TY - JOUR
AU - Li,, Nan
AB - Abstract In order to study the mechanism of rock bursts in a mined-out area of a gypsum mine, in this paper acoustic emission testing of the uniaxial compression of gypsum and sandstone samples is carried out. The case of rupture of the specimen is observed, and the load axial deformation curve and acoustic emission parameters are obtained for the whole process of specimen rupture. The similarities and differences between the gypsum and sandstone samples are determined in terms of their mechanical properties, their damage evolution laws and frequency band energy distributions, and the instantaneous energy characteristics of their acoustic emission. The results show that the main fracture morphology of gypsum is ‘eight’-type, and the macroscopic fracture morphology of sandstone is mainly of partial ‘Y’-type and inverted Y-type. The intensity and uniformity of the gypsum and sandstone of the medium are different; because the gypsum is more uniform, it does not show as much variation as sandstone, instead suddenly increasing and decreasing. The maximum value of the damage variable D of gypsum reached 1, but the maximum value of D of the sandstone only reached 0.9. The frequency band of the maximum energy of gypsum and sandstone gradually decreased across the the four stages of rupture, while the maximum energy percentage increased gradually. From the stage where damage gradually increases to the stage of integral fracture of the specimen, the instantaneous energy showed a certain degree of increase. With an increase in the strength of the sample, the maximum energy percentage of the two materials corresponding to each phase gradually increases, and from the stage where damage gradually increases to the stage of integral fracture of the specimen, the value of instantaneous energy obviously increases. The results indicate that gypsum mines will also experience rock bursts, as coal mines do, but the intensity will be different. Therefore, using the three indicators, the frequency band of the maximum energy, the maximum energy percentage, and the maximum instantaneous energy, the rupture of the sample can be predicted, which can be used to improve the accuracy and efficiency of early warning systems for rock bursts in gypsum mines. rock bursts in gypsum mines, fracture instability, damage evolution law, acoustic emission, wavelet packet analysis, HHT analysis, instability precursor 1. Introduction Gypsum resources play an important role in the non-metal industries, but gypsum mines, in particular, leave behind a large mined-out area. With the increase of mining depths and the enlargement of mined-out areas, the long-term stability of mined-out areas is becoming increasingly important, and is directly related to the safe operation of mines and mining plans for deep ore bodies (Wang et al2008). In recent years there have been significant accidents in gypsum mines, causing heavy casualties and property losses (Zhang and Zhang 2005). More and more people have been paying attention to accidents in gypsum mines, however, research on the mechanisms of such accidents is yet to satisfy the needs of mine production and development. In the process of mining, it is necessary to address the problem of rock bursts. For many years, there have achievements in the research of rock bursts in coal, which have played a positive guiding role in the prediction and control of rock bursts in mines. Acoustic emission techniques can be used to study the occurrence mechanism, evolution process, prediction and prevention of rock bursts in mines, and can be important in the prevention and reduction of accidents and disasters in gypsum mines. The so-called rock acoustic emission phenomenon (Wen et al2016, Zhang et al 2016, Mark and Gauna 2016, Katsuyama 1996, Shkuratnik et al2005) refers to the phenomenon in which the internal micro-crack initiation, propagation and fracture of rock material releases energy in the form of stress waves. This phenomenon is closely related to the physical and mechanical properties of the rock material, and the process and mode of loading. In the 1940s Obert and Hodgson put forward the concept of acoustic emission and in the 1950s Kaiser found the Kaiser effect. After more than 70 years of development, acoustic emission technology has been successfully used in many fields, such as the petrochemical industry, power industry, aerospace and aviation industry, metal processing, mining and geological disaster prevention, and control work. The acoustic emission of rock reflects the degree of damage, and is directly related to the evolution of the internal defects of the rock (Mandelbrot 1977, Tang 1993). A lot of researchers have studied and applied acoustic emission (Suzuki and Ohtsu 2004, Mba and Rao 2006, ISO Standard 2007, Suzuki et al2007, Grosse and Ohtsu 2008). It is helpful to understand the fracture mechanism of rock by analyzing the relationship between the characteristics of acoustic emission and lithology, so as to provide a theoretical and technical basis for monitoring dynamic disasters in rock mass using acoustic emission. He et al (2012) simulated the surface subsidence of a mined-out area of a gypsum mine in Heng Mountain County using FLAC3D with different structure parameters. They analyzed the factors influencing the surface subsidence in the mined-out area, and established an ANFIS model for prediction of surface subsidence. Huang et al (2013) studied the effect of loading rate on the velocity of gypsum under uniaxial compression. Uniaxial compression tests were carried out under four different loading rates and the variation of longitudinal wave velocity with stress during the loading process was tested. Li et al (2015) used an RMT-150B rock mechanics testing machine, and seven different kinds of high-diameter-ratio gypsum specimens were tested using uniaxial compression tests, providing analysis of mechanical properties and failure characteristics. Zhang et al (2015) performed acoustic emission tests for single-axis compression of dry and saturated coal gangue. They analyzed the fracture frequency and the entropy change of the acoustic emission signals of the whole process, based on spectrum analysis and information entropy theory, examining the precursor of acoustic emission in the deformation and failure of coal gangue. Also, some well-known scholars have performed research on the acoustic emission signal of rock (Wen et al2017, Alkan et al2007, Jan et al2008, Chmel and Shcherbakov 2013). Most of the studies in the literature are based on rocks, and the features of the acoustic emission sequences of different rocks are occasionally considered. However, there are few reports considering the failure process of gypsum samples, or analyzing the acoustic emission sequences of gypsum and sandstone samples using multi-angle analysis under the same test conditions. In the field of engineering, the stability of different lithology rock is different. On the basis of the study on instantaneous frequency by Chinese-American N E Huang in 1998, the signal processing method of the Hilbert–Huang transform (HHT) was proposed. In the process of decomposition (Huang et al1998), the HHT does not require a predetermined basis function. The modal decomposition is carried out according to the time scale characteristics of the data itself, and the characteristics of the data are preserved; it has strong self adaptability. The method has been studied and applied in many fields (Zhao and Huang 2001, Shen et al2003, Li et al2012). For example, signal denoising (Liu et al2009), blasting vibration signal characteristic analysis (Li et al2005, Ling and Li 2005, Yang et al2014), ground motion characteristic research (Wu et al2015), and other fields have achieved good results. Therefore, it is of engineering significance to study the sequence of acoustic emission parameters in the fracture process of gypsum and sandstone. Vibration signal and ground motion characteristics research have achieved good results. In this paper, uniaxial compression tests of gypsum and sandstone samples are carried out, with analysis of the form of their fractures and their damage evolution laws. Based on wavelet packet analysis and HHT, the acoustic emission parameters of gypsum and sandstone are obtained, with frequency band energy analysis and Hilbert–Huang diagrams of the two different stages of the loading process. The differences and similarities between the gypsum and rock samples in terms of fracture mode, damage evolution law, and acoustic emission characteristics are obtained in order to provide a basis for improving the accuracy of the monitoring and forecasting of rock blasts in gypsum mines. 2. Acoustic emission tests of different strength samples 2.1. Preparation of sample The gypsum and sandstone used for the tests are in strict accordance with the international rock mechanics test specification. Both ends of the gypsum and sandstone are carefully ground to make the upper and lower surfaces parallel as per the test requirements. The basic parameters of the samples are listed in table 1. Table 1. Basic parameters of the samples. Sample . Sample ID . Size (mm) . Quality (kg) . Volume (cm3) . Density (g cm-3) . Gypsum SG 50 × 100.30 0.426 196.84 2.164 Sandstone SY 49 × 100.40 0.515 189.23 2.722 Sample . Sample ID . Size (mm) . Quality (kg) . Volume (cm3) . Density (g cm-3) . Gypsum SG 50 × 100.30 0.426 196.84 2.164 Sandstone SY 49 × 100.40 0.515 189.23 2.722 Open in new tab Table 1. Basic parameters of the samples. Sample . Sample ID . Size (mm) . Quality (kg) . Volume (cm3) . Density (g cm-3) . Gypsum SG 50 × 100.30 0.426 196.84 2.164 Sandstone SY 49 × 100.40 0.515 189.23 2.722 Sample . Sample ID . Size (mm) . Quality (kg) . Volume (cm3) . Density (g cm-3) . Gypsum SG 50 × 100.30 0.426 196.84 2.164 Sandstone SY 49 × 100.40 0.515 189.23 2.722 Open in new tab 2.2. Test equipment and equipment parameters settings The test system for measuring the acoustic emission response and fracture evolution of the samples is mainly composed of a loading system, strain acquisition system, acoustic emission data acquisition system, and shielding system. A schematic diagram of the test set-up is shown in figure 1; the loading system and PCI-8 software are as shown in figure 2. Figure 1. Open in new tabDownload slide Schematic diagram of test set-up. Figure 1. Open in new tabDownload slide Schematic diagram of test set-up. Figure 2. Open in new tabDownload slide Loading system and PCI-8 software. Figure 2. Open in new tabDownload slide Loading system and PCI-8 software. The sample IDs of the gypsum and sandstone are SG and SY, respectively. A uniaxial compression test was carried out on each sample; in order to ensure the consistency of the data collected, the parameters in the experiments were consistent. The threshold value of the acoustic emission instrument is set to 42 dB, and the sampling rate is 1 MSPS. The loading mode for the sandstone is force control and the loading rate is 150 N s-1; the loading mode of gypsum is force control and the loading rate is 50 N s-1. The uniaxial compressive strength, elastic modulus, and stress–strain of the different rock samples were studied. 3. Mechanical properties and fracture damage of gypsum and sandstone samples 3.1. Mechanical properties of gypsum and sandstone samples Figure 3 shows the full stress–strain curves of the gypsum and sandstone samples. Table 2 provides the calculation results for the mechanical parameters of the sample. Figure 3. Open in new tabDownload slide Full stress–strain curves of the gypsum and sandstone samples. Figure 3. Open in new tabDownload slide Full stress–strain curves of the gypsum and sandstone samples. Table 2. Calculation results for the mechanical parameters of the samples. Sample ID . Uniaxial compressive strength (MPa) . Elastic modulus (MPa) . SG 6.31 0.8 × 103 SY 45.2 8325.6 Sample ID . Uniaxial compressive strength (MPa) . Elastic modulus (MPa) . SG 6.31 0.8 × 103 SY 45.2 8325.6 Open in new tab Table 2. Calculation results for the mechanical parameters of the samples. Sample ID . Uniaxial compressive strength (MPa) . Elastic modulus (MPa) . SG 6.31 0.8 × 103 SY 45.2 8325.6 Sample ID . Uniaxial compressive strength (MPa) . Elastic modulus (MPa) . SG 6.31 0.8 × 103 SY 45.2 8325.6 Open in new tab In the process of loading, the gypsum specimen undergoes obvious compaction, linear elasticity, elastic–plastic, and failure stages, and the residual strength after failure of the sample is very small. The full stress–strain curves of the sandstone and gypsum samples are similar in general. The sandstones experiences obvious compaction, linear elasticity, elastic–plastic, and failure stages, and the residual strength after failure of the sample is very small. However, due to the different intensities, the proportions of each stage are different. 3.2. Uniaxial compressive failure of gypsum and sandstone specimens Uniaxial compression tests were performed on gypsum and sandstone, and the fracture forms of the two were compared and analyzed. Figures 4 and 5 show the typical fracture forms of the gypsum and sandstone samples and corresponding sketch maps. Figure 4. Open in new tabDownload slide The main fracture morphology of gypsum samples. Figure 4. Open in new tabDownload slide The main fracture morphology of gypsum samples. Figure 5. Open in new tabDownload slide The main fracture morphology of sandstone samples. Figure 5. Open in new tabDownload slide The main fracture morphology of sandstone samples. The failure types of the gypsum and sandstone samples are mainly wedge-type. There are a multiple shear rupture surfaces, and symbiotic characteristics of shear planes in local and global. The macroscopic rupture forms an ‘eight’, ‘Y’, or inverted Y shape along the axial direction of the rupture. The study indicates that the two cleavage planes are not formed synchronously. The rupture process occurs on a control sheet deformed under two longitudinal side cracks, by a series of smaller integrated wedges. The sample is then under the joint action of tension deformation and wedge splitting forces, and the sample is split and destroyed. The failure mechanism shows obvious wedge wedge action. Although most of the specimens are intact, the specimen can be split immediately when the wedge is formed in the sample. There are some differences in the cracking of gypsum and sandstone. Gypsum mainly exhibits eight-type failures. This type of crack is due to the low strength of the specimen, the lack of obvious end effects, together with the very low strength, smoothness and heterogeneity of the coal samples. Sandstone mainly exhibits Y- and inverted Y-type fracture cracks. This type is caused by the relative strength of the sample being relatively high, the end effect is more prominent and the end is in the three direction stress state due to the friction suppression. 3.3. Damage evolution analysis of gypsum and sandstone samples Coal, rock, and concrete, as non-uniform materials, have a large number of internal defects such as micro-cracks, micro-holes etc, and the physical and mechanical properties of samples of the same materials also show some differences. The specimen is deformed up to the final macroscopic rupture in the course of uniaxial compression; the internal damage is a quantitative change until the destruction occurs. Damage mechanics research studies the transformation of the micro-cracks, holes, and other defects in these raw materials into macroscopic cracks, that is, the whole process of micro-crack initiation, propagation, evolution, nucleation, formation of the macroscopic crack, crack growth, and macroscopic instability of samples. By using damage theory to analyze the damage state of the internal mechanics during uniaxial compression, first we need to choose the appropriate damage variable to describe the internal damage state of the material. In this paper, the damage variable is D, as introduced by Andrianof et al Its physical meaning is ‘loss of bearing capacity’ and it is the percentage of initial cross-sectional area of material. Taking into account the defects of micro-cracks and micro-holes in the load-bearing section of the material, the calculation formulas are D=A*A=A-A˜A,1 where A is the initial cross-sectional area of the material, A* is the damaged area and A˜ is the effective area. The value range of the damage variable [0, 1] is D, which corresponds to the different degrees of damage of the material. When D = 0 the material is in the state of no damage, and D = 1 is the complete damage state of the material. According to material mechanics, the cross-sectional area, A, of force, F, is the stress, σ, σ=F/A.2 The effective stress on the effective cross section is: σ˜=F/A˜.3 Now putting (2) and (3) into (1) we obtain σ˜=σ1-D.4 Known as the strain equivalence principle, the sigma force in material damage caused by the strain and the effective force in the non-destructive material caused by the strain equivalence is ε=σ/E˜=σ˜/E.5 Using (4) and (5) can obtain damage variable D as D=1-σ˜/εE6 where E is the elastic modulus of the material, σ is stress, and ε is strain. Uniaxial compression tests were carried out on different strength specimens, and the evolution law of the loss variable D was studied. The damage variable D of different samples was calculated using equation (6), and the relationship curves of the damage variable curves D and strain curves, stress–strain curves (two-dimensional graph), were calculated. It can be seen from figure 6 that under the conditions of uniaxial compression, the damage variable D of the gypsum and sandstone samples in the strain evolution process can be divided into three stages: declining stage of damage reduction, rising stage of rapid damage growth and saturation stage of slow damage growth. As they are heterogeneous media, the samples have different shapes and sizes of micro-cracks and micro-pores, which lead to initial damage in accordance with these micro-cracks and pores. In the initial stage of loading, the initial micro-cracks and micro-holes in the sample are changed into real damage, the volume becomes smaller, which leads to the effective load of the event. According to equation (6), the damage variable D is gradually reduced. With continuous loading, there are a large number of micro-cracks in the specimen expansion which reach through into the core. This results in a sharp reduction of the effective bearing area, and damage variable D shows a rapid growth trend. Specimens were loaded to peak stress. Due to this, the inside of the specimen had a large number of cracks propagating, coalescing, and forming macro-cracks. After new internal fissures form less frequently, the damage growth rate becomes very slow, with the damage curve close to level, and damage variable D reaches maximum values (close to 1). The damage variable D can reflect well the evolution process of the internal crack during loading of the specimen. Figure 6. Open in new tabDownload slide The damage evolution laws of gypsum and sandstone. Figure 6. Open in new tabDownload slide The damage evolution laws of gypsum and sandstone. The intensity and uniformity of the gypsum and sandstone are different. Because the gypsum is more uniform, it does not show as much variation as sandstone, instead suddenly increasing and decreasing. The maximum value of damage variable D for gypsum essentially reaches 1, but the maximum value of D for the sandstone only reaches 0.9. 4. Analysis of the characteristics of the acoustic emission signal of gypsum and sandstone Acoustic emission signal waveform frequency analysis can be obtained for samples during loading and damage evolution, with different acoustic emission signal spectrum characteristics corresponding to gypsum and sandstone in the different stages of damage propagation. The spectral characteristics of the acoustic emission signal are little affected by other factors, and almost remain unchanged. In addition, the energy of the signal is also an important parameter that reflects the damage to the material. Therefore, the analysis of the spectrum and energy characteristics of acoustic emission may be an effective method to distinguish the type of damage. 4.1. Wavelet packet analysis of the acoustic emission signal in the frequency domain We used MATLAB to prepare a wavelet packet analysis program to analyze microseismic events. The microseismic monitoring system M/E ARAMIS has a sampling frequency of 500 Hz; according to the sampling theorem of Nyquist the frequency is 250 Hz. Db8 is chosen as the wavelet function of the microseismic signal using five-layer wavelet packet decomposition. In the fifth layer, a total of 25 = 32 wavelet packets, then the whole frequency domain is divided into 32 sub-bands; each sub-band has a bandwidth of 7.8125 Hz. The calculation of the microseismic signals in the wavelet packet decomposition is the percentage of nodes of the fifth layer energy and total energy, and is expressed in the form of a histogram. Based on wavelet packet analysis, the acoustic emission waveforms of gypsum and sandstone under uniaxial loading were compared and analyzed. Figure 7 shows the waveform and frequency band energy diagrams for gypsum in the four stages of rupture, and figure 8 the equivalent for sandstone. Figure 7. Open in new tabDownload slide The waveform and frequency band energy diagrams of gypsum in the four stages of rupture. Figure 7. Open in new tabDownload slide The waveform and frequency band energy diagrams of gypsum in the four stages of rupture. Figure 8. Open in new tabDownload slide The waveform and frequency band energy diagrams of sandstone in the four stages of rupture. Figure 8. Open in new tabDownload slide The waveform and frequency band energy diagrams of sandstone in the four stages of rupture. Analyzing the load damage waveform and frequency band energy of gypsum across the whole damage process (the initial damage, slow development, gradual intensification, and entire specimen fracture stages), the frequency band of the maximum energy of gypsum gradually decreased from 13 → 6 → 5 → 1, but the maximum energy percentage gradually increased from 13% → 13.5% → 15.2% → 28.8%. Analyzing the load damage waveform and frequency band energy of sandstone across the whole damage process, the frequency band of the maximum energy of sandstone gradually decreased from 13 → 5 → 5 → 1, but the maximum energy percentage gradually increased from 14% → 14.5% → 18% → 35.6%. The initial stage of damage is the expansion of the original micro-cracks in the matrix, the slow development stage corresponds to the damage as the matrix cracks. The more severe stage of matrix cracking and interface cracking follows, and then the entire specimen fractures in the final stage, with a variety of forms of damage to the sample. From figure 9, we can see the frequency bands of the maximum energy for gypsum and sandstone for different rupture stages, which gradually decrease, and the maximum energy percentages, which gradually increase. However, at each stage the frequency band of the maximum energy is highest for sandstone, and the maximum energy percentage for each stage is highest for gypsum. The four stages of rupture of gypsum and sandstone can be differentiated by the frequency band, where the maximum energy gradually would decrease, and the maximum energy percentage, which would gradually increases with stratum failure. Using the different points of these characteristics can effectively divide samples into different load damages. Figure 9. Open in new tabDownload slide The analysis chart of gypsum and sandstone in the four stages of rupture. Figure 9. Open in new tabDownload slide The analysis chart of gypsum and sandstone in the four stages of rupture. 4.2. HHT analysis of the frequency domain characteristics of acoustic emission signals HHT technology is a new signal processing method, which was proposed by E Huang Norden and other scholars in 1998 at NASA. The HHT method is widely used in the analysis of non-linear and non-stationary signals, and is an important breakthrough in the analysis of Fourier transforms. In essence, the HHT method is a smooth processing of a signal. The time signal is decomposed by empirical mode decomposition (EMD), to make the real existence of different scale fluctuations or trends gradually decomposed. A series of data sequences with different characteristic scales are generated, each of which is called an intrinsic mode function (IMF). The HHT method for signal analysis and processing generally has the following steps. First, empirical mode decomposition, where the signal, either stochastic non-linear or non-stationary, is decomposed into a number of IMFs. Then for each IMF component the HHT transform is used to draw the corresponding Hilbert spectrum. Finally, pooling the various components of the Hilbert spectrum will allow one to obtain the initial signal of the Hilbert spectrum, and determine the signal energy and the frequency. Based on the HHT analysis of acoustic emission signals, the gypsum and sandstone in the increasing damage stages and overall fracture stage are compared in the HHT three-dimensional analysis chart and instantaneous energy diagram of figures 10 and 11. Figure 10. Open in new tabDownload slide HHT analysis diagram and instantaneous energy diagram of the increasing damage stages and overall fracture stage of gypsum. Figure 10. Open in new tabDownload slide HHT analysis diagram and instantaneous energy diagram of the increasing damage stages and overall fracture stage of gypsum. Figure 11. Open in new tabDownload slide HHT analysis diagram and instantaneous energy diagram of the increasing damage stages and overall fracture stage of sandstone. Figure 11. Open in new tabDownload slide HHT analysis diagram and instantaneous energy diagram of the increasing damage stages and overall fracture stage of sandstone. According to the HHT analysis of the three stages of gradually increasing damage and the sample fracture stage, the maximum instantaneous energy of the fracture stage increased greatly compared to the instantaneous energy of the increasing damage stage, for gypsum by 1.4–1.98, sandstone by 1.41–2.18. The increase in magnitude is also different. For the gradually increasing damage stage, gypsum had a maximum instantaneous energy of 1.39, and for the overall fracture stage the maximum instantaneous energy was 1.98, an increase of 43.17%. For sandstone the gradually increasing damage stage has a maximum instantaneous energy of 1.41, and the overall fracture stage has a maximum instantaneous energy of 2.18, an increase of 54.61%. The increased instantaneous maximum energy of gypsum and sandstone in the overall fracture stage compared to the gradually increasing damage stage can be used in failure of stratum monitoring as an indicator, and according to the different stages can effectively distinguish between samples of different load damage. Research shows that when certain conditions are met, a gypsum layer can behave like a layer of sandstone, with rupture instabilities leading to rock bursts as in coal mines. Therefore it is important to take measures to develop effective rock burst early warning systems for gypsum mines. 4.3. Analysis of the fracture precursor characteristics of different strength samples For the gypsum sample, when there are two adjacent frequency bands where the maximum energy is ≤6, the maximum energy percentage is ≥25%, and the maximum instantaneous energy is ≥1.9, the occurrence probability of fracture and instability is large. For the sandstone sample, when there are two adjacent frequency bands where the maximum energy is ≤6, the maximum energy percentage is ≥30%, and the maximum instantaneous energy is ≥2, there is a great probability of rupture instability. For practical application in a gypsum mine, through field investigation and on the basis of previous research, there needs to be an understanding of the existing regional rock environment and monitoring, and the structure and force characteristics of the rock environment. Then, based on suitable acoustic emission parameters, a joint multi-parameter analysis and prediction can be established to obtain more accurate precursory information, and improve coal and rock dynamic disaster forecasting accuracy. 5. Conclusions The failures of the gypsum and sandstone samples are mainly of wedge-type; the macro-fracture morphology is ‘eight’-type, ‘Y’-type, partial ‘Y’-type, or inverted ‘Y’-type. With increasing lithologic strength, the macro-form rupture of the gypsum and sandstone gradually changes from ‘eight’-type to partial and inverted ‘Y’-type. For the gypsum and sandstone samples under conditions of uniaxial compression, the damage variable D of the strain evolution process can be divided into three stages: declining stage of damage reduction, rising stage of rapid damage growth and saturation stage of slow damage growth. The intensity and uniformity of the gypsum and sandstone are different. Because the gypsum is more uniform, it does not show as much variation as sandstone, instead suddenly increasing and decreasing. The maximum value of damage variable D for gypsum reached 1, but the maximum value of damage variable D for sandstone only reached 0.9. The frequency band of the maximum energy of gypsum and sandstone gradually decreased across the four stages of rupture. The maximum energy percentage increased gradually, and from the stage where damage gradually increased to the stage of overall sample fracture, instantaneous energy shows a certain degree of increase. With the increase of the strength of the sample, the maximum energy percentage across the four stages is correspondingly gradually increased, and the increase of instantaneous energy is obvious. For the gypsum sample, when there are two adjacent the frequency bands where the maximum energy is ≤6, the maximum energy percentage is ≥25%, and the maximum instantaneous energy is ≥1.9, the probability of occurrence of fracture and instability is large. For the sandstone sample, when there are two adjacent frequency bands where the maximum energy is ≤6, the maximum energy percentage is ≥30%, and the maximum instantaneous energy is ≥2, there is a great probability of rupture instability. Acknowledgments This work was supported by the Fundamental Research Funds for the Central Universities (2017CXNL02) and a Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD). References Alkan H , Cinar Y , Pusch G . , 2007 Rock salt dilatancy boundary from combined acoustic emission and triaxial compression tests , Int. J. Rock Mech. Min. Sci. , vol. 44 (pg. 108 - 119 ) 10.1016/j.ijrmms.2006.05.003 Google Scholar Crossref Search ADS WorldCat Crossref Chmel A , Shcherbakov I . , 2013 A comparative acoustic emission study of compression and impact fracture in granite , Int. J. Rock Mech. Min. Sci. , vol. 64 (pg. 56 - 59 ) 10.1016/j.ijrmms.2013.08.025 Google Scholar Crossref Search ADS WorldCat Crossref Grosse C U , Ohtsu M . , 2008 , Acoustic Emission Testing: Basics for Research—Applications in Civil Engineering Heidelberg Springer (pg. 3 - 10 ) 10.1007/978-3-540-69972-9 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Crossref He G-C , Ding D-X , Liu Y , Zhang Z-J . , 2012 ANFIS prediction of the surface subsidence of the old goaf of the gypsum mine in Hengshan , J. Min. Saf. Eng. , vol. 06 (pg. 876 - 881 ) OpenURL Placeholder Text WorldCat Huang N E , Shen Z , Long S R . , 1998 The empirical mode decomposition and the Hilbert spectrum for nonlinear and nonstationary time series analysis , Proc. R. Soc. Lond., Ser. A , vol. 454 (pg. 903 - 995 ) 10.1098/rspa.1998.0193 Google Scholar Crossref Search ADS WorldCat Crossref Huang X-C , LI H-B , Xia X , Wang M , Song Q-J , Liu T-T . , 2013 Experimental study of the effect of loading rate on the velocity of gypsum under uniaxial compression , Water Res. Power , vol. 9 (pg. 89 - 91 ) (see also p 181) OpenURL Placeholder Text WorldCat ISO Standard . , 2007 ISO 22096:2007. Condition monitoring and diagnosis of machines—acoustic emission (International Organization for Standardization) Jan V , Rudajevb V , Lokajíček T . , 2008 Application of autocorrelation analysis for interpreting acoustic emission in rock , Int. J. Rock Mech. Min. Sci. , vol. 45 (pg. 1068 - 1081 ) 10.1016/j.ijrmms.2007.11.004 Google Scholar Crossref Search ADS WorldCat Crossref Katsuyama K . , 1996 , Application of Acoustic Emission (AE) Technology Beijing Metallurgical Industry Press (translated by Xiating Feng) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Li C-W , Xie B-J , Yang W . , 2012 Coal impact damage SHPB testing signal de-noising based on HI-IT method , J. China Coal Soc. , vol. 37 (pg. 1796 - 1802 ) 10.13225/j.cnki.jccs.2012.11.014 OpenURL Placeholder Text WorldCat Crossref Li X-B , Zhang Y-P , Liu Z-X . , 2005 Wavelet analysis and Hilbert–Huang transform of blasting vibration signal , Explosion and Shock Waves , vol. 06 (pg. 528 - 535 ) OpenURL Placeholder Text WorldCat Li Y-L , Liang W , He G-C , Zhang Z-J . , 2015 Failure characteristics and energy dissipation of the gypsum sample in diffferent height to diameter ratios , J. Univ. South China (Sci. Technol.) , vol. 02 (pg. 30 - 36 ) OpenURL Placeholder Text WorldCat Ling T-H , Li X-B . , 2005 Analysis of energy distributions of millisecond blast vibration signals using the wavelet packet method , Chin. J. Rock Mech. Eng. , vol. 07 (pg. 1117 - 1122 ) OpenURL Placeholder Text WorldCat Liu Q , Zhou R-Z , Liu Y-H . , 2009 Computation and analysis of seismic response and energy based on Hilbert–Huang transform , Eng. J. Wuhan Univ. , vol. 06 (pg. 780 - 784 ) OpenURL Placeholder Text WorldCat Mandelbrot B B . , 1977 , Fractals: Forms, Chance and Dimension San Francisco, CA Freeman Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Mark C , Gauna M . , 2016 Evaluating the risk of coal bursts in underground coal mines , Int. J. Mining Sci. Technol. , vol. 26 (pg. 47 - 52 ) 10.1016/j.ijmst.2015.11.009 Google Scholar Crossref Search ADS WorldCat Crossref Mba D , Rao R B K N . , 2006 Development of acoustic emission technology for the condition monitoring and diagnosis of rotating machines: bearings, pumps, gearboxes, engines and rotating structures , Shock Vib. Dig. , vol. 38 (pg. 3 - 16 ) 10.1177/0583102405059054 Google Scholar Crossref Search ADS WorldCat Crossref Shen J J , Yen W P , O’Fallon J . , 2003 Interpretation and application of Hilbert–Huang tranformation for seismeic performance analyses , Tech. Counc. Lifeline Earthquake Eng. Monogr. , vol. 25 (pg. 657 - 666 ) 10.1061/40687(2003)67 OpenURL Placeholder Text WorldCat Crossref Shkuratnik V L , Filimonov L , Kuchurin S V . , 2005 Regularities of acoustic emission in coal samples undertriaxial compression , J. Min. Sci. , vol. 41 (pg. 44 - 52 ) 10.1007/s10913-005-0062-8 Google Scholar Crossref Search ADS WorldCat Crossref Suzuki T , Ohtsu M . , 2004 Quantitative damage evaluation of structural concrete by a compression test based on AE rate process analysis , Constr. Build. Mater. , vol. 18 (pg. 197 - 202 ) 10.1016/j.conbuildmat.2003.10.009 Google Scholar Crossref Search ADS WorldCat Crossref Suzuki T , Ohtsu M , Shigeishi M . , 2007 Relative damage evaluation of concrete in a road bridge by AE rate-process analysis , Mater. Struct. , vol. 40 (pg. 221 - 227 ) 10.1617/s11527-006-9133-9 Google Scholar Crossref Search ADS WorldCat Crossref Tang C-A . , 1993 , The Disaster of Rupture Process of Rock Beijing Coal Industry Press (pg. 70 - 71 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Wang J A , Shang X C , Ma H T . , 2008 Investigation of cata-strophic ground collapse in Xingtai gypsum mines in China , Int. J. Rock Mech. Min. Sci. , vol. 5 (pg. 1480 - 1499 ) 10.1016/j.ijrmms.2008.02.012 Google Scholar Crossref Search ADS WorldCat Crossref Wen Z , Wang X , Chen L , Lin G , Zhang H . , 2017 Size effect on acoustic emission characteristics of coal-rock damage evolution , Adv. Mater. Sci. Eng. , vol. 2017 3472485 10.1155/2017/3472485 OpenURL Placeholder Text WorldCat Crossref Wen Z , Wang X , Tan Y , Zhang H , Huang W , Li Q . , 2016 A study of rockburst hazard evaluation method in coal mine , Shock Vibration , vol. 2016 (pg. 1 - 9 ) 10.1155/2016/8740868 OpenURL Placeholder Text WorldCat Crossref Wu Z-J , Zhang J-J , Wang Z-J . , 2015 Research on seismic instantaneous frequency characteristics based on modified HHT algorithm , Earthquake Eng. Eng. Dyn. , vol. 2 (pg. 86 - 93 ) 10.13197/j.eeev.2015.02.86.wuzj.010 OpenURL Placeholder Text WorldCat Crossref Yang R-S , Gao X-T , Che Y-L . , 2014 Joint time-frequency analysis of blast strain wave based on Hilbert–Huang transformation , J. Vib. Shock , vol. 10 (pg. 17 - 21 ) 10.13465/j.cnki.jvs.2014.10.004 OpenURL Placeholder Text WorldCat Crossref Zhang J , Zhang S-X . , 2005 Cause and prevention of rock burst disaster in gypsum deposit , New Build. Mater. , vol. 06 (pg. 9 - 11 ) OpenURL Placeholder Text WorldCat Zhang Y-B , Liang P , Liu X-X , Liu S-J , Tian B-Z . , 2015 Experimental study on precursor of rock burst based on acoustic emission signal dominant frequency and entropy , Chin. J. Rock Mech. Eng. , vol. S1 (pg. 2959 - 2967 ) 10.13722/j.cnki.jrme.2014.0654 OpenURL Placeholder Text WorldCat Crossref Zhang Z , et al. , 2016 The observation of AE events under uniaxial compression and the quantitative relationship between the anisotropy index and the main failure plane , J. Appl. Geophys. , vol. 134 (pg. 183 - 190 ) 10.1016/j.jappgeo.2016.09.004 Google Scholar Crossref Search ADS WorldCat Crossref Zhao J-P , Huang D-J . , 2001 Mirror extending and circular spline function for empirical mode decomposition method , Zhejiang Univ. (Sci.) , vol. 2 (pg. 247 - 252 ) 10.1631/jzus.2001.0247 Google Scholar Crossref Search ADS WorldCat Crossref © 2017 Sinopec Geophysical Research Institute
TI - Analyzing the rules of fracture and damage, and the characteristics of the acoustic emission signal of a gypsum specimen under uniaxial loading
JF - Journal of Geophysics and Engineering
DO - 10.1088/1742-2140/aa676e
DA - 2017-08-01
UR - https://www.deepdyve.com/lp/oxford-university-press/analyzing-the-rules-of-fracture-and-damage-and-the-characteristics-of-e0zseVXJPQ
SP - 780
VL - 14
IS - 4
DP - DeepDyve
ER -