Nothing Special   »   [go: up one dir, main page]

Next Article in Journal
Paleoenvironmental Transition during the Rhuddanian–Aeronian and Its Implications for Lithofacies Evolution and Shale Gas Exploration: Insights from the Changning Area, Southern Sichuan Basin, South-West China
Previous Article in Journal
Characterization of Humic Acid Salts and Their Use for CO2 Reduction
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Microseismic Monitoring and Disaster Warning via Mining and Filling Processes of Residual Hazardous Ore Bodies

1
School of Resources and Safety Engineering, Central South University, Changsha 410083, China
2
State Key Laboratory of Safety Technology of Metal Mines, Changsha Institute of Mining Research Co., Ltd., Changsha 410012, China
3
Key Laboratory of the Ministry of Education of China for High-Efficient Mining and Safety of Metal Mines, School of Civil and Resource Engineering, University of Science and Technology Beijing, Beijing 100083, China
4
China Academy of Safety Science and Technology, Beijing 100012, China
*
Author to whom correspondence should be addressed.
Minerals 2024, 14(9), 948; https://doi.org/10.3390/min14090948
Submission received: 19 August 2024 / Revised: 10 September 2024 / Accepted: 17 September 2024 / Published: 18 September 2024
Figure 1
<p>Project overview. (<b>a</b>) Mine location. (<b>b</b>) Ore body occurrence. (<b>c</b>) Goaf. (<b>d</b>) Pillar.</p> ">
Figure 2
<p>Sensor position relationship and monitoring system network topology diagram within the microseismic monitoring area.</p> ">
Figure 3
<p>Simplified diagram of residual ore body mining.</p> ">
Figure 4
<p>The eastern and western parts of the mine and the overall ore output. (<b>a</b>) 2013–2018. (<b>b</b>) In the months of 2018.</p> ">
Figure 5
<p>Comparison chart of actual ore production and filling volume underground.</p> ">
Figure 6
<p>Spatial distribution of underground microseismic positioning events from 2013 to 2018.</p> ">
Figure 7
<p>The magnitude frequency cumulative relationship and its fitting formula for microseismic positioning events.</p> ">
Figure 8
<p>Probability distribution of cumulative occurrence of extreme magnitude values in 2019.</p> ">
Figure 9
<p>Probability density distribution of extreme magnitude events in 2019.</p> ">
Figure 10
<p>Probability index (P) distribution of microseismic energy release in 2019.</p> ">
Figure 11
<p>Establishment of precursor patterns: (<b>A</b>) Time window for roof collapse occurrence; (<b>B</b>) Microseismic event rate curve; (<b>C</b>) Time window for roof collapse occurrence.</p> ">
Figure 12
<p>37# probe area empty zone distribution map.</p> ">
Figure 13
<p>Trend chart of microseismic event rate for probe 37#.</p> ">
Figure 14
<p>Photos of roof-caving site.</p> ">
Figure 15
<p>Description of the destruction of the goaf where the 43# probe is located.</p> ">
Figure 16
<p>Microseismic event rate variation near probe 43#.</p> ">
Versions Notes

Abstract

:
The thick ore bodies in the Xianglushan tungsten mine have been irregularly mined, forming a super large, connected irregular goaf group and tall, isolated irregular pillars inside. At the same time, there is a production capacity task of recovering residual and dangerous ore bodies. This poses the potential for serious ground-pressure disasters, such as roof caving, pillar collapse, and large-scale goaf collapse during mining. Based on the actual needs of the site, we established a microseismic monitoring system. After analyzing the mining and filling processes and their relationships, and, combined with the distribution characteristics of microseismic multiple parameters, we constructed a ground-pressure disaster warning mode and mechanism. We analyzed the stability of the goaf, further formed a warning system, and achieved disaster warning. In response to the current situation of the difficulty of early warning of ground pressure in the Xianglushan tungsten mine, continuous on-site monitoring of existing goaves, point pillars, and strip pillars, as well as analysis of stress changes during dynamic mining and filling processes, we explored scientific and reasonable early warning mechanisms and models, understanding the relationship between the changes in microseismic parameters during dynamic mining and filling processes and ground pressure, studying and improving the reliability of underground microseismic monitoring and early warning, and achieved the internal connection between building early warning systems and the prevention of ground-pressure disasters. The results indicate that the mining and filling process of the ore body is the main factor in maintaining a stable and balanced distribution of underground ground pressure in mining engineering. Microseismic monitoring can invert the evolution of ground pressure and form a feedback system with ground-pressure warning, achieving mine safety management.

1. Introduction

The amount of goaf formed by global underground mining increases rapidly with the increase of mining scale, and the threat of ground-pressure disasters caused by goaf is becoming increasingly serious [1,2,3,4,5]. Especially in metal mines, which constitute the main strategic reserve resources, there are a large number of abandoned mines. With the widespread promotion of backfill mining technology and techniques, goaf disasters have been alleviated to a certain extent [6]. However, due to limitations in technology and equipment, there are still safety hazards in mine goafs that have not been properly resolved. In recent years, many scholars have proposed various warning methods for disasters such as ground pressure in the goaf [7] which have achieved some results in preventing goaf collapse and instability of mining pillars, thereby optimizing on-site mining and filling processes and promoting safe production.
In terms of combining microseismic monitoring and goaf research, many scholars have proposed different new methods [8,9,10]. Among them, Liu C.’s research is the most typical [9]; an in-depth analysis of the relationship between three-dimensional microseismic localization event groups, fault zones, and structural collapse patterns was conducted. In terms of the combination research of microseismic monitoring and mining and filling processes, good progress has also been made [11,12,13,14,15,16,17,18,19,20]. Among them, Zhao Y.’s research is the most outstanding [20]; it explains the inherent mechanism and connection between water inrush and microseismic monitoring during mining processes. A method for extracting crack evolution under different failure mechanisms was mainly proposed, which improved the applicability of microseismic monitoring data in this type of problem. In terms of microseismic monitoring and ground-pressure warning research, domestic and foreign scholars have studied and summarized some methods [21,22,23,24,25,26,27,28]. Among them, Lebert F. [21] used microseismic monitoring technology for underground mining collapse to record microseismic precursor signals that may characterize the onset of rock failure to determine its stability. Furthermore, Liu J. [29] and others believe that the goaf near the working face is the dominant factor in the occurrence of disasters, posing a serious threat to mine safety production. It mainly combines microseismic monitoring technology and numerical simulation methods to systematically analyze the stress law of the surrounding rock in the goaf and reveal the precursor characteristics of ground-pressure disasters in adjacent working faces of the goaf.
In recent years, microseismic monitoring technology has become increasingly mature in the stability monitoring and disaster warning applications of rock mass structures, and microseismic activities related to underground mine ground-pressure warnings have been widely studied [30,31,32,33]. The internal damage of rock mass induced by mining changes its compressive strength, especially in the large amount of goaf left behind, which is prone to roof collapse and shear failure of pillars [34,35]. In the process of underground mining, microseismic monitoring can reproduce the occurrence of microdamage and macro ground-pressure disasters in rock masses. Meanwhile, we can analyze the inherent relationship between multiple parameters of microseismic monitoring and ground-pressure activity, thus establishing a warning mechanism for ground-pressure disasters and microseismic monitoring parameters. However, due to the complex causes, long activity cycles, and high randomness of ground-pressure disasters in mining goaves, many early warning methods for ground-pressure disasters still cannot play a key role in stabilizing goaves under large-scale mining and filling alternation. In general, the failure of goaf in underground mines is closely related not only to the stability of the roof-surrounding rock and supporting pillars but also to the mining scale and filling volume. From the perspective of monitoring and early warning of ground-pressure disasters, the pre-disaster mining and filling process, as well as the multi-parameter interpretation of microseismic monitoring, should be the focus of our attention [36]. On the one hand, the synergistic effect of mining progress and filling scale can maintain the dynamic stability of ground pressure, mainly manifested in the maintenance and inhibition effect of cemented filling on rock-mass damage to a certain extent. However, the constraints on mining progress are production tasks, while the constraints on filling scale are process and cost. The inherent contradiction of this economic benefit hinders the relationship between the mining and filling process and ground-pressure disasters. On the other hand, the conventional application of microseismic monitoring is real-time online and all-round monitoring of passive-rupture sources [37,38,39]. The study of passive-rupture source localization and mechanical characteristics analysis through microseismic monitoring systems has serious lag. This will lead to a lag in our research on disaster prediction. Therefore, we need to use deep interpretation of microseismic parameters to extract warning indicators from key parameters. In summary, by analyzing the evolution patterns of historical ground-pressure records and existing microseismic data, we can grasp and evaluate the fragmentation of surrounding rocks and pillars, as well as areas with weak mechanical properties, especially by studying comprehensive and novel monitoring and early warning methods based on the scale of on-site mining and filling, to achieve early warning of ground-pressure disasters [40,41].
Based on the above analysis, to study a new method for early warning of goaf and ground-pressure disasters in mining engineering, we focus on a detailed analysis of and research into the mining and filling processes and microseismic monitoring parameters. Taking the Xianglushan tungsten mine as an example, we conducted microseismic monitoring and disaster warning research on the mining and filling process of residual hidden danger ore bodies. Firstly, an analysis was conducted on the mining and filling process of residual ore bodies in the goaf, mainly including the mining scale, filling volume, and the proportional relationship between the mining and filling processes. This laid the foundation for early warning methods from the perspectives of mining technology and proactive risk avoidance. Secondly, multi-parameter analysis and interpretation of microseismic monitoring were carried out through the established multi-channel microseismic monitoring system in the Xianglushan tungsten mine. This mainly includes the b-value parameter in the frequency magnitude cumulative relationship, the extreme value distribution of the maximum magnitude based on the Gumbel distribution model, and the multiple analysis of the probability relationship between microseismic release energy and rock mass stability. This is necessary to build practical and feasible early warning models and mechanisms and fully demonstrate their reliability. Finally, through a typical case study of the mine, practical analysis and verification were conducted, using local and small-scale microseismic event parameters as examples to study the potential disaster warning methods and their practicality in the mining process of residual ore bodies. This provided a reference and basis for the stability analysis and disaster warning of residual ore body mining in large-scale and complex goaf areas.

2. Engineering Background and Microseismic Monitoring

2.1. Summary of Engineering Background

The Xianglushan tungsten mine is located in the northwest of Jiangxi Province, China (as shown in Figure 1a). The mine has abundant reserves of scheelite resources (discovered in 1958 and put into operation in 1991), making it the second-largest scheelite source discovered in China [42]. The occurrence of ore bodies and anticline structures in the mining area are shown in Figure 1b. Its morphology is variable, and the rock strata are widely metamorphosed, resulting in significant differences in the distribution of ground pressure in mines [42]. In addition, due to the special working conditions of early mining methods, mining scale, and residual ore bodies, since 2013, a large number of goaf areas (as shown in Figure 1c) have been left underground, and the key pillars inside them (as shown in Figure 1d) have been severely damaged and destroyed. Thus, the foundation of disaster-warning research for analyzing and studying the mining of residual hidden dangers in ore bodies has been established. They are mainly using microseismic data for monitoring and early warning of mine ground pressure, to ensure safe production and provide guidance for the management of the goaf.

2.2. Monitoring Network

To reproduce and analyze the ground-pressure manifestation patterns in key areas of the mine, we have built and optimized a microseismic monitoring system at the Xianglushan tungsten mine. Figure 2 shows the topology structure of the main monitoring areas and the microseismic monitoring system in the mine.
In terms of horizontal distribution, it mainly includes two major regions: the east (oval shaped by blue dashed lines) and the west (oval shaped by red dashed lines). Between them, based on the occurrence characteristics of the ore body and the mining area, it is more detailed and divided into four regions.
We divided the eastern and western regions based on the mining stage and scale, using the 16 # exploration line as the boundary. A total of 24 new sensors have been added. We allocated and installed 18 sensors for amplification between lines 16 and 18 in the western region. We installed 6 sensors for amplification east of 16 lines. The sensors in the eastern area mainly cover the pillars of thick and large goaves. As shown in Figure 2, the cylinder is a sensor numbered from 1 to 84. The expanded microseismic monitoring network has a total of 84 sensors.
Figure 2 shows the placement and distribution of sensors in different zones. This topology consists of sensors for signal acquisition hardware, Paladin for data acquisition and processing systems, dedicated cables for signal transmission and time synchronization systems, software for comprehensive positioning and parameter calculation systems, a “brain” processing system for remote collaborative monitoring, and a machine subsystem system for information transmission and network sharing systems [42].

3. Analysis of the Mining and Filling Process of Residual Ore Bodies in the Goaf

3.1. Mining Scale

The main mining area of the mine is divided into two parts: east and west. The mining method in the eastern residual mining area is mainly a comprehensive mining method with irregular point pillars, as shown in Figure 3. The mining method in the western regular mining area is the subsequent cemented filling mining method.
We calculated the annual output of the eastern and western parts of the mine separately, including the dynamic alternation relationship of specific mining points, mainly to fully understand the transfer and concentration characteristics of the overall ground pressure affected by mining. At the same time, on the plane, we determined the overall spatial direction and sequence of the underground mining points. This ensured the timeliness and regionality of ground-pressure warning and avoided the disorderly stacking and multiple alternating concentrations of ground pressure in the mining area that may affect the monitoring and warning effectiveness of ground pressure.
Figure 4a shows the mining output of the east and west underground and the overall ore body of the mine in each year from 2013 to 2018. Every year, the number of mining sites in the eastern residual mining area is about 7–8, and there is no unified planning for mining sites in the eastern residual mining area. The number of mining sites in the western regular mining area is about 10–11.
Figure 4b shows the monthly mining output of the eastern residual mining area and the western regular mining area in 2018. The western region accounts for 58% of the total ore production, while the eastern region accounts for 42%. Under normal production months, the total ore production is approximately 50,000 tons per month.

3.2. Filling Volume

The filling in this article is a process in which the cemented filling material is transported through a filling pipeline to the underground mining area, where it undergoes consolidation and dehydration and finally forms a cemented solid to fill the original goaf. The main purpose of this study is to investigate and statistically analyze the filling materials during the research period and to conduct a detailed, comprehensive comparison and analysis of the ground-pressure manifestation. On the one hand, it is necessary to fully understand the specific location of ground-pressure phenomena such as roof caving and pillar burst, as well as the distance distribution and related relationship between the filling mining area. On the other hand, it is necessary to determine the degree of microseismic activity (decrease or increase) around the filling body. Overall, analyzing specific quantitative filling data to compare and verify the mutual influence of filling on overall ground pressure and local mining pressure is essential. This lays the foundation for the subsequent analysis of the distribution of ground pressure under the comprehensive, synergistic effect of mining and filling and the realization of macroscopic ground-pressure warning.
Filling affects the distribution of ground pressure, and the quality of filling directly determines the effectiveness of ground-pressure control. According to existing statistical data, the amount of underground filling carried out in each year from 2013 to 2018 shows that the filling height of each mining site varies depending on the height of the mining site. The top connection area of each filling area underground is relatively small, and it is only carried out in areas that are greatly affected by mining. The topping rate of cemented filling in the east is very low, with almost no topping, and the filling height is about 10 m. The topping rate of cemented filling in the western region is 50% to 60%, and the height of the filling is equal to the thickness of the ore body.
The actual filling volume completed underground each year during the research period is shown in Figure 4a. It can be seen that after trial operations from 2013 to 2015, the filling volume from 2016 to 2018 remained stable at 320,000 m3 per year. As of the end of 2018, the total underground filling volume was 1.5636 million m3; the total filling amount in the eastern residual mining area was 1.1766 million m3, and the total filling volume in the western region was 387,000 m3; the filling amount in the eastern residual mining area accounts for approximately 75% of the total filling amount.
The actual underground filling volume in the eastern residual mining area and the western regular mining area in 2018 is shown in Figure 4b. The actual filling capacity per month is about 30,000 to 40,000 m3, and the eastern filling volume accounts for approximately 60% of the total filling capacity.

3.3. The Process of Mining and Filling and Its Proportional Relationship

The exposed area of the goaf covered by the filling body during the research period is shown in Table 1. In the eastern residual mining area, the filling material covers a total of approximately 129,000 m2 of goaf, while in the western area, it covers a total of approximately 26,000 m2 of goaf. The total exposed area of the covered goaf is about 155,000 m2, and the total exposed area underground is about 338,300 m2. Therefore, the exposed area of the filled goaf accounts for about 45.82% of the total exposed area. According to the predetermined filling and production capacity, the newly added coverage area increases by about 10% annually.
Based on the annual production capacity and filling capacity of the ore, it is assumed that ∆Vr is the volume of underground goaf reduction each year. ∆Ve is the volume of goaf increase caused by the annual underground mining volume. ∆Vf is the volume of filling material transported underground each year. The following relationship holds:
Δ V r = Δ V f Δ V e
The decrease in the total volume of goaf underground is shown in Table 2. It can be seen that from 2013 to 2018, the amount that the goaf increased due to raw ore production was 928,000 m3, and during this period, the volume of underground goaf filled due to cemented filling was approximately 1.5636 million m3. Overall, during this period, the total volume of underground goaf decreased by approximately 632,900 cubic meters. At the end of 2012, the total amount of goaf was about 3 million m3, so the percentage of reduced goaf volume was 21.1%. With the existing production capacity and filling capacity of the ore remaining unchanged, the percentage of reduced goaf volume per year was 5% to 6%.
Figure 5 is a comparison chart between the actual ore output and the filling volume of the mine in 2018. As of December 2018, the cumulative filling volume of the entire mine is 1.5636 million m3. Due to the reduction in mining scale, the total volume of goaf is gradually decreasing, but the overall reduction is not significant. It should be noted that in the eastern residual mining area, the filling of goaf is carried out in a scattered and fragmented manner. The already-filled bodies are not connected as a whole. There is a common phenomenon of unfilled or non-topped filling in the filling area. By the end of 2018, the scale of goaf–goaf still exceeded 2.2 million m3. Although the expansion of goaf has been effectively controlled and is gradually decreasing, the huge scale of goaf still exists. This leads to a close correlation between the concentration and migration of ground pressure and the uneven mining and filling process and its relationship.

4. Microseismic Data and Ground-Pressure Warning Analysis

4.1. Multi-Parameter Analysis of Microseismic Monitoring

4.1.1. Statistics and Evaluation of b-Value

  • Theoretical basis for frequency magnitude relationship and b-value
The maximum likelihood value of parameter b in the cumulative distribution function of the frequency magnitude relationship β ^ . The premise of the calculation is that there is a frequency magnitude relationship, as follows:
lg n = a b m
In the above equation, if n is the number of microseismic events with a magnitude greater than or equal to m within a specific period, then the above equation is the cumulative relationship between frequency and magnitude. If n is the number of earthquakes with a magnitude of around m within a specific period, then the above equation is the density law relationship between frequency and magnitude. The physical meaning of parameter a is a measure of the level of microseismic activity. The physical meaning of parameter b is the relative number of small and large earthquakes within a given period.
This article draws on the effective statistical methods introduced by previous researchers for evaluating parameter b. Assuming that the magnitudes of microseismic events are independent of each other and have the same random distribution variation and that the magnitudes are continuous and have no upper limit, the frequency magnitude relationship of Gutenberg–Richter can be expressed as follows:
F m = 0 , m < m m i n 1 1 e β m m m i n m i n ,     m m m i n
F(m) is the cumulative probability distribution function of magnitude m, which is considered a continuous variation value greater than the threshold value of m min at any time. If the magnitude is considered to be independent, then the joint probability density of N sets of magnitudes of mi(i = 1,2,…, N) is equal to the product of each probability density. Let the maximum likelihood calculation of parameter b be β ^ . The value, which is the maximum value of the likelihood function L(·), can be determined as follows:
L b m i , . . . , m N = c o n s t i = 1 N f m i b m a x
Or equivalent to
i = 1 N β ln f m i β = 0
Taking f m i β logarithm and calculating its partial derivative, we can write the maximum likelihood condition as follows:
N b i = 1 N m i N m m i n = 0
Performing the above equation yields the following:
b ^ = 1 m m m i n
In the above equation:
m = i = 1 N m i N
According to the central limit theory, for sufficiently large N, b ^ is approximately normally distributed near its mean value.
b ^ standard deviation is as follows:
s ^ b = b 2 ln L b 2 1 / 2 = b ^ / N
The standard deviation of parameter b value is as follows:
s ^ b = s ^ b ln 10 = b ^ ln 10 N
2.
Statistical analysis of magnitude and energy
Between 2013 and 2018, 122 microseismic positioning events were detected underground in the mine (e.g., Figure 6). Detailed information on the spatial distribution positions of these 122 microseismic positioning events is shown in Table 3, which lists in detail the occurrence time, energy magnitude, and current earthquake magnitude of each positioning event.
The dataset in Table 3 contains information on microseismic events from different dates from 2013 to 2018, covering changes in event energy and local magnitude. These data features can be used to study the regularity and trend of microseismic events, providing data support for disaster warning and prevention. By analyzing the characteristics of magnitude and energy in the given data, it can be found that the range of energy variation is wide: the energy values in the data are reflected from very small values (such as 0.01) to relatively large values (such as 4290.00), showing a large range of variation. The magnitude distribution is relatively concentrated: the magnitude values are concentrated between −1.25 and −4.54, with a relatively narrow overall distribution and no extreme values. For the correlation between time and magnitude/energy, observing the data reveals the possibility of multiple microseismic events occurring within a short period (such as one day or several consecutive days) with different magnitudes and energies. For outlier handling: magnitude −4.54 and energy 4290.00 may be outliers, and further research is needed to identify potential warning boundaries during data collection and warning processing. For time series analysis: trend analysis can be performed on time series data to understand the changes in magnitude and energy over time and whether there are seasonal or periodic patterns of change. Overall, it contributes to more effective monitoring and early warning of microseismic activity and potential risks.
3.
Analysis of the magnitude, energy, and b-value of the mine
By calculating the magnitude frequency cumulative relationship for the 122 microseismic positioning events mentioned above, the following calculation results can be obtained:
The minimum local magnitude is as follows:
m = 4.54 m i n
The cumulative relationship between magnitude and frequency is as follows:
y = lgn = a bm
In the above equation: a = −0.0601, b = 0.555.
As:
β = bln 10 2.3 b = 2.3 × 0.555 = 1.2765
The cumulative probability distribution function F(m) of magnitude m is as follows:
F m = 0                             m < m m i n 1 1 e 1.2765 m + 4.54             m m m i n
The average magnitude is as follows:
m = i = 1 N m i N = 2.84
Parameter β maximum likelihood calculation value of β ^ for:
β ^ = 1 m m m i n = 1 2.84 + 4.54 = 0.5882
The standard deviation of parameter b value is as follows:
σ ^ b = σ ^ β ln 10 = β ^ ln 10 N = 0.5882 2.3 × 11.0454 = 0.0232
The magnitude frequency cumulative relationship and linear fitting results of 122 microseismic positioning events from 2013 to 2018 are shown in Figure 7. According to the calculation results, the parameter b, representing the relative number of small and large earthquakes, is 0.555. This value is relatively low, indicating that there are relatively more small earthquakes and relatively fewer large earthquakes. On the one hand, it indicates that the intensity of microseismic activity underground during this period is relatively low. On the other hand, when regressing the average value of parameter b, there will be a relative increase in the number of major earthquakes. This also indicates that in the next step of mining activities, the underground microseismic activity may be further enhanced.

4.1.2. Extreme Distribution of Maximum Magnitude Based on the Gambell Distribution Model

Low-magnitude earthquake records are relatively incomplete, leading to deviations from specific prediction results based on frequency magnitude relationships. On the contrary, the magnitude of the largest microseismic event is relatively easy to understand, with a clear timeline, and it is easier to accurately determine than small earthquakes. Usually, they play a major role, such as the energy they release and the seismic moment that may cause harm. Therefore, it can be expected that the method of prioritizing the analysis of the strongest microseismic events is more effective than relying solely on complete data.
This section intends to use relevant theoretical analysis to determine the following relationship: based on the cumulative relationship between past magnitudes and frequencies between 2013 and 2018 to estimate the probability of a certain magnitude event occurring in 2019.
The extreme value theory of magnitude extreme distribution based on the Gumbel distribution model Fmax maximum magnitude extreme value distribution characteristics analysis is as follows:
Let x be a random variable with a definite cumulative probability function F(x), and n mutually independent samples, with the same distribution function F(x). If x is the maximum value in x 1 , . . . , x n , then the probability of this event is as follows:
F x x 1 x , x 2 x , . . . , x n x [ F ( x ) ] m a x n
The above equation is the exact distribution function of extreme values.
According to Gumbel’s research, there are only three types of distribution functions. Their asymptotic form is equivalent to the above relationship, and each distribution function assumes a specific form of the absolute maximum value of the variable x.
Gambell Type I distribution:
F I x 1 e a x u m a x
In the above equation, a and u are distribution parameters, and a > 0.
Gambell type II distribution:
F I I x u x m i n x x m i n k m a x
In the above equation, k > 0, u > xmin > 0, and xmin is the lower limit of x.
Gambell III distribution:
F I I I x x m a x x m a x k m a x
In the above equation, k > 0 and u < xmax, where xmax is the upper bound of x.
Nordguist (1945) first applied extreme statistics to seismology and proved that the maximum earthquake that occurred in California was consistent with the Gambell Type I distribution. Epstein and Lomnitz’s proof [43]: assuming that microseismic events are generated by a simple Poisson process and follow the frequency magnitude relationship, the Gumbel Type I distribution can be directly derived.
For random variables with the same independent distribution, the maximum value of a fixed length n sequence satisfies the distribution of Equation (17), where n is defined as a random number with a probability distribution P(n), and the probability that the maximum value of the sequence does not exceed m is equal to:
P n F m n
Therefore, if every time period’s number of seismic events n follows a Poisson distribution, and its probability formula is as follows:
P n Δ t = λ Δ t n e λ Δ t n !
And, if the magnitude m is a random variable that follows the Gutenberg–Richter cumulative probability distribution function, then during the period Δ t , the cumulative distribution of the maximum magnitude m during the period is as follows:
F m Δ t n = 0 P n Δ t F m m a x n = n = 0 λ Δ t n e λ Δ t n ! F m n
Using the symbol “Λ”, Λ = λΔt F(m), due to n = 0 n / n ! = e , the cumulative distribution of the maximum magnitude m becomes the following:
F m Δ t λ Δ t e β m m a x
About β = a and Δ t = e a u , the above equation is exactly Gambell Type I distributed. After calculating λ, β afterwards, we can obtain useful microseismic activity safety characteristic values.
According to the above theoretical methods and existing basic data, the cumulative probability distribution function F(m) of magnitude m is as follows:
F m = 0                             m < m m i n 1 1 e 1.2765 m + 4.54             m m m i n
Based on this function formula, we can obtain the probability distribution functions for the extreme values of different magnitudes in 2019, including the cumulative probability distribution function and the probability density distribution function. Based on the actual situation of the mine, the range of magnitude extremes from −1 to +2, cumulative probability distribution function, and probability density distribution function curves were plotted, as shown in Figure 8 and Figure 9.
The calculation results show that the range of magnitude extremes with a probability density above 0.03 is approximately [−1, 0.3]. The cumulative probability of this interval is about 0.75, indicating that in 2019, the magnitude of microseismic positioning events occurring underground was more likely to be distributed in the [−1, 0.3] interval. For seismic sources with a magnitude of 0.3, their rupture scale is about 7–8 m.

4.1.3. Probability Analysis of Stability and Microseismic Energy Release

The strength of microseismic activity in mining areas is related to mining conditions and methods. Assuming that the energy released from the excavation area can serve as a criterion for observing signs of rockburst, high energy release often implies an increase in the number of rock fractures, implying the possibility of rockburst events occurring. This section intends to determine the following relationship through relevant analysis: the probability of seismic energy release caused by mining volume exceeding the given value in the next specific time interval year starting from January 2019.
McGarr [44] proved that microseismic activity induced by mining can be represented by the sum of seismic moments and by the volume of converging elastic deformation around mining. Later, based on the sum of the deformation volumes of the surrounding rock and the physical connection between the fractured rock volumes caused by seismic activity, the following transformations were made based on the relationship proposed by Kijko [45,46]:
M 0 = const θ Δ V
In the above equation, θ is a parameter that depends on the mining depth, mining method, mechanical properties of the mined rock, geometric shape of the goaf formed during mining, and type of mining support. When backfilling, the parameter θ is descending. When the hanging wall bends, or the bottom of the goaf is compressed, the parameter θ is rising.
To use the above formula for statistical evaluation of mine microseismic activity safety in mining practice, the value of seismic moment M0 will be replaced by the magnitude parameter of nearby small earthquakes. The magnitude of a small earthquake near the earthquake is directly proportional to the logarithm of the seismic moment, and the specific formula is as follows:
lg M 0 = p + q M L
In the formula, p and q are constants. For simple models of seismic sources and microseismic events at small seismic scales, the constant q approaches 1. Organize and obtain the following:
1 0 q M L = c o n s t θ Δ V
The assessment of seismic event intensity in mining areas is based on energy rather than magnitude, which can be determined by empirical formulas.
lg E = c + d M L
Then, the following equation is obtained:
E d / q = c o n s t θ Δ V
Accompanied by a certain volume Δ V , the total microseismic energy released by the excavation is not considered a measurable constant but can be well regarded as a random variable and estimated from a known cumulative energy distribution function.
We assume that microseismic events related to the extraction of a certain volume of ore and rock processes follow the Gutenberg–Richter relationship. The upper bound of the magnitude of the near earthquake is mmax, and the relationship between magnitude and energy follows the above equation. The number dN of microseisms in mines with energy E and energy interval (E, E + dE) can be expressed as follows:
d N = const E b / q + 1 d E
In the above equation, the constant = 1 0 a + c b / d / d ln ( 10 ) .
So, the total amount of released seismic energy is as follows:
E = const E m i n E m a x E b / d d E = const 1 b d ( E m a x 1 b / d m i n 1 b d )
In the above equation, Emin is the energy threshold, and all this above energy, determined by the mining volume events induced by Δ V , can be recorded.
ln M 0 = 1 B ln E + ln B c 2 c 3 c 1
In the formula, B and c3 are characteristic parameters of the rock mass state, and the calculation formula for the constants of each parameter is as follows:
B = d b / q b
We assume that the parameters during mining θ = const, then organize the above equation to obtain:
E = c o n s t Δ V B
It is worth noting that the above model has a flaw, which assumes that the stress in the rock mass is the result of mining disturbance while ignoring the role of tectonic stress. However, tectonic stress in the mining area is important, and only after certain modifications can the given model be successfully applied.
Assumption: replace ∑E with the symbol E and ΔV with V. Symbol ΔV represents a specific period’s volume of rocks and ore excavated. We define microseismic safety as a probability in the next specific period within Δt. The probability of seismic energy release Δ E caused by excavation volume Δ V exceeds the given value E 0 .
In the next time period Δti = ti+1 − ti, the expected release of energy is as follows:
Δ E i = C V i + 1 B C V i B
In the above equation, B and C can be obtained from earlier energy release calculations. The energy released Δ E i clearly has random characteristics. We assume that Δ E i follows a normal distribution, with zero truncation on the left and an average value of Δ E ; the standard deviation is σ E .
Therefore, the probability that the released seismic energy Δ E will exceed a specific value E 0 within the newly excavated V , V + Δ V volume is P; that is, the probability index P of microseismic activity and microseismic release energy in rock mining is as follows:
P = 1 2 π σ E 0 E 0 e 1 2 σ E 2 ( E Δ E ) 2 d E
In the above equation, Δ E is the average of the total expected seismic energy release during the time interval, and parameters B and C can be calculated from the volume of rock previously released and mined.
In this mine, the empirical relationship between moment magnitude and local magnitude can be rewritten as follows:
lg M 0 = 10.32 + 0.94 M L
The empirical relationship between microseismic energy and local magnitude is as follows:
lg E = 4.8 + 1.5 M L
According to the calculation results in the previous section, the value of b is 0.555.
According to the 2019 original ore production task plan of the mine, the tungsten concentrate production target for 2019 was 3800 tons. Compared to the ore production grade in 2018, the ore production in 2019 may increase from 450,000 tons in 2018 to 520,000 tons.
The cumulative microseismic energy release for the whole year of 2013 was 5388 J, and the cumulative microseismic energy release for the whole year of 2018 was 5091 J. We drew diagrams to illustrate the probability of an annual ore mining volume of 450,000 tons and the release of seismic energy exceeding a specific value of 5388 J, as well as the probability of an annual ore mining volume of 520,000 tons and the release of seismic energy exceeding a specific value of 5091 J, as shown in Figure 10.
From the graph, it can be seen that when the ore extraction amount in 2019 was 450,000 tons, the probability of the microseismic energy released throughout the year exceeding the 5091 joules in 2018 was 0.3. The probability of the microseismic energy released throughout the year exceeding the 5388 joules in 2013 is 0.27. When the ore extraction volume in 2019 was 520,000 tons, the probability of the microseismic energy released throughout the year exceeding 5091 joules in 2018 was 0.36. The probability of the microseismic energy released throughout the year exceeding the 5388 joules in 2013 is 0.35.

4.2. Construction of Warning Modes and Mechanisms

4.2.1. Microseismic Warning Precursor Mode

There have been nearly a hundred successful cases of using microseismic monitoring for monitoring and early warning of roof caving in Xianglushan tungsten mine, but there have also been dozens of missed cases. By analyzing and organizing the microseismic event rate curve and the occurrence time of roof collapse in each case, the precursor modes of using microseismic monitoring for short-term warning of roof collapse in mining areas can be broadly divided into three types: First, the occurrence of roof collapse after a rapid increase in microseismic event rate. Second, there were no roof-caving incidents after the rate of microseismic events surged. The third reason is that there has been a rapid increase in the rate of no microseismic events, but there has been a roof collapse event. The above three modes are respectively referred to with symbols A, B, and C, as shown in Figure 11.
For the microseismic event rate curve, t0 is the moment when the microseismic event rate begins to increase rapidly. The moment when the microseismic event rate returns to a normal level after the active period ends is t2. The duration of the active period of microseismic events is defined as T1 = t2 − t0. The time window for roof collapse indicates that roof collapse may occur at any moment within this time window. The initial time window for roof collapse is t1 and the end time is t3. Therefore, the length of the roof collapse time window is defined as T2 = t3 − t1.
Precursor mode A is the most common type, in which T1 typically lasts for several days to weeks. The time interval between t1 and t0 is generally several hours to several weeks. The time point t1 is sometimes before the time point t2 and sometimes after it. The duration of T2 can sometimes reach several months. There are fewer cases of precursor mode B, while there are slightly more cases of precursor mode C. In precursor mode B, the duration of T1 is generally several days to weeks or even months.

4.2.2. Mechanism of Precursor Mode

The use of microseismic monitoring methods for short-term monitoring and early warning of roof collapse in mining areas aims to achieve the locking and analysis of the specific location, time, and scale of roof collapse in mining areas. The problem of microseismic monitoring and early warning is similar to the prediction of natural earthquakes in terms of scientific essence and key issues. The short-term prediction of natural earthquakes is an unresolved problem in the world. The foreshock activity of natural earthquakes varies greatly with different situations. The sequence of foreshocks and earthquakes can be roughly divided into three types: main shock type, swarm type, and isolated type, ranging from a single event to a swarm type. Different seismic types are caused by the stress mode of rocks, the uniformity of rock structure, the historical stress of rocks, the cementation process after rock fracture, and the interaction of water, either alone or in mutual coupling.
Compared to natural earthquakes, the depth of the epicenter is generally several thousand meters or more. For roof caving in the mining area, we can conduct a close-up survey of the primary weak surface and its lithology, as well as the influence of water and other factors. This allows us to analyze the causes and underlying mechanisms of different precursor patterns. After clarifying the mechanisms of different precursor modes, with the help of theoretical foundations, we can monitor, warn, and treat complex precursor modes such as roof collapse and then develop corresponding preventive measures for the occurrence of roof collapse events.
Through the detailed analysis of the typical cases mentioned above, we found that in precursor mode A, the fracture surface of the roof caving is sometimes newly formed and more often develops along the primary fractures or weak planes. The combination of fracture surface and free face results in a zero constraint degree of the collapsed rock mass, allowing it to move freely towards the free face, which is the mechanism of precursor mode A. The mechanism of precursor mode B is that the incubation and expansion of the fracture surface have generated many microseismic events, but the formed fracture surface and free face cannot allow a certain part of the rock mass to form an unconstrained free body, and the occurrence of roof-caving events is temporarily delayed. In precursor mode C, the fracture surface is the primary weak plane in the collapsed rock mass. The original weak surface is filled with weak materials, or the rock on the fracture surface is weakened and mud is formed due to the weathering of water. This results in the failure of the fracture surface to release elastic waves during the expansion process due to the inability to gather elastic strain energy. Therefore, before the roof falls, the surrounding probes cannot detect any microseismic events. The weathering of the primary soft filling of groundwater is the basis for the mechanism of precursor mode C. In addition, in some cases of missed reports of roof caving, the probe is far away from the roof-caving point, or multiple pillars and voids are blocking the probe from the roof-caving point. Although the rupture of the roof slab releases elastic waves in all directions, it cannot be detected due to attenuation, which is a limitation of engineering conditions.

4.3. Early Warning and Reliability Analysis

The mechanism of roof and pillar collapse in goaf is complex and variable. At the same time, due to the complexity of the geometric shape of the goaf underground, it is scientifically impossible to achieve a success rate of 100% and a failure rate of 0% in early warning and prediction of all roof-caving and collapse events from the perspective of microseismic monitoring. After practical application in Xianglu mountain, the success rate of early warning and forecasting is about 84.3%, and the missed reporting rate is about 33.4%. The objective reasons and theoretical reality for this situation are as follows:
Monitoring of the impact of goaf obstruction: the complex geometric shape of goaf and pillars results in significant attenuation of precursor rupture signals due to goaf obstruction, even if the distance between the sensor and the roof falling slope is within 50 m and within the monitoring range, which may lead to missed reporting of the roof falling slope events.
The obstruction of joint structure to signal propagation: The roof and pillars of the goaf are exposed to external forces for a long time, such as weathering, groundwater erosion, blasting vibration, etc. These dynamic loads constantly change in their duration and magnitude. In the area where roof caving occurs, the rock mass joints are more developed, resulting in the accumulation of numerous small fractures to form a macroscopic fracture surface before the final roof caving. These rupture signals are locally aggregated into surfaces in space, with varying periods. If the joints in the surrounding rock mass are developed, even if the precursor rupture signal is generated, it is difficult to effectively propagate, which may lead to missed monitoring of roof-caving and slope-collapse events.
Unpredictable toppling incidents: Some toppling incidents did not show significant precursor rupture for several days to several hundred days before the final occurrence. During the long-term evolution process, the bonding force between the detached rock mass and the parent rock mass gradually weakens, and the gravity of the detached rock mass tends to balance. Under external forces such as blasting vibration and impact, the bonding force of rock mass in the ultimate equilibrium state may suddenly be lower than gravity, leading to macroscopic roof collapse. Such incidents of roof collapse without obvious precursors are difficult to predict and, therefore, prone to being missed.

4.4. Warning Cases

4.4.1. 37# Probe Case

The 37# probe is installed inside a rectangular mining pillar with a plane distance of about 30 m from the roof-caving point, as shown in Figure 12. There are also probes 36# and 38# that are relatively close to the roof-caving point, with plane distances of 43 m and 57 m, respectively. Compared with the 37# probe, other pillars or voids block the position between the pillar and the roof, caving where these two probes are located.
On 8 October 2018, the number of microseismic events detected by probe 37# slightly increased compared to usual, reaching 10 per day, while the usual number was only about 1–2 per day. On 13 October, the microseismic event rate of probe 37# suddenly increased sharply to 130 per day, as shown in Figure 13. For the 36# probe, due to the barrier between the pillar and the goaf, the elastic waves released by the microfracture of the rock mass at the roof collapse were rarely sensed. The 36# probe only received a small number of microseismic events from 12 to 15 October, as shown in Figure 13.
On 13 October, the microseismic event rate of probe 37# significantly increased, and monitoring technicians issued a warning notice, suspending production operations at surrounding mining sites and evacuating production workers. The microseismic activity rate of probe 37# continued to be at a very high level on the 14th and 15th. On 16 October, underground safety inspectors conducted safety inspections, and a large-scale roof collapse event was discovered at the top plate position shown in Figure 14. The roof collapse is located on the transportation channel for entering and exiting the production site. The volume of the collapsed rock mass is about 8–10 m3. The lithology is limestone. The newly formed exposed surface is very smooth, and after exploration, a weak surface of carbonaceous mudstone is hidden before the roof rock mass falls. The newly formed exposed surface and falling rock mass of the roof rock mass are shown in Figure 14.

4.4.2. 43# Probe Case

The 43# probe is installed in an isolated point column, relatively close to the location of the top plate cracking. There is no other pillar or void between the location of the roof cracking and the pillar where the 43# probe is located, with a plane distance of about 15 m, as shown in Figure 15. The average span of the empty area where the 43# probe is located is 18.2 m, and the average height of the empty area is 16.3 m. It belongs to a tall, empty area, and there are no mining and blasting operations in the surrounding area, but filling operations are being carried out.
Starting from 5 June 2019, the microseismic event rate of probe 43# began to increase. From 16 June to the end of June, the number of microseismic events per day exceeded the warning value, with an average of 60–80 per day. In July, the average level was 40–60 per day. In August, the average level was 30–50 per day. From 1 to 10 September, the average level was 15–25 per day. The microseismic event rate exceeded the warning value for more than three consecutive months until after 9 September, when the microseismic event rate reached zero and then remained at a normal level, as shown in Figure 16.
After the abnormal increase in the microseismic activity rate of the 43# probe, mining technicians regularly visit the site for safety inspections and surveys. During the entire period of microseismic activity, only two small cracks were observed on the roof, as shown in Figure 15. There are no other abnormal phenomena, such as cracking on the roof and pillars, and there has been no roof collapse event.

5. Conclusions

In this study of on-site work and microseismic monitoring, the overview of the engineering location and the layout of the microseismic monitoring network were first analyzed. Subsequently, an in-depth analysis was conducted on the mining and filling process of residual ore bodies in the goaf, including mining scale, filling volume, and the relationship between the mining and filling process and proportion. Through an overall analysis of 64 cases, the regularity and specificity were revealed. Subsequently, combined with microseismic monitoring data, ground-pressure analysis was conducted, and a warning mode and mechanism were constructed using multi-parameter analysis methods. Based on this, research on warning and reliability was carried out, further verifying the effectiveness of warning. Through the verification and comparison of two warning cases, the value and practicality of the warning system have been further demonstrated. In the discussion section, further exploration of the above research results will be conducted to conclude, providing theoretical support and practical guidance for future work in related fields. The main conclusions are as follows:
(1)
By studying the engineering location overview and microseismic monitoring network in on-site work and microseismic monitoring, we have gained a deeper understanding of the characteristics of the geological environment and the deployment of monitoring systems. The analysis of the engineering location overview provides us with a comprehensive understanding of geological conditions and engineering structures, laying the foundation for subsequent research. The construction of microseismic monitoring networks provides us with real-time monitoring and data support, helping us to more accurately grasp the dynamic changes of underground microseismic activities and providing a solid foundation and support for the prevention and early warning of geological disasters.
(2)
The mining and filling process of residual ore bodies in goaf is a complex engineering system that requires comprehensive consideration of multiple factors such as mining scale, filling volume, and mining filling ratio. Through in-depth analysis of these factors and case studies, we can better understand the inherent laws of the mining and filling process, providing a scientific basis for achieving efficient resource utilization and safe environmental protection.
(3)
Microseismic monitoring technology is crucial for mine safety production. It improves the accuracy of predicting ground-pressure activities through multi-parameter analysis, provides a scientific decision-making basis for mine managers, and effectively reduces geological disasters. The constructed warning mode and mechanism can quickly respond to the initial occurrence of ground-pressure anomalies, ensuring the safety of personnel and equipment. Continuous reliability analysis and system optimization ensure the stable operation of the early warning system in complex environments, improving the accuracy and timeliness of early warning. The combination of these technologies and methods provides a solid guarantee for mine safety production. With the continuous development of technology, microseismic monitoring will play a greater role in preventing ground-pressure disasters.
(4)
By analyzing actual warning cases, the performance of the warning system can be further improved. Successful warning cases not only validate the effectiveness of the warning model but also provide us with valuable practical experience. By comparing and analyzing different cases, shortcomings in the early warning system can be identified, providing direction for future improvement and upgrading.

Author Contributions

Z.Z.: Conceptualization, Funding acquisition, Writing—Review and Editing; Y.H.: Writing—Original Draft, Data, Project administration. C.Z.: Methodology, Writing—Original Draft, Software, Visualization. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by National Natural Science Foundation grant number 52274249 & 52334003, and the China Postdoctoral Science Foundation.

Data Availability Statement

Data will be made available on request.

Acknowledgments

The authors would like to acknowledge the National Natural Science Foundation of China (Project No. 52274249, 52334003).

Conflicts of Interest

Author Yinghua Huang was employed by the company Changsha Institute of Mining Research Co., Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Feng, X.-T.; Young, R.P.; Reyes-Montes, J.M.; Aydan, Ö.; Ishida, T.; Liu, J.-P.; Liu, H.-J. ISRM Suggested Method for In Situ Acoustic Emission Monitoring of the Fracturing Process in Rock Masses. Rock Mech. Rock Eng. 2019, 52, 1395–1414. [Google Scholar] [CrossRef]
  2. Zhou, Z.; Chen, L.; Cai, X.; Shen, B.; Zhou, J.; Du, K. Experimental Investigation of the Progressive Failure of Multiple Pillar-Roof System. Rock Mech. Rock Eng. 2018, 51, 1629–1636. [Google Scholar] [CrossRef]
  3. Zhou, Z.; Zhao, Y.; Jiang, Y.; Zou, Y.; Cai, X.; Li, D. Dynamic behavior of rock during its post failure stage in SHPB tests. Trans. Nonferrous Met. Soc. China 2017, 27, 184–196. [Google Scholar] [CrossRef]
  4. Cai, X.; Zhou, Z.; Tan, L.; Zang, H.; Song, Z. Water Saturation Effects on Thermal Infrared Radiation Features of Rock Materials During Deformation and Fracturing. Rock Mech. Rock Eng. 2020, 53, 4839–4856. [Google Scholar] [CrossRef]
  5. Wang, F.; Ren, Q.; Jiang, X.; Jiang, A.; Zhao, C.; Liu, W. Engineering geology and subsidence mechanism of a mountain surface in the Daliang Lead-zinc Ore Mine in China. Bull. Eng. Geol. Environ. 2022, 81, 488. [Google Scholar] [CrossRef]
  6. Yuan, Z.; Ban, X.; Han, F.; Zhang, X.; Yin, S.; Wang, Y. Integrated three-dimensional visualization and soft-sensing system for underground paste backfilling. Tunn. Undergr. Space Technol. 2022, 127, 104578. [Google Scholar] [CrossRef]
  7. Bednarczyk, Z. Identification of flysch landslide triggers using conventional and ‘nearly real-time’ monitoring methods—An example from the Carpathian Mountains, Poland. Eng. Geol. 2018, 244, 41–56. [Google Scholar] [CrossRef]
  8. Jia, H.; Yan, B.; Yang, Z.; Yilmaz, E. Identification of goaf instability under blasting disturbance using microseismic monitoring technology. Geomech. Geophys. Geo-Energy Geo-Resour. 2023, 9, 142. [Google Scholar] [CrossRef]
  9. Liu, C.; Tang, C.A. Analysis of structure collapse regularity for overlying strata in stope based on microseismic technique. Mater. Res. Innov. 2011, 15, S522–S526. [Google Scholar] [CrossRef]
  10. Wei, L.; Qi, Q.; Li, H.; Zhang, B.; Wang, Y.; Kong, L. A Case Study of Damage Energy Analysis and an Early Warning by Microseismic Monitoring for Large Area Roof Caving in Shallow Depth Seams. Shock Vib. 2015, 2015, 709459. [Google Scholar] [CrossRef]
  11. Cai, W.; Dou, L.; Si, G.; Hu, Y. Fault-Induced Coal Burst Mechanism under Mining-Induced Static and Dynamic Stresses. Engineering 2021, 7, 687–700. [Google Scholar] [CrossRef]
  12. Contrucci, I.; Namjesnik, D.; Niemz, P.; Primo, P.; Kotyrba, A.; Mutke, G.; Konicek, P.; Dominique, P.; Rudolph, T.; Moellerherm, S.; et al. European feedback on post-mining seismicity. J. Sustain. Min. 2023, 22, 195–218. [Google Scholar] [CrossRef]
  13. Dong, L.; Yan, X.; Wang, J.; Tang, Z. Case study of microseismic tomography and multi-parameter characteristics under mining disturbances. J. Cent. South Univ. 2023, 30, 2252–2265. [Google Scholar] [CrossRef]
  14. Kinscher, J.L.; De Santis, F.; Poiata, N.; Bernard, P.; Palgunadi, K.H.; Contrucci, I. Seismic repeaters linked to weak rock-mass creep in deep excavation mining. Geophys. J. Int. 2020, 222, 110–131. [Google Scholar] [CrossRef]
  15. Li, P.; Wang, Z.; Chen, B.; Peng, J.; Zhang, B.; Zhao, J.; Xie, H. Rockburst hazard assessment and prevention and control strategies in the Ruihai gold mine. Front. Earth Sci. 2024, 11, 1348366. [Google Scholar] [CrossRef]
  16. Liang, Z.; Xue, R.; Xu, N.; Li, W. Characterizing rockbursts and analysis on frequency-spectrum evolutionary law of rockburst precursor based on microseismic monitoring. Tunn. Undergr. Space Technol. 2020, 105, 103564. [Google Scholar] [CrossRef]
  17. Palgunadi, K.H.; Poiata, N.; Kinscher, J.; Bernard, P.; De Santis, F.; Contrucci, I. Methodology for Full Waveform Near Real-Time Automatic Detection and Localization of Microseismic Events Using High (8 kHz) Sampling Rate Records in Mines: Application to the Garpenberg Mine (Sweden). Seismol. Res. Lett. 2020, 91, 399–414. [Google Scholar] [CrossRef]
  18. Wang, S.-C.; Dou, L.-M.; Wang, Z.-Y.; Bai, J.-Z.; Chai, Y.-J. Mechanism of Coal Bursts Induced by Horizontal Section Mining of Steeply Inclined Coal Seams and Application of Microseismic Multiparameter Monitoring in Early Warning. Adv. Civ. Eng. 2020, 2020, 1048624. [Google Scholar] [CrossRef]
  19. Yu, Y.; Zhu, B.; Guo, H.; Chen, B.; Geng, D. Warning Index Associated with Rock Burst in Deeply Buried Tunnels. Int. J. Geomech. 2021, 21, 04021211. [Google Scholar] [CrossRef]
  20. Zhao, Y.; Yang, T.; Zhang, P.; Xu, H.; Wang, S. Inversion of seepage channels based on mining-induced microseismic data. Int. J. Rock Mech. Min. Sci. 2020, 126, 104180. [Google Scholar] [CrossRef]
  21. Lebert, F.; Bernardie, S.; Mainsant, G. Hydroacoustic monitoring of a salt cavity: An analysis of precursory events of the collapse. Nat. Hazards Earth Syst. Sci. 2011, 11, 2663–2675. [Google Scholar] [CrossRef]
  22. Mercerat, E.D.; Driad-Lebeau, L.; Bernard, P. Induced Seismicity Monitoring of an Underground Salt Cavern Prone to Collapse. Pure Appl. Geophys. 2010, 167, 5–25. [Google Scholar] [CrossRef]
  23. Occhiena, C.; Coviello, V.; Arattano, M.; Chiarle, M.; di Cella, U.M.; Pirulli, M.; Pogliotti, P.; Scavia, C. Analysis of microseismic signals and temperature recordings for rock slope stability investigations in high mountain areas. Nat. Hazards Earth Syst. Sci. 2012, 12, 2283–2298. [Google Scholar] [CrossRef]
  24. Siddhamshetty, P.K.Z. Modeling of Hydraulic Fracturing and Design of Online Optimal Pumping Schedule for Enhanced Productivity in Shale Formations. Ph.D. Thesis, Texas A&M University, College Station, TX, USA, 2020. [Google Scholar]
  25. Zhai, J.; Wang, Q.; Yuan, D.; Zhang, W.; Wang, H.; Xie, X.; Shahrour, I. Clogging Risk Early Warning for Slurry Shield Tunneling in Mixed Mudstone-Gravel Ground: A Real-Time Self-Updating Machine Learning Approach. Sustainability 2022, 14, 1368. [Google Scholar] [CrossRef]
  26. Wang, X.Y.; Ma, Z.; Zhang, Y.T. Research on Safety Early Warning Standard of Large-Scale Underground Utility Tunnel in Ground Fissure Active Period. Front. Earth Sci. 2022, 10, 828477. [Google Scholar] [CrossRef]
  27. Kuldeev, E.I.; Rysbekov, K.B.; Donenbayeva, N.S.; Miletenko, N.A. Modern Methods of Geotechnic—Effective Way of Providing Industrial Safety in Mines. Eurasian Min. 2021, 36, 18–21. [Google Scholar] [CrossRef]
  28. Zhang, P.; Chen, R.-P.; Wu, H.-N.; Liu, Y. Ground settlement induced by tunneling crossing interface of water-bearing mixed ground: A lesson from Changsha, China. Tunn. Undergr. Space Technol. 2020, 96, 103224. [Google Scholar] [CrossRef]
  29. Liu, J.; Si, Y.; Wei, D.; Shi, H.; Wang, R. Developments and prospects of microseismic monitoring technology in underground metal mines in China. J. Cent. South Univ. 2021, 28, 3074–3098. [Google Scholar] [CrossRef]
  30. Shu, L.; Liu, Z.; Wang, K.; Zhu, N.; Yang, J. Characteristics and Classification of Microseismic Signals in Heading Face of Coal Mine: Implication for Coal and Gas Outburst Warning. Rock Mech. Rock Eng. 2022, 55, 6905–6919. [Google Scholar] [CrossRef]
  31. Wang, K.; Ma, K.; Tang, C.; Liu, H.; Wang, X.; Li, Q. Study on Deep Mining-Induced Strata Behavior Based on the Evolutional Laws of Multiple Indices from Microseismic Monitoring. Rock Mech. Rock Eng. 2023, 56, 6481–6501. [Google Scholar] [CrossRef]
  32. Zhang, C.; Jin, G.; Liu, C.; Li, S.; Cheng, R.; Wang, X.; Zeng, X. Prediction of rockbursts in a typical island working face of a coal mine through microseismic monitoring technology. Tunn. Undergr. Space Technol. 2021, 113, 103972. [Google Scholar] [CrossRef]
  33. Zhang, Q.; Zhang, X.-P.; Liu, Q.; Pan, Y.; Chi, J.; Qiu, J.; Yin, X. Microseismic Monitoring and Rockburst Characteristics in a Deep-Buried Tunnel Excavated by TBM. Rock Mech. Rock Eng. 2024, 57, 1565–1578. [Google Scholar] [CrossRef]
  34. He, M.; Wang, Q.; Wu, Q. Innovation and future of mining rock mechanics. J. Rock Mech. Geotech. Eng. 2021, 13, 1–21. [Google Scholar] [CrossRef]
  35. Wang, Y.; Zheng, G.; Wang, X. Development and application of a goaf-safety monitoring system using multi-sensor information fusion. Tunn. Undergr. Space Technol. 2019, 94, 103112. [Google Scholar] [CrossRef]
  36. Qi, W.; Zhang, J.; Zhou, N.; Wu, Z.; Zhang, J. Mechanism by Which Backfill Body Reduces Amount of Energy Released in Deep Coal Mining. Shock Vib. 2019, 2019, 8253269. [Google Scholar] [CrossRef]
  37. Mao, H.; Xu, N.; Zhou, Z.; Sha, C.; Xiao, P.; Li, B. Failure mechanism and deformation forecasting of surrounding rock mass in an underground cavern based on engineering analogy method. Tunn. Undergr. Space Technol. 2024, 143, 105497. [Google Scholar] [CrossRef]
  38. Zhao, J.-S.; Chen, B.-R.; Jiang, Q.; Lu, J.-F.; Hao, X.-J.; Pei, S.-F.; Wang, F. Microseismic Monitoring of Rock Mass Fracture Response to Blasting Excavation of Large Underground Caverns Under High Geostress. Rock Mech. Rock Eng. 2022, 55, 733–750. [Google Scholar] [CrossRef]
  39. Zhao, Z.; Chen, B.; Wu, Z.; Zhang, S. Multi-sensing investigation of crack problems for concrete dams based on detection and monitoring data: A case study. Measurement 2021, 175, 109137. [Google Scholar] [CrossRef]
  40. Cui, X.J. Disaster Monitoring and Warning in Mining Process of Multi-Disasters Underground Mine. Master’s Thesis, Central South University, Changsha, China, 2012. [Google Scholar]
  41. Xie, X.; Li, S.; Guo, J. Study on Multiple Fractal Analysis and Response Characteristics of Acoustic Emission Signals from Goaf Rock Bodies. Sensors 2022, 22, 2746. [Google Scholar] [CrossRef]
  42. Zhou, Z.; Huang, Y.; Zhao, C. Distribution Law of Mine Ground Pressure via a Microseismic Sensor System. Minerals 2023, 13, 649. [Google Scholar] [CrossRef]
  43. Epstein, B.; Lomnitz, C. A Model for the Occurrence of Large Earthquakes. Nature 1966, 211, 954–956. [Google Scholar] [CrossRef]
  44. McGarr, A. Seismic moments and volume changes. J. Geophys. Res. 1976, 81, 1487–1494. [Google Scholar] [CrossRef]
  45. Kijko, A. Seismicity and seismic risk in the offshore North Sea area. Pure Appl. Geophys. 1985, 123, 348–349. [Google Scholar] [CrossRef]
  46. Kijko, A. A modified form of the Gutenberg-Richter magnitude-frequency relation: Maximum likelihood estimation of its parameters. Bull. Seismol. Soc. Am. 1985, 75, 319–322. [Google Scholar] [CrossRef]
Figure 1. Project overview. (a) Mine location. (b) Ore body occurrence. (c) Goaf. (d) Pillar.
Figure 1. Project overview. (a) Mine location. (b) Ore body occurrence. (c) Goaf. (d) Pillar.
Minerals 14 00948 g001
Figure 2. Sensor position relationship and monitoring system network topology diagram within the microseismic monitoring area.
Figure 2. Sensor position relationship and monitoring system network topology diagram within the microseismic monitoring area.
Minerals 14 00948 g002
Figure 3. Simplified diagram of residual ore body mining.
Figure 3. Simplified diagram of residual ore body mining.
Minerals 14 00948 g003
Figure 4. The eastern and western parts of the mine and the overall ore output. (a) 2013–2018. (b) In the months of 2018.
Figure 4. The eastern and western parts of the mine and the overall ore output. (a) 2013–2018. (b) In the months of 2018.
Minerals 14 00948 g004
Figure 5. Comparison chart of actual ore production and filling volume underground.
Figure 5. Comparison chart of actual ore production and filling volume underground.
Minerals 14 00948 g005
Figure 6. Spatial distribution of underground microseismic positioning events from 2013 to 2018.
Figure 6. Spatial distribution of underground microseismic positioning events from 2013 to 2018.
Minerals 14 00948 g006
Figure 7. The magnitude frequency cumulative relationship and its fitting formula for microseismic positioning events.
Figure 7. The magnitude frequency cumulative relationship and its fitting formula for microseismic positioning events.
Minerals 14 00948 g007
Figure 8. Probability distribution of cumulative occurrence of extreme magnitude values in 2019.
Figure 8. Probability distribution of cumulative occurrence of extreme magnitude values in 2019.
Minerals 14 00948 g008
Figure 9. Probability density distribution of extreme magnitude events in 2019.
Figure 9. Probability density distribution of extreme magnitude events in 2019.
Minerals 14 00948 g009
Figure 10. Probability index (P) distribution of microseismic energy release in 2019.
Figure 10. Probability index (P) distribution of microseismic energy release in 2019.
Minerals 14 00948 g010
Figure 11. Establishment of precursor patterns: (A) Time window for roof collapse occurrence; (B) Microseismic event rate curve; (C) Time window for roof collapse occurrence.
Figure 11. Establishment of precursor patterns: (A) Time window for roof collapse occurrence; (B) Microseismic event rate curve; (C) Time window for roof collapse occurrence.
Minerals 14 00948 g011
Figure 12. 37# probe area empty zone distribution map.
Figure 12. 37# probe area empty zone distribution map.
Minerals 14 00948 g012
Figure 13. Trend chart of microseismic event rate for probe 37#.
Figure 13. Trend chart of microseismic event rate for probe 37#.
Minerals 14 00948 g013
Figure 14. Photos of roof-caving site.
Figure 14. Photos of roof-caving site.
Minerals 14 00948 g014
Figure 15. Description of the destruction of the goaf where the 43# probe is located.
Figure 15. Description of the destruction of the goaf where the 43# probe is located.
Minerals 14 00948 g015
Figure 16. Microseismic event rate variation near probe 43#.
Figure 16. Microseismic event rate variation near probe 43#.
Minerals 14 00948 g016
Table 1. Statistical table of the exposed area of goaf covered by filling bodies (10,000 m2).
Table 1. Statistical table of the exposed area of goaf covered by filling bodies (10,000 m2).
YearEastern Residual
Mining Area
Western Regular
Mining Area
Total Exposed
Area Covered
Percentage of
Coverage Area
20131.1001.103.25%
20141.100.311.414.17%
20152.120.362.487.33%
20162.450.502.958.72%
20173.210.283.4910.32%
20182.921.154.0712.03%
Total12.902.6015.5045.82%
Table 2. Statistical data of reduced volume of goaf from 2013 to 2018 (10,000 m3).
Table 2. Statistical data of reduced volume of goaf from 2013 to 2018 (10,000 m3).
YearMining Volume (10,000 tons)Increase in Space
∆Ve
Filling Volume
∆Vf
Empty Area Variation
∆Vr
Reduction of the Percentage of Space Volume
201361.7119.3411.967.34*
201443.1614.2720.00−6.232.07%
201546.6115.0324.70−8.772.92%
201647.7315.0634.90−19.886.63%
201745.3914.8430.15−15.315.10%
201845.5114.2634.70−20.446.81%
Total290.1192.8156.36−63.2921.10%
Note: * represents the total volume of existing goaf at the end of 2012, which is 3 million m3.
Table 3. Energy and magnitude statistics of underground microseismic positioning events from 2013 to 2018.
Table 3. Energy and magnitude statistics of underground microseismic positioning events from 2013 to 2018.
No.DateEnergy/JLocal
Magnitude
No.DateEnergy/JLocal
Magnitude
11 February 2013562.00−1.726211 July 201384.60−2.02
22 March 20134.36−2.816311 July 201369.80−2.22
312 March 2013995.00−1.256414 July 20139.86−2.39
430 March 2013323.00−1.716529 July 201313.30−2.38
531 March 201317.90−2.64662 August 201324.80−2.07
631 March 20130.29−3.35674 August 20130.35−3.58
731 March 20130.16−3.716818 August 20131.72−3.17
831 March 201314.70−2.62691 December 201324.20−2.42
931 March 201323.80−2.437021 December 2013280.00−1.59
1031 March 20130.19−3.73719 May 20140.74−3.16
116 April 20130.04−3.90726 June 201422.00−3.02
126 April 20130.05−3.947312 August 20141.64−3.08
136 April 20130.07−4.547419 August 20147.39−2.59
148 April 20130.24−3.717528 April 20163.35−2.88
158 April 20130.28−3.60764 July 20160.63−3.54
163 May 2013153.00−1.78774 July 20160.04−4.16
1721 June 201318.90−2.527821 July 20160.08−3.95
1828 June 20134.12−2.527921 July 20160.10−4.07
1928 June 2013869.00−0.78802 August 2016151.00−1.84
2028 June 2013210.00−1.738131 October 20162.87−2.01
2128 June 201373.60−1.558211 February 201721.00−2.89
2229 June 20132.48−3.40834 June 2017102.00−1.98
232 July 20132.22−3.02844 June 201714.50−2.36
242 July 201316.20−2.328513 June 20174.82−3.14
253 July 20135.62−2.86861 July 20177.19−2.44
263 July 20135.68−2.72873 September 201785.20−2.66
273 July 2013229.00−1.83884 November 20174.51−2.72
283 July 201328.20−2.38895 June 20182.99−2.83
293 July 201327.40−2.309014 June 20181.38−3.15
303 July 20130.16−3.9391141 June 20180.25−3.81
313 July 20130.24−3.819222 June 20180.01−3.59
323 July 20130.46−3.379320 August 2018119.00−2.20
333 July 2013112.00−2.53943 October 20186.20−2.59
343 July 2013134.00−1.94953 October 2018125.00−1.58
353 July 20137.05−2.68968 October 20189.19−2.58
363 July 201313.70−2.809712 October 20180.06−4.12
374 July 20130.60−3.109816 October 20181.42−3.33
384 July 20136.08−3.069922 October 20180.09−3.88
394 July 201331.50−2.3610027 October 20184.42−2.79
404 July 2013240.00−2.1610127 October 20184.27−2.98
414 July 20131.50−3.0710227 October 20181.63−3.06
424 July 201311.90−2.7810327 October 20181.36−3.30
436 July 201353.40−1.7410427 October 20186.43−2.81
446 July 20133.09−2.9010527 October 20180.21−3.87
456 July 201339.40−1.9010627 October 20180.55−3.79
466 July 2013115.00−1.5810727 October 20181.95−3.21
477 July 201336.60−2.4210827 October 201812.30−2.51
488 July 20133.71−2.3610927 October 20180.61−3.69
498 July 20133.27−3.1911027 October 20181.01−3.44
509 July 20130.09−3.9011127 October 20180.12−4.24
519 July 20130.51−3.5911227 October 20186.06−2.56
529 July 201367.90−2.3611327 October 20180.28−3.87
539 July 201381.80−2.4111427 October 20183.46−2.81
549 July 2013315.00−1.6711527 October 20180.16−3.95
5510 July 20135.65−2.4311627 October 20184290.00−0.37
5610 July 20133.29−2.931179 November 20180.94−3.32
5710 July 20133.84−2.8411811 November 20182.79−2.92
5811 July 20130.28−3.8511911 November 20183.54−2.87
5911 July 20130.14−3.5412014 November 201824.10−2.60
6011 July 20130.24−3.7012118 November 20180.15−3.83
6111 July 20130.30−3.9812213 December 2018460.00−1.27
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zhou, Z.; Huang, Y.; Zhao, C. Microseismic Monitoring and Disaster Warning via Mining and Filling Processes of Residual Hazardous Ore Bodies. Minerals 2024, 14, 948. https://doi.org/10.3390/min14090948

AMA Style

Zhou Z, Huang Y, Zhao C. Microseismic Monitoring and Disaster Warning via Mining and Filling Processes of Residual Hazardous Ore Bodies. Minerals. 2024; 14(9):948. https://doi.org/10.3390/min14090948

Chicago/Turabian Style

Zhou, Zilong, Yinghua Huang, and Congcong Zhao. 2024. "Microseismic Monitoring and Disaster Warning via Mining and Filling Processes of Residual Hazardous Ore Bodies" Minerals 14, no. 9: 948. https://doi.org/10.3390/min14090948

APA Style

Zhou, Z., Huang, Y., & Zhao, C. (2024). Microseismic Monitoring and Disaster Warning via Mining and Filling Processes of Residual Hazardous Ore Bodies. Minerals, 14(9), 948. https://doi.org/10.3390/min14090948

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop