble and granite when they fail under two different loading rates: monotonic loading and cyclic loading.

5. The theoretical stress-strain curve can reasonably reflect the actual stress-strain characteristics of rock by combining the number of acoustic emission events with the rock damage model. The results are consistent with the acoustic emission test, which verifies the inevitable relationship between acoustic emission and damage to the rock.

The data that support the findings of this study are available from the corresponding author upon reasonable request.

The authors declare that they have no conflicts of interest to report regarding the present study.