Abstract
In order to investigate the acoustic emission (AE) features of rocks in different (medium and strong) rockburst areas, the Sangzhuling Railway Tunnel granite in China was taken as an example, the mineral composition of rocks in different rockburst areas was analyzed by using XRD (X-ray diffraction) method. The Original rock cores of different rockburst areas were processed into standard rock specimens (diameter 50 mm, height 100 mm) in different directions (transverse, oblique and longitudinal). AE feature parameters (event number, ringing count and energy) of standard rock specimens during indoor uniaxial compression test were obtained by using AE technique. The variation law of AE feature parameters of rocks in different rockburst grade areas was then analyzed. The AE features of failure precursor of rocks in different rockburst areas were therefore discussed. It shows that compared with rocks in medium rockburst area, the content of quartz and feldspar of rocks in strong rockburst area is high, while the content of biotite is low; the rock in the strong rock burst area released more energy during the failure process with about 2–3 times that of the rock in the medium rock burst area; the cumulative ringing curve of rock in medium burst area is a stepped type, while the cumulative ringing curve of rock in strong burst area is the smoothed type; the end of the second and first AE quiet period may be regarded as the failure precursor of rocks in medium and strong rockburst area, respectively. The results presented herein are important for understanding the mechanisms of rockburst.
Keywords: Rockburst grade areas, Granite, Acoustic emission feature, Mineral composition
1. Introduction
Rockburst is a common dynamic instability disaster in the excavation of deep buried tunnel engineering [[1], [2], [3], [4], [5], [6], [7], [8]]. The occurrence of rockburst depends on the initiation and expansion of crack in rocks, and the evolution process of cracks is always accompanied by the release of energy [9]. Acoustic emission (AE) is a visual representation of the energy released during rock failure [[10], [11], [12], [13], [14]], and the deformation and failure characteristics of rocks in different rockburst grades areas could be well characterized by using AE feature parameters.
Recently, there are many reports on the relationship between AE features and rockburst activity by using AE technique. For example, He et al. [15] conducted rockburst simulation tests on limestone and analyzed the variation law of AE frequency and amplitude during rockburst stage; Chen et al. [16] studied the influence of high temperature on rockburst tendency of hard rock based on AE technology; Zhang et al. [17] established the concrete expression of rockburst tendency index based on AE cumulative ringing count; Su et al. [18] conducted rockburst simulation tests of granite under different high temperature conditions and analyzed the AE features; Zhang et al. [19] proposed a rockburst warning method by clustering AE signals during tunnel rockburst simulation test; Akdag et al. [20] discussed the influence of temperature on rockburst under true triaxial loading and unloading conditions based on AE and kinetic energy analysis; Liu et al. [21] studied the effect of different horizontal stresses on tunnel rockburst by AE monitoring technique; Su et al. [22] adopted the improved rockburst test system to conduct rockburst simulation tests on granite and analyzed the variation rules of AE feature parameters; Gong et al. [23] explored the time-frequency features of AE during limestone rockburst test based on Singular Spectrum Analysis algorithm; Hu et al. [24] researched the propagation law of cracks in the granite failure process by using AE technique and explored the occurrence principle of rockburst; Liu et al. [25] studied the effect of different gradient stresses on rockburst by using AE technology; Mei et al. [26] investigated the waveform characteristics during the occurrence of rockburst in Dali rock, and then discussed the AE main frequency precursor characteristics of rock failure; Wang et al. [27] and Ren et al. [28] analyzed the spatio-temporal evolution features and damage laws of rock fractures in the process of rockburst by using AE location technology; Zhao et al. [29] performed indoor rockburst simulation tests on granite with different heights and analyzed the variation features of AE main frequency value during the process of rockburst; Yang and Zhang [30] used AE technique to carry out unloading tests on rocks with single defects, and discussed the mechanism of rockburst; Askdag et al. [31] conducted uniaxial and triaxial compression tests on rocks, studied the post-peak energy distribution characteristics of brittle rocks based on AE technology, and explored the mechanism of strain rockburst; Liu et al. [32] adopted AE technology to study the fracture failure characteristics of basalts with different defects and revealed the occurrence process of rockburst in hard brittle rocks; Ren et al. [33] investigated the fractal characteristics of rockburst in porous siltstones by utilizing AE technique; Fan et al. [34] discussed the AE activity rules of rock during indoor uniaxial rockburst experiment based on statistics and effective kinetic energy theory; Li et al. [35] analyzed the effect of unloading rate on rockburst occurrence under high pressure based on AE method; Li et al. [36] used the true triaxial rockburst test instrument to study the AE characteristics of rock in the process of horizontal multi-plane rapid unloading under high stress conditions, and put forward the precursor of rockburst; Huo et al. [37] studied the AE characteristics of rock failure process under different stress gradients, and explored the influence of stress gradient of surrounding rock on the occurrence mechanism of rockburst.
The above studies have analyzed the temporal and spatial evolution characteristics of cracks in the process of loading and unloading failure of rock, and then discussed the occurrence process of rockburst by using AE method. However, these studies mainly focus on the damage of rock in a certain grade of rockburst location, ignoring the damage characteristics of rock in different grades of rockburst location. Actually, for the same type of rock, with the change of buried depth and in-situ stress environment, the mechanical properties of rocks change, and then in the process of rock failure often show different failure phenomena. The analysis of AE parameters mainly focuses on the instantaneous parameters or the cumulative parameters, and the comprehensive analysis of the two parameters is not much. Furthermore, the effect of different processing directions (longitudinal, oblique and transverse) on the AE features of rock deformation and failure process was usually ignored, while rocks with different processing directions often show different mechanical characteristics [38]. At present, there is little research in these aspects. Therefore, the Sangzhuling Railway Tunnel in southwest China was used as the research background, the rockburst characteristics of surrounding rocks at DK185 + 207.4 (strong rockburst) and DK185 + 859.4 (medium rockburst) locations was introduced in briefly. The original cores were obtained by using drilling coring method, and the mineral composition of rocks in different rockburst grade areas was then identified. The original rock cores were thereafter processed into cylinder specimens along different processing directions. the AE feature parameters of the specimens during the process of failure were collected, and the variation rules of AE event number, ringing count and energy of rocks in different rockburst grades areas were then analyzed. The failure precursor AE features of rocks in different rockburst grades areas were also discussed.
2. Rockburst characteristics and mineral composition analysis of rocks
2.1. Rockburst characteristics of surrounding rock
The granite specimens were extracted from the Sangzhuling Railway Tunnel in China at DK185 + 207.4 and DK185 + 809.4. At DK185 + 207.4 of Sangzhuling Tunnel, the buried depth is 772 m, the maximum in-situ stress is 36.6 MPa, and the surrounding rock is Grade II. During the process of tunnel construction, slabs of rock fall first in the roof and face, followed by a lot of big rocks quickly burst out with a strong breaking sound, the rocks exceeded 30 cm thick. Furthermore, the rock flour ejection is serious on site, and the crater is irregular in shape and exceeds 4 m in deep. Rockbursts were classified into different grades according to the occurrence frequency of rockbursts, sound characteristics, rock block size, rock ejection speed, etc. The rockburst classification method are shown in Table 1 [39]. Hence, the rockburst at DK185 + 207.4 of Sangzhuling Tunnel was classified as a strong grade based on the preceding classification standard.
Table 1.
Rockburst classification method (National Standards Compilation Group of People's Republic of China, 2016).
| Features of rock burst | Rock burst grades |
||||
|---|---|---|---|---|---|
| no | slight | medium | strong | severe | |
| Sound features | None | Cracking sounds and tearing sounds | Crisp popping sounds | Very loud popping sounds | A sharp dull cracking sound |
| Movement features | None | loosen or peel off |
Burst loose, strong stripping and a small amount of ejection | Lots of bursts and ejections | Large continuous burst, large pieces of rock ejection |
| Aging features | None | Sporadic and intermittent bursts | Lasts for a long time and develops to the depth with time | Continuity and rapidly extends to the depth of the surrounding rock | Suddenly and rapidly extended to the depth of surrounding rock |
| Influence depth | None | <0.5 m | 0.5∼1 m | 1∼3 m | >3 m |
| Degree of hazard to the project | None | Small | Relatively large | Large | Heavy |
At the DK185 + 809.4 of Sangzhuling Tunnel, the buried depth is 591.6 m, the maximum in-situ stress is 23.5 MPa, and the surrounding rock is Grade II. The rockburst occurred mainly near the tunnel face, and the time of rockburst was usually within 3 h after drilling and blasting, but some were relatively slow; rockburst occurs near the vault; when rockburst occurs, it could be seen that large pieces of surrounding rockburst and fall off, the thickness of rock block is 20∼60 cm, and occasionally rockburst and fly out; at this time, a strong sound of rock cracking is heard, and the phenomenon of rock powder ejection is relatively obvious. Based on the above classification method of rockburst grade, the rockburst at this location is classified as a medium grade.
2.2. Identification of mineral composition of rocks in different rockburst areas
Original rock cores of medium and strong burst area were respectively crushed and grounded, and the grounded rock fragments were then passed through 200 mesh (0.075 μm) vibrating screen filter. The sieve residue was thereafter made into specimens for mineral composition identification, as shown in Fig. 1(a) and (b).
Fig. 1.
Crushing and grinding of core.
The mineral compositions of the rock powder were analyzed by using an X diffraction instrument (model: Ultima IV). The test voltage was 40 kV, the current was 40 mA, the scanning range was 5°–90°, and the scanning speed was 8°/min. The mineral compositions of rocks in different rockburst grade areas were obtained, as shown in Fig. 2(a) and (b).
Fig. 2.
The rock mineral compositions of rocks in different rockburst grade areas.
As shown in Fig. 2, the mineral composition types of rocks in different rockburst grade areas are basically the same, mainly including biotite, amphibole, gypsum, quartz, plagioclase and pyrite. The contents of these mineral compositions are shown in Table 2.
Table 2.
Mineral compositions of rocks in different areas.
| Specimens in different areas | Mineral compositions |
|||||
|---|---|---|---|---|---|---|
| Biotite | Amphibole | Gypsum | Quartz | Plagioclase | Pyrite | |
| Medium rockburst areas | 15.1 | 0.9 | 1.0 | 29.7 | 52.8 | 0.5 |
| Strong rockburst areas | 10.8 | 1.8 | 1.7 | 24.2 | 60.7 | 0.8 |
As shown in Table 1, the mineral compositions of rock specimens in different rockburst grade areas are mainly plagioclase, quartz and biotite, accounting for more than 95%; Compared with the rocks in the area of medium rockburst, the content of biotite in the area of strong rockburst is lower, while the content of quartz and plagioclase is higher.
The distributions of different mineral compositions of rock specimens in medium and strong rockburst areas are respectively shown in Fig. 3(a) and (b).
Fig. 3.
Distribution of mineral compositions of rocks in different rockburst areas.
As shown in Fig. 3, the rocks in different rockburst areas are massive structures and crystal structures, which belong to intrusive rocks in magmatic rocks. The whole rock surface is grayish-white, and the different mineral colors are mainly grayish-white, light grey and black. The grayish-white minerals are mainly plagioclase, most of which are tabular, a few of which are granular, and some of which are pinstriped; the particle size is mostly medium grains of 2.0–5.0 mm, followed by fine grains less than 2.0 mm, and some coarse grains greater than 5.0 mm; the distribution is chaotic; when the mineral is rotated in the sunlight, some light and dark patches could be faintly seen. The light grey minerals are mainly quartz, which are granular and have glass luster; the particle size is medium grains of 2.0–4.8 mm, and a small amount of fine grains of 0.8–2.0 mm, with disorderly distribution. The black minerals are mainly biotite, which is granular or flaky and have a glassy luster; the biotite in the strong rockburst grade area is mainly fine grained from 0.1 mm to 1.4 mm, while the biotite in the medium rockburst grade area is mainly medium grained from 2.0 mm to 3.8 mm, and the distribution is disorderly.
In conclusion, the mineral composition types of rocks in medium and strong rockburst areas are basically the same; the content of quartz and plagioclase of rocks in medium and strong rockburst area is relatively high, which make the rocks have strong strain energy storage capacity.
3. AE testing process of rock specimens under uniaxial compression
3.1. AE test equipment and rock specimens preparation
In the process of small cracks changing (closing, expanding, penetrating) and new cracks, the phenomenon of transient elastic waves emitted by rapid energy release in the material is called acoustic emission (AE). For rock materials, the AE characteristic parameters corresponding to AE phenomena are related to the generation, expansion and evolution of cracks in rocks. AE instruments could be used to detect, record and process AE signals, and the deformation and failure process of rock could be studied by analyzing the changes of AE characteristic parameters in the deformation and failure process of rock. Fig. 4 shows a schematic diagram of AE signal.
Fig. 4.
A schematic diagram of AE signal.
As shown in Fig. 4, the maximum value of the signal is the AE amplitude, the time from the start of recording to reach the maximum amplitude value is the rising time, the time when the signal amplitude exceeds the set threshold is called the duration, and each signal peak exceeding the threshold is called a ringing. The common parameters and definitions of acoustic emission are described as follows.
-
(1)
AE events. From the time an acoustic emission signal exceeds the threshold value until the signal continues to fade below the threshold value, a rectangular pulse is generated during this process, and this rectangular pulse is counted as an event. An AE event is usually used to represent the local change of a rock material, and the event count is a reflection of the activity of AE sources in the rock.
-
(2)
AE ringing count. The ringing count in AE represents the number of oscillations of the signal when the signal amplitude exceeds a set threshold in an AE event. In the process of rock failure, the internal damage of rock could be evaluated by ringing count.
-
(3)
AE energy. AE energy represents the area below the envelope of the AE signal waveform, reflecting the relative energy and intensity of the AE signal. The propagation characteristics of cracks could be reflected by energy in rock materials.
To sum up, the definition of AE event number and AE ringing count is obviously different, and an AE event number often contains one or more AE ringing counts; compared with the AE ringing count, the AE energy is not limited by the threshold value.
3.2. AE test equipment and rock specimens preparation
The AE test system (See Fig. 5) for rock mainly consists of three parts: AE data collection instrument, rock loading testing machine and control platform. The model of the AE collection instrument is PXDAQ16172G. In order to ensure the accuracy of test results, indoor temperature need be kept in the range of 0∼55 °C when the instrument is working. The AE equipment needs to be as far away from the noise signal source as possible during the test.
Fig. 5.
Outline of AE testing system.
As shown in Fig. 5, two AE sensors were respectively fixed on the upper and lower ends of the side of the rock specimen. The collection instrument, rock hydraulic loading testing machine and sensors were connected with the control platform through the data line, and the loading failure process of rock specimens was then recorded.
The specimen preparation process in uniaxial compression test is shown in Fig. 6(a) and (b), 6(c). The original cores are machined into transverse, oblique and longitudinal directions cylindrical specimens, and the three directions correspond to 0°, 45° and 90°, respectively. Specimen surface was processed by grinding and sliding so that the non-parallelism and non-verticality were controlled within 0.02 mm.
Fig. 6.
Original rock core acquisition and processing.
3.3. Process of AE test under uniaxial compression
The AE test process is mainly divided into five steps: prepare the specimens, connect the instrument and set the parameters, apply the AE coupling agent, install the AE sensor, load the specimen and collect the data.
-
(a)
Prepare the specimens (see Fig. 7). Prepare the rock specimen for loading as required, and record the relevant information, such as: specimen's name, direction, mineral composition, indoor ambient temperature, loading start and end time, stress-strain data and AE acquisition data file number, etc.
-
(b)
Connect the instrument and set the parameters. Set loading parameters (such as graded load, loading rate and holding time) and AE instrument parameters. According to the actual test situation on site, the amplifier's pre-gain is set as 40 dB, the threshold as 55 dB, the high pass as 100 kHz, low pass as 200 kHz, the lower limit of digital filter as 200k, and the upper limit as 400k.
-
(c)
Apply AE coupling agent. The debris on the specimen surface need be carefully removed before the sensors were installed. In order to ensure a good acoustic coupling between the surface of the sensor and the surface of the specimen, an appropriate amount of AE coupling agent was applied to the surface of the AE sensor. The coupling agent could fill the tiny gap between the contact surface of the specimen and the sensor, so that the acoustic impedance difference between the sensor and the detection surface was then reduced, and thus the energy reflection loss at the interface was reduced, as shown in Fig. 8.
-
(d)
AE sensors installation. Slowly lower the upper pressure plate of the testing machine until it is very close to but not in contact with the upper surface of the specimen. Clean the surface of the bottom pressure plate, and repeatedly adjust the position of the specimen on the pressure plate at the bottom of the press to ensure that the specimen could be evenly compressed during the test. Fig. 9 shows where the two sensors are installed. Because the specimen size is relatively small, small diameter sensor is adopted. The AE sensors were connected to the acquisition instrument and the control computer through the acquisition data line. During the test, data are continuously transmitted to the computer in real time.
-
(e)
Load the specimen and collect the data. After the sensor is installed, uniaxial graded loading of the specimen is carried out. Loading method is shown in Fig. 10. In this study, the single-axis multi-stage loading method was adopted to approximate the excavation process of the tunnel, and the loading rate is 0.1 mm/s. Moreover, the pressure is stabilized once when the load reaches 20, 150, 250 kN, that is, after the completion of each stage of loading, keep the load unchanged and stable for a period of time and then reload. The period time is 45s (according to the field test situation), the purpose is to make the sample cracks as fully developed as possible after each stage of loading.
Fig. 7.

Original AE specimen.
Fig. 8.

Application of acoustic emission sensor coupling agent.
Fig. 9.

AE sensors installation.
Fig. 10.
Loading method.
The specimen is loaded until completely destroyed (see Fig. 11), and the AE information is collected and processed.
Fig. 11.

Appearance of AE specimen failure.
4. AE features of rocks in different rockburst areas under uniaxial compression
According to the rock stress-strain curve and crack evolution characteristics, the rock deformation and failure process could be divided into four stages: compaction stage (stage I), elastic stage (stage II), crack expansion stage (stage III) and failure stage (stage IV). The AE features of rock specimens of different rockburst areas at different stages were then studied respectively.
4.1. Number of AE events
The number of AE events could directly reflect the number and frequency of crack in rocks. Under uniaxial compression load, the AE event number curves of the rocks in the medium rockburst area various at different deformation stages are shown in Fig. 12 (a1), (a2), (b1), (b2), (c1), (c2).
Fig. 12.
AE events of rocks in medium rockburst area.
As shown in Fig. 12, the specimen's uniaxial compressive strength varies with its directions longitudinally, obliquely, and transversely, of which the longitudinal specimen is the largest, followed by the oblique specimen and the transverse specimen is the smallest; the AE activities are relatively quiet and the number of AE events is small during the compaction stage; compared with the transverse and oblique specimens, the compaction stage time of longitudinal specimen is the shortest, which may be due to the high compaction degree of rock caused by the highest in-situ stress level in the longitudinal; in the elastic stage, the AE events of the specimens in different directions have a small increase, but no large surge, which means that the original cracks in the rock have been completely closed, and a small number of new cracks begin to sprout gradually; In the crack expansion stage, AE activities enter the active stage, the AE events occur in large quantities, and when the stress is close to the peak, AE events also reach the maximum value; in the failure stage, the stress of the specimens in different directions drops precipitously, showing typical brittle failure characteristics, indicating that the rock has a strong rockburst tendency, and the AE events of the rock at this stage rapidly decrease to a low level after a short and rapid increase.
Fig. 13 shows the AE event number curves of rocks in different deformation stages in the strong rockburst area under uniaxial compression load. In the strong rockburst area, as shown in Fig. 13 (a1), (a2), (b1), (b2), (c1), (c2), the uniaxial compressive strength of rocks is longitudinal, oblique, and transverse in order of increasing strength; compared with the rocks in the medium rockburst area, the uniaxial compressive strength of rocks in the strong rockburst area is relatively high, showing the characteristics of high strength and high rockburst tendency; in the compaction and elastic stage, the AE activities of rock specimens of different directions in the strong rockburst area are relatively calm, while in the crack expansion stage, the AE events surge and reach a high level rapidly, and the AE events in the crack expansion stage account for more than 80% of the total AE events; compared with the rock specimens in the medium rockburst area, the total duration of the compaction and elasticity stage of the rock specimens in the strong rockburst area lasts longer, which could gather more energy, and the explosion of AE events in the crack expansion stage is more sudden and concentrated. There may be a reason for this in that the rocks with strong rockbursts have significantly greater buried depths and in-situ stress levels than those with medium bursts, so the rocks in the strong rockburst area are more capable of collecting elastic energy. When the inner pores are generated and expanded, a large amount of strain energy is often released suddenly at the moment of rock failure, thus increasing the number of AE events.
Fig. 13.
AE events of rocks in strong rockburst area.
In conclusion, the rock specimen's uniaxial compressive strength relates closely to its initial in-situ stress in the tunnel where it is found, and the uniaxial compressive strength of the longitudinal specimens is the highest, followed by that of the oblique specimens and that of the transverse specimens is the least; the uniaxial compressive strength, buried depth and in-situ stress of rocks in the strong rockburst area are slightly higher than those in the medium rockburst area, which indicates that rocks in the strong rockburst area have the characteristics of high strength and high in-situ stress; the rocks in the rockburst area often have obvious brittle failure characteristics and rockburst tendency; with the increase of burial depth and initial in-situ stress level, the smaller the pores, the higher the densification, and the shorter the compaction stage; compared with the rock specimens in the medium rockburst stage, the AE events of the rock specimens in the strong rockburst stage are less in the compaction and elasticity stage, and the AE events of the rock specimens in the crack expansion stage and the stress close to the peak stress are more obvious. This indicates that the AE signals of the rock specimens in the strong rockburst stage are concentrated in the number of AE events.
4.2. AE ringing count
AE ringing count represents the number of times that AE signals exceed the threshold value in a unit time, and AE cumulative ringing count represents the cumulative number of ringing counts. The variation law of AE ringing count and cumulative ringing count well reflect the activity degree of AE signal in the process of crack initiation and propagation in rocks. Under uniaxial compression load, AE ringing count curve of rocks in medium rockburst area is shown in Fig. 14 (a1), (a2), (b1), (b2), (c1), (c2).
Fig. 14.
AE ring count of rocks in medium rockburst area.
As shown in Fig. 14, in the compaction stage, the AE ringing count of specimens in different directions changes little, and the value is almost zero; in the elastic stage, a small amount of ringing count begins to appear, and the cumulative ringing count increases very slowly; at this stage, the accumulation of elastic strain energy is the main factor; in the crack expansion stage, the ringing counts of specimens in different directions increase greatly, and the cumulative ringing counts increase in a stepped manner; in the failure stage, the ringing count of the specimens increases rapidly for a short time and then becomes stable; AE ringing count has two obvious quiet periods before the failure of rock specimens, the second quiet period of AE occurs before the rock begins to completely fail, the ratio of stress to peak stress at the beginning is about 0.75, and the ratio of stress to peak stress at the end is 0.85, and after the second quiet period, cracks expand rapidly and enter the failure stage soon. Therefore, the second quiet period of ringing count may be regarded as a precursor of rock failure.
Under uniaxial compression load, AE ringing count curve of rocks in strong rockburst area is shown in Fig. 15 (a1), (a2), (b1), (b2), (c1), (c2).
Fig. 15.
AE ring count of rocks in strong rockburst area.
As shown in Fig. 15, in the compaction and elasticity stages, the ringing count of rock specimens in the strong rockburst area does not increase significantly, the value approaches zero, and the AE activity is in a quiet period; compared with rock specimens in the medium rockburst area, the ringing count of rock specimens in the strong rockburst area has a longer quiet period, which is closely related to the initial stress level of rocks; in the stage of crack expansion, AE activities enter the active stage, and AE ringing counts mainly occur near the peak stress; the AE ringing counts of rock specimens in the stage of crack expansion account for more than 90% of all AE ringing counts; in the failure stage, the ringing count increases briefly and then becomes stable. The cumulative ringing count curve of transverse and oblique specimens has no obvious step-type change, while the cumulative ringing count curve of longitudinal specimens has a more obvious smooth-type increase, and after entering the crack expansion stage, AE ringing count increases rapidly without obvious quiet period. Compared with the specimens in the medium rockburst area, the cumulative ringing count of the specimens in the strong rockburst area does not increase significantly in the step type and gradually changes to the smooth type. The rock failure is almost instantaneous, indicating that with the increase of buried depth and initial in-situ stress, the second quiet period of the ringing count in the rock compression failure process is shorter, and the precursor of rock failure is less obvious.
In conclusion, the cumulative ringing count curve of rocks in the medium rockburst area often has the growth form of “two steps and two surges”, that is, there are two AE quiet periods and two active periods, and after the second quiet period, the rock cracks are quickly connected and the cumulative ringing count surges to near the peak, which may be used as the precursor of rock failure; compared with the rocks in medium rockburst area, the cumulative ringing curve of the rocks in the strong rockburst area changes from “two steps and two surges” to “one step and one surge”, and the curve is relatively smooth; after the end of the first calm period, the cumulative ringing count of the rock surges to near the peak, which may be used as the precursor of rock failure; the ringing count of the rocks in the strong rockburst area first stabilizes at a low value in the initial loading stage, and then approaches the peak value quickly after entering the active period, which is faster than the surge rate of the rock ringing count in the medium rockburst area. This also means that the rock in the strong rockburst area often has more AE ringing count when the rockburst occurs.
4.3. AE energy
Under uniaxial compression load, AE energy curve of rocks in medium rockburst area is shown in Fig. 16 (a1), (a2), (b1), (b2), (c1), (c2).
Fig. 16.
AE energy of rocks in medium rockburst area.
As shown in Fig. 16, in the stage of compaction and elasticity, the stress at this time reaches 45–60% of the peak stress, the pores inside the rock are compressed, and the AE energy has no obvious change; in the stage of crack expansion, the stress reaches 75–90% of the peak stress, the cracks inside the rock begin to sprout and expand rapidly, and the AE energy begins to increase obviously; when the stress reaches near the peak stress, the crack inside rocks is connected and the crack surface is formed, and the AE energy surges; the failure stage lasts for a short time and the AE energy becomes stable after a short and rapid increase; the energy released by specimens in different directions is longitudinal, oblique and transverse in order from high to low, this is because the initial in-situ stress level of longitudinal specimens is the highest, followed by oblique specimens, and the initial in-situ stress level of transverse specimens is the lowest. This also indicates that at the same buried depth of tunnel, the greater the initial ground stress is, the more energy will be released when the rock is loaded and damaged. The relationship between the cumulative AE energy of rock and time has an obvious step type growth phenomenon, and the two steps correspond to two quiet periods of AE activity, which appear in the compaction and elasticity stage and the rapid expansion stage of cracks respectively. At the end of the second quiet period, the AE energy is close to the peak values and after the end of the second quiet period, the cracks rapidly passes through and the rock loses the bearing capacity. Therefore, the second quiet period of the accumulated AE energy in the medium rockburst period may be taken as the precursor of rock failure.
Under uniaxial compression load, AE energy curve of rocks in strong rockburst area is shown in Fig. 17 (a1), (a2), (b1), (b2), (c1), (c2).
Fig. 17.
AE energy of rocks in strong rockburst area.
As shown in Fig. 17, in the compaction-elasticity stage, the AE energy release of rock specimens in the strong rockburst area is small, and the AE activity is relatively quiet, indicating that there is almost no energy release in the strong rockburst area in the compaction-elasticity stage; compared with the rocks in the medium rockburst area, the quiet period of AE energy in the compaction and elasticity stage of the rocks in the strong rockburst area is longer, indicating that the rocks in the strong rockburst area could accumulate more energy before failure; in the early stage of crack expansion, the energy release of rock in the strong rockburst is still not obvious, which is conducive to the continued accumulation of energy, while in the late stage of crack expansion, when the stress is close to the peak, AE activities suddenly become active, which is manifested as a rapid increase of AE energy and a surge of accumulated AE energy; after rock failure, the AE energy curve still has a small increase, which is because the residual strength exists after rock failure, and the rock continues to fail under the external load; compared with the specimens in the medium rockburst area, the AE energy of the rock specimens in the strong rockburst area only has a surge near the peak stress, and the AE energy reaches a large value in an instant, indicating that the rocks in the strong rockburst area could accumulate more strain energy before the stress reaches the peak stress; when the internal stress of the rock reaches the peak stress, the energy is released suddenly and violently, resulting in brittle failure of the rock; in the uniaxial compression deformation and failure process, the energy released by rock specimens in the strong rockburst area is about 2∼3 times that in the medium rockburst area; the AE energy of rocks in the strong rockburst area has a long first calm period (including compaction, elastic stage and the early stage of crack expansion stage), a sharp release of AE energy at the end of the calm period, a rapid increase in the accumulated AE energy, and then the rock quickly loses its bearing capacity, which may be used as the precursor of rock failure.
In conclusion, the cumulative AE energy of rock specimens in the medium rockburst area presents an obvious double-step growth law during loading failure process, that is, there are two AE quiet periods, and the end of the second quiet period may be taken as the precursor of rock failure. However, rock specimens in the strong rockburst area, there is an obvious single-step growth law, that is, there is a quiet period of AE, and the end of the first quiet period of AE may be taken as the precursor of rock failure. Compared with the rock specimens in the medium rockburst area, the AE energy of the rock specimens in the strong rockburst area has a longer quiet period and a more concentrated release period, which is also in line with the characteristics of sudden and large amount of energy release in the process of strong rockburst. Under the same lithology condition, the rocks in the high-grade rockburst area usually have relatively large burial depth and initial in-situ stress, and the rock structure is relatively dense. During the process of uniaxial compression deformation and failure, the rocks in the high-grade rockburst area tend to release more energy than those in the low-grade rockburst area.
5. Discussions
-
(1)
Using true triaxial test equipment to simulate rockburst unloading may be more in line with the actual stress condition of rock. However, the equipment of these tests is more complex and expensive, and the popularization and application are greatly limited. In addition, previous studies mainly focused on the AE characteristics of rocks at a specific rock burst location in the process of uniaxial, biaxial and triaxial loading and unloading, while there were few studies on the AE characteristics of rocks at different rock burst locations during the damage process. Therefore, it is still a reasonable choice to use uniaxial compression test method to study the acoustic emission characteristics in the deformation and failure process of rockburst in different grades, and then explore the mechanism of rockburst.
-
(2)
Sensors 1 and 2 were located at the upper and lower part of the rock specimens, respectively (see Fig. 5). Taking the transverse specimen in the medium rock burst area as an example, the test results of AE parameters of the specimen at these two positions were analyzed. It shows that the number of AE events of sensor 2 in the quiet period is lower than that of sensor 1; the amplitude of cumulative AE energy curve of sensor 1 is great, while that of sensor 2 is small; the maximum amplitudes of sensor 1 and 2 ringing counts appeared before and after the complete failure of the specimen, respectively.
In summary, compared with sensor 2, sensor 1 could collect more AE feature information and better reflect the actual deformation and failure process of rocks. Therefore, AE feature information collected by sensor 1 was selected.
-
(3)
The AE characteristics were significantly correlated with the in-situ stress directions and rock structure features. Considering the influence of different directions on AE feature parameters, the transverse specimen in the medium rock burst area was taken as an example, the differences between specimens in different directions in the process of deformation and failure were investigated. It shows that the angle between the transverse specimen cracks and the specimen axis is large, and the cracks intersect at the upper or lower end face of the specimen; compared with the transverse specimen, the longitudinal specimen has a relatively small angle between the crack and the axis of the specimen and the distribution is relatively concentrated, the spacing between the cracks is small, and the cracks are approximately parallel; the cracks of the oblique specimen are approximately parallel to the axis of the specimen, and the distribution of cracks is relatively dispersed; the uniaxial compressive strength of the specimens varies with the longitudinal, oblique and transverse directions, in which the longitudinal specimens are the largest, the oblique specimens are the second, and the transverse specimens are the smallest.
-
(4)
The stages of uniaxial compression deformation and failure of rock were divided mainly according to the axial stress-loading time curve and crack initiation and expansion. Taking the transverse rock specimen in the medium rock burst position as an example, the division process of rock failure stage was illustrated. The axial stress-loading time curve of the specimen during deformation and failure and the distribution of cracks at different times were shown in Fig. 18, Fig. 19 (a), 19 (b), 19 (c), 19 (d), respectively.
Fig. 18.

Division of rock deformation and failure stages.
Fig. 19.
The appearance of the rock specimen surface at different times.
-
(a)
Compaction stage (stage I), the loading time range is 0–200 s; the axial stress-loading time curve is approximately concave upward, and no cracks occur.
-
(b)
Elastic stage (stage II), the loading time range is 200–386 s; the axial stress-loading time curve is approximately linear; there is no obvious new crack on the surface of rock specimen.
-
(c)
Compaction stage (stage III), the loading time range is 386–556 s; the axial stress-loading time curve is approximately concave upward; obvious cracks began to appear on the rock surface at 295 s; at 556 s, the cracks spread rapidly from the top to the bottom, forming a through crack.
-
(d)
Failure stage (stage IV), the loading time range is 556–575 s; the axial stress-loading time curve drops rapidly in a straight line; the through crack continues to expand until the rock specimen is completely destroyed.
-
(5)
Considering the relationship between rock specimens in different directions and in-situ stress, the rock at the medium rock burst area was taken as an example, the in-situ stress test results of the rock mass at this area were shown in Table 3.
Table 3.
The in-situ stress test results.
| Mileage | Maximum horizontal principal stress |
Vertical principal stress |
Minimum horizontal principal stress |
|||
|---|---|---|---|---|---|---|
| Value/MPa | azimuth angle/° | Value/MPa | azimuth angle/° | Value/MPa | azimuth angle/° | |
| DK185 + 8094 | 23.5 | 188.9 | 14.4 | 256.6 | 14.9 | −75.6 |
As shown in Table 3, the maximum horizontal principal stress is obviously greater than the vertical principal stress; the in-situ stress of rock mass at this location is mainly horizontal tectonic stress.
Based on the in-situ stress test results in Table 3, the relationship between tunnel direction, specimen direction and maximum horizontal principal stress direction is shown in Fig. 20. As shown in Fig. 20(a) and (b), the direction of drilling the core is approximately in a straight line with the direction of the maximum horizontal principal stress; the angle between the axis of the specimens and the direction of drilling the core (that is, the direction of maximum horizontal principal stress) is longitudinal, oblique and transverse in order from low to high. Therefore, the in-situ stresses of specimens with different directions in their loading direction are longitudinal, oblique and transverse in the order from great to small.
Fig. 20.
The relationship between tunnel direction, specimen direction and maximum horizontal principal stress direction.
6. Conclusions
This study focuses on the granite from the Sangzhuling Railway Tunnel, and the mineral compositions of rocks in different rockburst areas were identified. Following the original cores, cylindrical specimens were formed (transversely, obliquely and longitudinally). The acoustic emission (AE) features of rocks in strong and medium rockburst areas under uniaxial compression load were thereafter analyzed respectively. The AE features of rock failure precursors in different rockburst areas were also discussed. It shows that.
-
(1)
The mineral composition types of rocks in different rockburst areas are basically the same, and the content of quartz and feldspar of rocks in the medium and strong rockburst area is relatively high, while the content of biotite is relatively low.
-
(2)
In the process of rock deformation and failure, the AE energy of longitudinal specimens is the highest, followed by oblique and transverse specimens.
-
(3)
In the process of deformation and failure, the energy released by the rock specimens in the strong rockburst area is about 2∼3 times that of the rock specimens in the medium rockburst area.
-
(4)
Compared with the rocks in the medium rockburst area, the growth form of AE ringing number or energy of rocks in the strong rockburst area changes from double step type to single step type.
-
(5)
The end of the second and first AE quiet period in the deformation and failure process of rocks may be regarded as a failure precursor of rocks in the medium and strong rockburst area, respectively.
Data availability statement
Data will be made available on request.
CRediT authorship contribution statement
Yimin Jiang: Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis, Data curation. Zhenyi Wang: Supervision. Xiaoliang Jin: Writing – review & editing, Funding acquisition. Yalei Wang: Writing – review & editing, Writing – original draft, Investigation, Conceptualization.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
Financial supports for the study were provided by the Science and technology project of Henan Province under Grant No. 222102320204.
References
- 1.Lai X.P., Ren J., Cui F., Shan P.F., Dai J.J., Xu H.C., Wang Z.Y. Study on vertical cross loading fracture of coal mass through hole based on AE-TF characteristics. Appl. Acoust. 2020;166 doi: 10.1016/j.apacoust.2020.107353. [DOI] [Google Scholar]
- 2.Song Z.X., Zhang J.W. Research on the progressive failure process and fracture mechanism of rocks with the structural evolution perspective. J. Struct. Geol. 2022;154 doi: 10.1016/j.jsg.2021.104484. [DOI] [Google Scholar]
- 3.Askaripour M., Saeidi A., Rouleau A., P Mercier-Langevin, Rockburst in underground excavations: a review of mechanism, classification, and prediction methods. Undergr. Space. 2022;7(4):577–607. doi: 10.1016/j.undsp.2021.11.008. [DOI] [Google Scholar]
- 4.Xu K., Song L.B., Song W.J., He J.T., Li L.P., Yan G., Chen Q., Wang G. Research on temporal and spatial distribution characteristics of microseismic events of slip-type rockburst. Shock Vib. 2022 doi: 10.1155/2022/2262401. [DOI] [Google Scholar]
- 5.Cui K., Yang Z. Early estimation method of rockburst and large deformation of surrounding rock based on the deep borehole test. Heliyon. 2023;9(6) doi: 10.1016/j.heliyon.2023.e16638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Li P.X., Chen B.R., Xiao Y.X., Feng G.L., Zhou Y.Y., Zhao J.S. Rockburst and microseismic activity in a lagging tunnel as the spacing between twin TBM excavated tunnels changes: a case from the Neelum-Jhelum hydropower project. Tunn. Undergr. Space Technol. 2023;132 doi: 10.1016/j.tust.2022.104884. [DOI] [Google Scholar]
- 7.Luo S., Gong F.Q., Peng K., Liu Z.X. Influence of water on rockburst proneness of sandstone: insights from relative and absolute energy storage. Eng. Geol. 2023;323 doi: 10.1016/j.enggeo.2023.107172. [DOI] [Google Scholar]
- 8.Luo S., Gong F.Q., Peng K., Liu Z.X. Rockburst proneness considering energy characteristics and sample shape effects. J. Rock Mech. Geotech. Eng. 2023 doi: 10.1016/j.jrmge.2023.09.003. [DOI] [Google Scholar]
- 9.Zhang P.L., Gong F.Q., Luo S., Si X.F., Xu L. Damage constitutive model of uniaxially compressed coal material considering energy dissipation. J. Mater. Res. Technol. 2023;27:920–931. doi: 10.1016/j.jmrt.2023.09.281. [DOI] [Google Scholar]
- 10.Zhou X.P., Zhang J.Z., Qian Q.H., Niu Y. Experimental investigation of progressive cracking processes in granite under uniaxial loading using digital imaging and AE techniques. J. Struct. Geol. 2019;126:129–145. doi: 10.1016/j.jsg.2019.06.003. [DOI] [Google Scholar]
- 11.Li K.S., Zhao Z., Ma D.P., Liu C.X., Zhang J.P. Acoustic emission and mechanical characteristics of rock-like material containing single crack under uniaxial compression. Arabian J. Sci. Eng. 2021;47(4):4749–4761. doi: 10.1007/s13369-021-06251-w. [DOI] [Google Scholar]
- 12.Wang Y.F., Wang S.R., Cui F., Wang W. Frequency spectrum and damage characteristics of saturated and dry red sandstone subject to shear test. Arabian J. Sci. Eng. 2022;48(4):4609–4618. doi: 10.1007/s13369-022-07135-3. [DOI] [Google Scholar]
- 13.Cui X.N., Wang J.M., Pan B. Comparative analysis of fracture characteristics between rock and rock-like materials. Heliyon. 2023;9(8) doi: 10.1016/j.heliyon.2023.e18486. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Dang Y.K., Yang Z., Y H. Study on failure mechanism of tight sandstone based on moment tensor inversion. Heliyon. 2023;9(8) doi: 10.1016/j.heliyon.2023.e19030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.He M.C., Miao J.L., Feng J.L. Rockburst process of limestone and its acoustic emission characteristics under true-triaxial unloading conditions. Int. J. Rock Mech. Min. Sci. 2010;47:286–298. doi: 10.1016/j.ijrmms.2009.09.003. [DOI] [Google Scholar]
- 16.Chen G.Q., Li T.B., Zhang G.F., Yin H.Y., Zhang H. Temperature effect of rockburst for hard rock in deep-buried tunnel. Nat. Hazards. 2014;72(2):915–926. doi: 10.1007/s11069-014-1042-6. [DOI] [Google Scholar]
- 17.Zhang Z.Z., Gao F., Shang X.J. Rockburst proneness prediction by acoustic emission test during rock deformation. J. Cent. S. Univ. 2014;21(1):373–380. doi: 10.1007/s11771-014-1950-3. [DOI] [Google Scholar]
- 18.Su G.S., Chen Z.Y., Ju J.W., Jiang J.Q. Influence of temperature on the strainburst characteristics of granite under true triaxial loading conditions. Eng. Geol. 2017;222:38–52. doi: 10.1016/j.enggeo.2017.03.021. [DOI] [Google Scholar]
- 19.Zhang Y.B., Yang Z., Yao X.L., Liang P., Tian B.Z., Sun L. Experimental study of rockburst early warning method based on acoustic emission cluster analysis and neural network identification. Rock Soil Mech. 2017;38(S2):89–98. doi: 10.16285/j.rsm.2017.S2.012. (in Chinese) [DOI] [Google Scholar]
- 20.Akdag S., Karakus M., Taheri A., Nguyen G., He M.C. Effects of thermal damage on strain burst mechanism for brittle rocks under true-triaxial loading conditions. Rock Mech. Rock Eng. 2018;51(6):1657–1682. doi: 10.1007/s00603-018-1415-3. [DOI] [Google Scholar]
- 21.Liu X.X., Zhan S.B., Zhang Y.B., Wang X.L., Liang Z.Z., Tian B.Z. The mechanical and fracturing of rockburst in tunnel and its acoustic emission characteristics. Shock Vib. 2018;2018 doi: 10.1155/2018/3503940. [DOI] [Google Scholar]
- 22.Su G.S., Shi Y.J., Feng X.T., Jiang J.Q., Zhang J., Jiang Q. True-triaxial experimental study of the evolutionary features of the acoustic emissions and sounds of rockburst processes. Rock Mech. Rock Eng. 2018;51(2):375–389. doi: 10.1007/s00603-017-1344-6. [DOI] [Google Scholar]
- 23.Gong Y.X., Song Z.J., He M.C., Gong W.L., Ren F.Q. Precursory waves and eigenfrequencies identified from acoustic emission data based on Singular Spectrum Analysis and laboratory rockburst experiments. Int. J. Rock Mech. Min. Sci. 2017;91:155–169. doi: 10.1016/j.ijrmms.2016.11.020. [DOI] [Google Scholar]
- 24.Hu X.C., Su G.S., Chen G.Y., Mei S.M., Feng X.T., Mei G.X., Huang X.H. Experiment on rockburst process of borehole and its acoustic emission characteristics. Rock Mech. Rock Eng. 2019;52(3):783–802. doi: 10.1007/s00603-018-1613-z. [DOI] [Google Scholar]
- 25.Liu X.Q., Xia Y.Y., Lin M.Q., Benzerzour M. Experimental study of rockburst under true-triaxial gradient loading conditions. Geomechanics and Engineering. 2019;18(5):481–492. doi: 10.12989/gae.2019.18.5.481. [DOI] [Google Scholar]
- 26.Mei F.D., Hu C.Y., Li P.Y., Zhang J.S. Study on main frequency precursor characteristics of acoustic emission from deep buried Dali rock explosion. Arabian J. Geosci. 2019;12(21):645. doi: 10.1007/s12517-019-4706-4. [DOI] [Google Scholar]
- 27.Wang Y., He M.C., Ren F.Q., Zhu C., Faramarzi L. Source analysis of acoustic emissions during granite strain burst, Geomatics. Natural Hazards & Risk. 2019;10(1):1542–1562. doi: 10.1080/19475705.2019.1593888. [DOI] [Google Scholar]
- 28.Ren F.Q., Zhu C., He M.C. Moment tensor analysis of acoustic emissions for cracking mechanisms during schist strain burst. Rock Mech. Rock Eng. 2020;53(1):153–170. doi: 10.1007/s00603-019-01897-3. [DOI] [Google Scholar]
- 29.Zhao F., Wang H.J., He M.C., Yuan G.X., Luo Y.W. Acoustic emission characteristics of granite specimens with different heights in rockburst tests. Rock Soil Mech. 2019;40(1):135–146. doi: 10.16285/j.rsm.2017.1017. [DOI] [Google Scholar]
- 30.Yang Y.Z., Zhang Z.N. Dynamic fracturing process of fissured rock under abrupt unloading condition: a numerical study. Eng. Fract. Mech. 2020;231 doi: 10.1016/j.engfracmech.2020.107025. [DOI] [Google Scholar]
- 31.Akdag S., Karakus M., Nguyen G.D., Taheri A., Bruning T. Evaluation of the propensity of strain burst in brittle granite based on post-peak energy analysis. Undergr. Space. 2019;6(1):1–11. doi: 10.1016/j.undsp.2019.08.002. [DOI] [Google Scholar]
- 32.Liu Z.J., Wang H.B., Zhou B. Effect of natural defects on the fracture behaviors and failure mechanism of basalt through mesotesting and FDEM modeling. Eng. Fract. Mech. 2022;271 doi: 10.1016/j.engfracmech.2022.108598. [DOI] [Google Scholar]
- 33.Ren J.J., Zhang W.L., Ma J. Experimental study on butterfly shape of failure area and fractal characteristics of rockburst. Eng. Fail. Anal. 2022;140 doi: 10.1016/j.engfailanal.2022.106636. [DOI] [Google Scholar]
- 34.Fan J.Q., Guo P., Kong F.L., Shi X.Y. Experimental study on rock failure characteristics of ejective rockburst based on energy compensation. Geotech. Geol. Eng. 2022;40(11):5547–5564. doi: 10.1007/s10706-022-02232-4. [DOI] [Google Scholar]
- 35.Li J.Y., He M.C., Liu D.Q., Elmo D., Zhang J.M. Comparative experimental investigation on true triaxial double-sided fast unloading rockburst and biaxial compression under high stress. Canidian Geotechnical Journal. 2023;60(8):1202–1215. doi: 10.1139/cgj-2022-0344. [DOI] [Google Scholar]
- 36.Li J.Y., Liu D.Q., He M.C., Guo Y.P. True triaxial experimental study on the variation characteristics of rockburst with the number of fast unloading surfaces. Rock Mech. Rock Eng. 2023;56(8):5585–5606. doi: 10.1007/s00603-023-03311-5. [DOI] [Google Scholar]
- 37.Huo M.Z., Xia Y.Y., Liu X.Q., Bowa V.M., Lin M.Q., Yan Y.F. Frequency domain analysis of AE waveforms in rockburst specimens under different stress gradients. Bull. Eng. Geol. Environ. 2023;82(7):273. doi: 10.1007/s10064-023-03288-z. [DOI] [Google Scholar]
- 38.Si X.F., Luo Y., Luo S. Influence of lithology and bedding orientation on failure behavior of “D” shaped tunnel. Theor. Appl. Fract. Mech. 2024;129 doi: 10.1016/j.tafmec.2023.104219. [DOI] [Google Scholar]
- 39.National Standards Compilation Group of People's Republic of China . China Planning; Beijing: 2016. GB 50287-2016 Code for Hydropower Engineering Geological Investigation. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data will be made available on request.















