Study on the spectrum characteristics of rock burst and the stability of the surrounding rock of a roadway in mines under dynamic load


 This study analyzes the impact pressure in the Yuejin and Qianqiu coal mines in the Yimei mine area, and shows that rock bursts may be caused by damage to the overburden gravel strata caused by coal seam mining and unreasonable mining layout. Rock burst microseismic signals from the Yuejin and Qianqiu mines show that the duration of the vibration waveform is greater than 0.06 s. The fast Fourier transform shows that the low-frequency component of the rock burst accounts for a large proportion, with the main frequency being concentrated in the range between 5 and 50 Hz. A numerical simulation scheme was designed, and the extended D-P strength criterion was adopted to select the distributed load of a sinusoidal pulse in the load waveform as the dynamic load. The plastic strain energy density distribution is used to measure the tendency of the surrounding rock to impact the roadway. By changing the shock position, wave frequency, disturbance intensity, tunnel section shape, and buried depth, it is seen that when a (vibration wave amplitude) = 2.0 m/s2, f (vibration wave frequency) = 40 Hz, H (roadway buried depth) = 1000 m, θ (the angle between the impact position of the seismic wave and the center of the roadway) = 180°, and the roadway section is horseshoe-shaped, the tendency of the surrounding rock to impact the roadway is higher. Under the same conditions, the impact tendency of the surrounding rock on the roadway is the smallest and second smallest when the roadway is circular and straight-wall arched, respectively.


Introduction
With the continuous increase in the mining depth and intensity of exploring and exploiting coal resources, dynamic disasters, such as rock bursts, during the process of mining coal resources have become increasingly exacerbated. By 2020, there were 133 rock bursts in mines in China [1], of which 50 mines with a mining depth of more than 1000 m had rock bursts. Rock bursts lead to serious roadway deformation, support breakage, rib spalling and roof fall, and casualties, which seriously threaten the safe and efficient mining of coal resources [2][3][4].
For a long time, scholars at home and abroad have paid attention to the mechanism and prevention of coal mine rock burst. Qian [5] divided rockburst into fault slip or shear fracture induced rockburst and strain type rockburst induced by rock failure according to different mechanism of rockburst. Combined with accident cases, the occurrence mechanism and characteristics of strain type and slip type rockburst and rock burst were analyzed. Pan et al. [6] established the dynamic response differential equation of surrounding rock and support under rock burst, and proposed the control method of surrounding rock of roadway under rock burst. Dou et al. [7] carried out the research on the identification of acoustic and seismic precursors in the process of coal and rock disaster by means of laboratory acoustic emission and stope microseismic monitoring, and proposed a comprehensive early warning model of rock burst based on microseismic precursor index system. Yang and Wei [8] used CDEM software to simulate and study the mechanical properties of surrounding rock of deep roadway under the conditions of uniaxial compressive strength, disturbance stress, frequency, buried depth and wave propagation angle. Jin et al. [9] conducted threedimensional dynamic compression tests on red sandstone under different static stress and impact velocity.
Xia et al. [10] established the static comprehensive evaluation index and method of impact risk, aiming at the problems of critical mutation and difficulty in weight quantification in impact risk evaluation by comprehensive index method and multi factor coupling method. Song et al. [11] analyzed the stress distribution characteristics and microseismic activity law of surrounding rock in coal seam gangue occurrence area under dynamic and static load conditions, which provided a reference for the prevention and control of rock burst in coal face with gangue. Jiao [12] quantitatively studied the influence law of single factor such as surrounding rock strength, source distance, source strength, original rock stress, support strength and multi factor interaction on the deformation of roadway anchorage bearing structure. Wang et al. [13] used UDEC numerical simulation to study the stress distribution and evolution in different pushing and mining directions of adjacent fault working face under hard and thick roof, and analyzed the influence and difference of pushing and mining direction on stress characteristics and induced impact potential risk main control area. He et al. [14] carried out the experimental research on mechanical response of commonly used pallets and their combined components in coal mines by using the drop weight impact test device, revealed the impact resistance mechanical performance of bolting surface components, and provided design basis for the selection of bolting surface components in rock burst roadway. Based on the particularity of graben structure and the occurrence characteristics of overburden. Wu et al. [15] analyzed the movement characteristics of overburden and the stress evolution law of coal and rock mass in the graben structure area, established the corresponding mechanical model, and studied the mechanism of rock burst in the graben structure area. Cui et al. [16] analyzed the mining disturbance characteristics under different advancing speed and mining thickness, and discussed the application of mining disturbance characteristics in mine productivity control and mining method optimization.
To summarize, during rock excavation, dynamic mine disasters caused by the sudden release of high stress from the surrounding rock and dynamic disturbance load have complex dynamic characteristics.
Therefore, this study analyzes the causes of rock bursts and the frequency spectrum characteristics of vibration waveforms of these rock bursts in the Yuejin and Qianqiu mines in the Yi coal mine area. A numerical simulation scheme is designed to simulate the response characteristics of different influencing factors on the surrounding rock of roadways under dynamic load to provide theoretical guidance to control the impact of surrounding rock under dynamic load on roadways.

Spectrum characteristics of rock bursts
Frequent microearthquakes have occurred in the Yuejin and Qianqiu mines, causing difficulties for safe and efficient mining. To predict the precursor information of rock bursts, an ESG microseismic monitoring system has been installed in the mine. In this study, the seismic waveform collected during the occurrence of rock bursts was analyzed, and a fast Fourier transform was carried out using MATLAB software to analyze the spectrum characteristics of the rock bursts to guide subsequent research.

Spectrum characteristics of the Yuejin Mine
There have been many rock burst accidents in the Yuejin coal mine during the mining of the working Since the probe used in the microseismic monitoring system is a velocity sensor, the vibration signal is a velocity time history curve. Fig. 1 and Fig. 2 show the velocity time vibration signal and corresponding spectrum characteristics of the three rock bursts respectively.

Spectrum characteristics of the Qianqiu mine
In February 2014, three rock burst events occurred in the 21032 return wind uphill of the Qianqiu mine, with an energy of 1.1% × 10 7 J and a magnitude of 1.9. The impact event occurred in the roof. When the rock burst occurred, the roadway 85 m above the return air connecting roadway and 20 m away from the lower slope changing point was damaged to varying degrees. The roadway 50 m away from the lower slope changing point was seriously damaged, and the roadway was basically closed. There was only about 0.8 m of space on the lower side of the roadway. Most of the 36 u embracing pillars in the roadway were bent, and the two air doors in the lower parking lot were damaged by the shock wave; the gas concentration was as high as 9%, and the 763 belt inclined roadway, the strong belt head chamber, and the three-meter winch room in mining area 21 were deformed to varying degrees. Fig. 2 shows the velocity time vibration signal diagram and the corresponding spectrum characteristic of the three rock bursts. The causes of rock burst in these two mines are as follows: coal mining affects the overlying breccia bed; activity such as coal seam deep hole pressure relief blasting and coal seam deep hole water injection failed to effectively relieve the risk of rock burst. The mining layout is unreasonable, the return air is arranged in the stress concentration coal pillar, and the stress change is unstable when driving through the coal seam.
Analyses of the shock waves show that the low-frequency component of the main shock event accounts for a large proportion, and the main frequency is mainly distributed between 5 and 50 Hz.

Model establishment and boundary condition setting
The The constitutive model adopted in this simulation is the extended D-P strength criterion, as shown in formula (1): Where, According to plastic theory, the flow rule associated with the extended D-P strength criterion is as follows: Where, c-cohesion, MPa; φ-Internal friction angle, °.

Scheme design
In this study, the impact location, roadway buried depth, vibration wave frequency, amplitude, and  (5) We set the buried depth, frequency, impact position, and seismic wave acceleration amplitude value, that is, when H = 1000 m, f = 40 Hz, θ = 180°, a = 2.0 m/s 2 , we changed the shape of the roadway section to be rectangular, circular, a straight wall arch, and horseshoe-shaped.

Selection of damping and numerical calculation of transient dynamics
Rayleigh damping is commonly used to simulate geotechnical dynamics. The expression of Rayleigh damping is as follows: In the formula: α is the mass damping coefficient, β is the stiffness damping coefficient, and δ is the damping ratio. Generally, the value is about 2% to meet the requirements of numerical calculation. To calculate the damping coefficient of the structure, a modal analysis was performed on the model. After completing the modal analysis, a list of the natural frequencies of each order was obtained, and the first natural frequency f1 and the sixth natural frequency f6 of the structure were selected. According to the damping, we calculated the mass damping coefficient and the stiffness damping coefficient with a ratio of δ = 2%. The results of the modal analysis of the structure using the ANSYS software are as follows: SET TIME/FREQ LOAD STEP SUBSTEP CUMULATIVE

Dynamic load conditions
This model adopts the method of excavating first and then applying the seismic source, which is more in line with the on-site shock situation. The action time of the applied acceleration waveform is 0.05 s, the acceleration amplitude is a = 2.0 m/s 2 , and the constitutive model used in the transient dynamics simulation is the extended D-P intensity criterion; the dynamic disturbance can be used in the numerical simulation calculation of a sine pulse distribution load that is a harmonic in the load waveform, as shown in Fig. 6.

Analysis of the simulation results under dynamic load
Change frequency Fig. 7 shows the distribution of plastic strain energy density at different frequencies.
(a) f = 20 Hz According to the experimental results in Fig. 7, the relationship curve between the maximum plastic strain energy density and frequency was drawn, as shown in Fig. 8. It can be seen from Fig. 8 that the distribution of the plastic strain energy density is mainly concentrated around the section of the roadway where the arch angle of the roadway is the largest. When the frequency is 20 Hz, the maximum plastic strain energy density is 1.77×10 6 J; when the frequency is 40 Hz, the maximum plastic strain energy density is 2.16×10 6 J; when the frequency is 60 Hz, the maximum plastic strain energy density is 1.44×10 6 J; when the frequency is 80 Hz, the maximum plastic strain energy density is 1.15×10 6 J.
Therefore, the order of magnitude of the plastic strain energy density caused by the frequency factor is: 40 Hz > 20 Hz > 60 Hz > 80 Hz; therefore, when a = 2.0 m/s 2 , θ = 180°, H = 600 m, and f = 40 Hz, the tendency of the surrounding rock to impact the roadway is higher.
Changing the incident angle of the shock wave Fig. 9 shows the distribution of the plastic strain energy density under different shock wave incidence angles.
(a)θ = 0° Fig. 9. Distribution of the plastic strain energy density under different shock wave incidence angles.
Based on the experimental results in Fig. 9, the relationship between the maximum plastic strain energy density and the impact position was drawn, as shown in Fig. 10. It can be seen from Fig. 10 that when θ = 0°, the maximum plastic strain energy density is 1.09 × 10 6 J.
When θ = 45°, the maximum plastic strain energy density is 1.67 × 10 6 J. When θ = 90°, the maximum plastic strain energy density is 1.72 × 10 6 J. When θ = 180°, the maximum plastic strain energy density is 2.16 × 10 6 J. Therefore, the order of the magnitude of the plastic strain energy density caused by the impact position is: 180° > 90° > 45° > 0°; therefore, when a = 2.0 m/s 2 , f = 40 Hz, H = 600 m, and θ = 180°, the tendency of the surrounding rock to impact the roadway is higher.
Changing the buried depth of the roadway Figure 11 shows the distribution of the plastic strain energy density at different depths. Changing the shock wave acceleration Changing the cross-section shape of the roadway It can be seen from Fig.15 that when the section shape is rectangular, the maximum plastic strain energy density of 5.39 × 10 6 J is mainly concentrated at the four arched corners of the roadway. When the section shape is circular, the maximum plastic strain energy density of 4.96 × 10 6 J is mainly concentrated on the roof of the roadway. When the section shape is a straight wall arch, the maximum plastic strain energy density of 5.20 × 10 6 J is mainly concentrated in the roof and arch foot positions of the roadway. When the section shape is horseshoe-shaped, the maximum plastic strain energy density of 6.06 × 10 6 J is mainly concentrated at the arch of the roadway floor.
Therefore, the order of magnitude of the plastic strain energy density caused by different section shapes is: horseshoe section > rectangular section > straight wall arch section > circular section; therefore, when a = 2.0 m/s 2 , f = 40 Hz, H = 1000 m, θ = 180°, and the roadway section is horseshoe-shaped, the tendency of the surrounding rock to impact the roadway is higher.

Conclusion
In this study, rock bursts in the Yi coal mine area are analyzed, and the response characteristics of the surrounding rock of the roadway under dynamic load are studied using a numerical simulation method; the following results are obtained: (1) The main causes of rock bursts are the damage to the overlying gravel strata caused by coal mining and an unreasonable mining layout.
(2) The duration of the shock wave is more than 0.06 s, and the proportion of low-frequency components is large; the main frequency is concentrated in the range of 5-50 Hz.
(3) Taking the distribution characteristics of plastic strain energy density to judge the tendency for rock bursts to occur on the roadway, this study shows that when a (vibration wave amplitude) = 2.0 m/s 2 , f (vibration wave frequency) = 40 Hz, H (roadway buried depth) = 1000 m, θ (the angle between the impact position of the seismic wave and the center of the roadway) = 180°, and the roadway section is horseshoeshaped, the tendency of the surrounding rock to impact the roadway is higher. Under the same conditions, the impact tendency of the surrounding rock on the roadway is the smallest and second smallest when the roadway is circular and straight-wall arched, respectively.