Dynamic early warning method for rock stratum instability based on microseismic spatio-temporal information entropy value

Abstract

The present application provides a dynamic early warning method for rock stratum instability based on the microseismic spatio-temporal information entropy value. In this method, based on the K-means clustering method, the microseismic monitoring data of microseismic events in an underground mining area are clustered according to spatial coordinates, and based on the clustering results, the information entropy value of each category of microseismic monitoring data is calculated; according to the information entropy value of the microseismic monitoring data, a graph of the change of the information entropy of the underground mining area over time is drawn to conduct dynamic early warning of rock stratum instability in the underground mining area. Thus, by using a discrimination method with a single index, the failure state of the rock can be accurately predicted, which can not only accurately early warn of rock instability disasters, but also has the advantages of simple single-index discrimination method, fast discrimination speed and good effect.

Description

Technical Field

[0001] The present application relates to the technical field of mining, and particularly relates to a dynamic early warning method for strata instability based on microseismic spatio-temporal information entropy value. Background Art

[0002] In the area of underground mines, the mining of coal seams often causes changes in the ground pressure state of the surrounding rock, resulting in stress concentration in local areas, and thus rock mass instability phenomena such as joint structure slip and roof rock rupture occur. The continuous accumulation of ground pressure activities will cause the development and penetration of microfractures in the surrounding rock, leading to large-scale collapse of the roof, forming shock air waves that pose a threat to underground personnel and equipment, and also causing waste of stope resources and seriously affecting the implementation of production plans.

[0003] Microseismic monitoring is based on acoustic emission and seismology. By monitoring the vibrations generated by rock mass fractures or the vibrations of other objects, using the seismic waves emitted by detected microfractures to determine the location of earthquakes, giving the intensity and frequency of seismic activity, judging the laws of potential rock bursts, and making evaluations on the damage status, safety status, etc. of the monitored object, so as to provide a basis for predicting and controlling disasters. It is a real-time and continuous monitoring technology.

[0004] Currently, in the technology of monitoring and early warning of underground mine mining through microseismic monitoring, multiple calculation indicators are used to monitor and early warn disasters. However, due to the large number of calculation indicators and the mutual influence between multiple calculation indicators, it is easy to miss the influence between some indicators during the early warning process, which is bound to cause missed or false alarms of risks.

[0005] Therefore, there is an urgent need to provide a technical solution to address the above deficiencies in the prior art. Summary of the Invention

[0006] The purpose of the present application is to provide a dynamic early warning method for strata instability based on microseismic spatio-temporal information entropy value to solve or alleviate the problems existing in the above prior art.

[0007] To achieve the above purpose, the present application provides the following technical solutions:

[0008] The present application provides a dynamic early warning method for strata instability based on microseismic spatio-temporal information entropy value for strata instability early warning in underground mining areas, including: Step S101, clustering the microseismic monitoring data of microseismic events in the underground mining area according to spatial coordinates based on the K-means clustering method; Step S102, calculating the information entropy value of each class of the microseismic monitoring data based on the clustering result; Step S103, drawing a graph of the change of the information entropy of the underground mining area over time according to the information entropy value of the microseismic monitoring data, so as to conduct dynamic early warning of strata instability in the underground mining area according to the graph of the change of the information entropy over time.

[0009] Preferably, step S101 includes: calculating the clustering K value of the microseismic monitoring data of the microseismic event based on the silhouette coefficient method; where K is a positive integer; clustering the microseismic monitoring data into K types of microseismic data according to the spatial coordinates based on the K-means clustering method.

[0010] Preferably, step S102 includes: extracting the time and magnitude information of each type of microseismic data among the K types of microseismic data obtained by the K-means clustering method, and sorting the magnitude information corresponding to each type of microseismic data according to the time series; dividing the sorted magnitude information corresponding to each type of microseismic data into slices based on a preset slice length; calculating the slice information entropy value of each group of slices corresponding to each type of microseismic data.

[0011] Preferably, the preset slice length is determined based on the dichotomy method.

[0012] Preferably, the calculation of the slice information entropy value of each group of slices corresponding to each type of microseismic data is specifically: according to the formula:

[0013]

[0014] calculate the slice information entropy value H of each group of slices;

[0015] where n is the number of slice groups obtained by dividing the sorted magnitude information corresponding to k types of microseismic data into slices, k j ∈K, j is a positive integer; A j is the bth group of slice groups among the obtained n groups of slice groups, b∈n, b is a positive integer; a b is the ith magnitude information in the bth group of slice groups A i is the ith magnitude information in the bth group of slice groups A b where i∈m, m is the number of magnitude information in the bth group of slice groups A b ; P(a i ) is the frequency of occurrence of the magnitude information a b in the bth group of slice groups A i .

[0016] Preferably, in step S103: the third low point of the information entropy value that continuously decreases below the average line in the information entropy vs. time graph is used as the instability warning point.

[0017] Preferably, in step S103: in response to the information entropy value in the information entropy vs. time graph being between [2.28, 2.60], dynamic warning of rock stratum instability in the underground mining area is performed.

[0018] Beneficial effects:

[0019] The dynamic early warning method for strata instability based on the microseismic spatio-temporal information entropy value provided by the embodiments of the present application first clusters the microseismic monitoring data of microseismic events in the underground mining area according to the spatial coordinates based on the K-means clustering method, and calculates the information entropy value of each category of microseismic monitoring data based on the clustering results; then, according to the information entropy value of the microseismic monitoring data, a graph of the change of the information entropy of the underground mining area over time is drawn to perform dynamic early warning of strata instability in the underground mining area. Thus, by using a discriminant method with a single index, the failure state of the rock can be accurately predicted, not only can the rock instability disaster be accurately warned, but also the single-index discriminant method is simple, the discrimination speed is fast, and the effect is good.

[0020] The degree of rock stratum chaos in the underground mining area is expressed by the information entropy value, the stability degree of the rock strata in the underground mining area is evaluated, and the rock fracture situation in the strata instability is quantified to achieve the effect of accurately warning the rock instability and improve the early warning accuracy of the strata instability in the underground mining area. Moreover, using the information entropy value as the judgment index for strata instability is simpler and easier to use than the multi-index judgment, saving time and effort, without the need to analyze and process the complex relationships between multiple indexes, being easier to understand and explain, and being conducive to decision-makers making decisions more easily, effectively avoiding the influence brought by multi-index chaos or omission.

[0021] At the same time, by using the method of cluster analysis, the microseismic data is grouped according to the spatial coordinates, the influence of discrete events is removed, and the failure response of the rock instability is accurately reflected in a more concentrated failure area, so as to discover the spatial distribution law and spatial characteristics of microseismic events, and more accurately capture the local failure information in the microseismic signal of microseismic events, further improving the prediction range and accuracy of strata instability in the underground mining area. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The specification drawings forming a part of the present application are used to provide a further understanding of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation to the present application. Among them:

[0023] Figure 1 is a schematic flowchart of a method for dynamic early warning of strata instability based on the microseismic spatio-temporal information entropy value according to some embodiments of the present application;

[0024] Figure 2 is a logic diagram of a method for dynamic early warning of strata instability based on the microseismic spatio-temporal information entropy value according to some embodiments of the present application;

[0025] Figure 3 is a clustering schematic diagram of microseismic monitoring data according to some embodiments of the present application;

[0026] Figure 4Schematic diagram of microseismic time series slices provided according to some embodiments of the present application;

[0027] Figure 5 Graph of the change in information entropy value of the first main roadway provided according to some embodiments of the present application;

[0028] Figure 6 Graph of the change in information entropy value of the second main roadway provided according to some embodiments of the present application. Detailed implementation manners

[0029] The present application will be described in detail below with reference to the drawings and in conjunction with embodiments. Each example is provided by way of explanation of the present application rather than limitation of the present application. In fact, those skilled in the art will clearly understand that modifications and variations can be made to the present application without departing from the scope or spirit of the present application. For example, features shown or described as part of one embodiment can be used in another embodiment to yield yet another embodiment. Therefore, it is desirable that the present application include such modifications and variations that fall within the scope of the appended claims and their equivalents.

[0030] Due to the complexity of the rock mass instability mechanism, at present, it is difficult to find an exact precursor characteristic parameter as a quantitative index for disaster early warning. During the process of disaster monitoring and early warning through microseismic monitoring, since there are multiple calculation indexes, the influencing factors between multiple calculation indexes are easily omitted or ignored, or the indexes selected from multiple calculation indexes are incorrect or incomplete, resulting in the phenomenon of false alarms or missed alarms of disasters occurring from time to time.

[0031] Based on this, the applicant proposes a dynamic early warning method for rock stratum instability based on microseismic spatio-temporal information entropy value, which uses a single index to monitor and early warn of rock instability, effectively avoiding the phenomena of false alarms and missed alarms that occur when using multiple calculation indexes. As Figure 1 、 Figure 2 shown, the dynamic early warning method for rock stratum instability based on microseismic spatio-temporal information entropy value includes:

[0032] Step S101: Cluster the microseismic monitoring data of microseismic events in the underground mining area according to spatial coordinates based on the K-means clustering method.

[0033] In the present application, through the microseismic monitoring system in the underground mining area, the microseismic monitoring data of microseismic events in the underground mining area are collected in real time, and the collected microseismic events are located to obtain data such as the coordinates, time, and magnitude of the occurrence of microseismic events, and a microseismic monitoring database about the underground mining area is established. According to the spatial coordinates of the occurrence of microseismic events, the three-dimensional display of the spatial positions of microseismic events in the underground mining area is performed to obtain the contour shape of the occurrence of microseismic events in the underground mining area, so as to clearly observe the positions where microseismic events occur in the underground mining area.

[0034] When clustering the microseismic monitoring data in the microseismic monitoring database according to spatial coordinates based on the K-means clustering method, the microseismic monitoring data in the microseismic monitoring database are divided into K groups (K is a positive integer), and a group of microseismic monitoring data is randomly selected from each of the K groups as the initial clustering center. The distances between the other data in the microseismic monitoring database and the K data selected as the initial clustering centers are calculated respectively. Then, according to the calculated distance magnitudes, the other data in the microseismic monitoring database are assigned to the initial clustering center closest to it.

[0035] Each initial clustering center and the data objects assigned to them form a clustering category. The clustering center of each clustering category is recalculated based on the existing data objects in the clustering until there are no data objects in the microseismic monitoring database that can be reassigned to different clustering categories, the clustering centers of each clustering category no longer change, and the sum of squared errors is locally minimized, completing the clustering of the microseismic monitoring data in the microseismic monitoring database. That is, based on the K-means clustering method, according to the spatial coordinates, the microseismic monitoring data are clustered into K categories of microseismic data, as Figure 3 shown.

[0036] In this application, based on the silhouette coefficient method, the clustering K value of the microseismic monitoring data of microseismic events is calculated. In a specific example, first, the evaluation indexes of the clustering algorithm effect are imported from the python database to evaluate the compactness and separation degree of the clustering results. Among them, the value range of the evaluation indexes of the clustering algorithm effect is [-1, 1], and the closer the evaluation indexes of the clustering algorithm effect are to 1, the better the clustering result. Then, the spatial coordinates of the microseismic data are imported, and the K value range is initialized (the K value range can be adjusted according to the silhouette coefficient calculated later); then, for each K value, a clustering is performed once, and its corresponding silhouette coefficient is calculated, and the K value with the highest silhouette coefficient is found, which is the optimal number of clustering categories.

[0037] Step S102: Calculate the information entropy value of each category of microseismic monitoring data based on the clustering result.

[0038] The microseismic monitoring data in the microseismic monitoring database are classified according to the clustering result, and the time and magnitude in each category of microseismic monitoring data are obtained, and each category of microseismic monitoring data is sorted according to the preset rules. Specifically, the time and magnitude information of each category of microseismic data in the K categories of microseismic data obtained by the K-means clustering method are extracted, and the magnitude information corresponding to each category of microseismic data is sorted according to the time series.

[0039] Then, based on a preset slice length, slice and divide the sorted magnitude information corresponding to each type of microseismic data, and calculate the information entropy value of each group of slices corresponding to each type of microseismic data. Specifically, select an appropriate number of slices according to the data volume of each type of microseismic data, and slice the microseismic events corresponding to each type of microseismic data according to time periods. As Figure 4 shown, then calculate the distribution frequency of the magnitude values in the slices, and substitute them into the information entropy calculation formula to obtain the change in information entropy related to the magnitude.

[0040] In this application, based on the dichotomy method, determine the preset slice length of each type of microseismic data. Specifically, arrange the microseismic events in chronological order and divide them into two slices, each with a length of n / 2, where n is a positive integer representing the number of microseismic events, and calculate the information entropy value of each slice; then divide each slice by half of the previous slice length. Compare the information entropy value of each slice with the large-magnitude events to determine whether it meets a reasonable information entropy value judgment criterion (that is, whether the information entropy value also increases when large-magnitude events occur); if it meets a judgment criterion of the information entropy value, take the corresponding slice length as the preset slice length.

[0041] When calculating the slice information entropy value of each group of slices corresponding to each type of microseismic data, according to the formula:

[0042]

[0043] Calculate the slice information entropy value H of each group of slices. Among them, n is the number of slice groups obtained by slicing and dividing the sorted magnitude information corresponding to k j types of microseismic data, h j ∈K, j is a positive integer; A b is the bth group of slice groups among the obtained n groups of slice groups, b ∈ n, b is a positive integer; a i is the ith magnitude information in the bth group of slice groups A b , i ∈ m, m is the number of magnitude information in the bth group of slice groups A b ; P(a i ) is the frequency of the magnitude information a b appearing in the bth group of slice groups A i .

[0044] Specifically, according to the ratio of the number of times the magnitude information a i appears in the bth group of slice groups A b to the number m of magnitude information in the bth group of slice groups A b , obtain the frequency P(a i ) of the magnitude information a b appearing in the bth group of slice groups A i)。For example, a slice group has a total of 20 data sources (i.e., 20 magnitude information), and the value is a i The magnitude information with the value of a i appears 5 times in the b-th slice group A b The frequency P(a i ) = 0.25

[0045] Step S103: According to the information entropy value of the microseismic monitoring data, draw a graph of the change of the information entropy of the underground mining area over time, so as to conduct dynamic early warning of rock strata instability in the underground mining area based on the graph of the change of the information entropy over time

[0046] In this application, taking the data of the slice information entropy value of each group of slices and the end time corresponding to the slice as the object, draw a graph of the change of the information entropy of the underground mining area over time, and conduct dynamic early warning of rock strata instability in the underground mining area through the graph of the change of the information entropy over time. Specifically, the degree of rock instability state is characterized by the information entropy value. In the stable state of the rock, the information entropy value is in a low level state, the internal structure of the rock is relatively orderly, and the information entropy value fluctuates slightly up and down

[0047] When the rock begins to generate cracks and expand, the information entropy value will rapidly rise from the previous low level state to a high level state, reflecting the increase in the internal chaos degree of the rock. With the expansion of the cracks, the information entropy value will further increase, reflecting the intensification of the chaotic degree inside the rock

[0048] When the rock breaks, the information entropy value will suddenly increase due to the increase in the scale of the fracture. This process is usually accompanied by the occurrence of large magnitudes. The information entropy value will enter a high level state and will fluctuate. This kind of fluctuation may be caused by the re-adjustment of the rock structure

[0049] When the rock is unstable and collapses, the information entropy value will suddenly reach the maximum value, the internal structure of the rock is completely destroyed, and the chaotic degree reaches the peak. Therefore, through the evolution law of the information entropy value during the rock instability process, the precursor characteristics of the rock instability and collapse can be revealed and used as the basis for the rock instability and collapse, so as to better help understand the change process of the rock strata in the underground mining area and improve the early warning ability of the rock instability and collapse

[0050] The instability and failure of the rock generally cause macroscopic damage, such as Figure 5As shown, in the first main roadway, the information entropy value ranges from 2.28 to 3.44. Among them, the main range of the information entropy value is between 2.60 and 3.20. Within this range, the magnitude fluctuation range is relatively stable, and the destructive effect on the rock mass is small, belonging to the safe range area. During this period, the roadway can be supported according to the normal situation. When the information entropy value is between 2.28 and 2.60, since there are many microseismic events with the same or similar magnitudes, this may be due to the period when most small cracks expand into large cracks, and the magnitudes are the same or similar. Observe whether the magnitude value will cause damage to the rock mass. Because if the large crack expands again, it will lead to through - failure, so early warning of strata instability can be carried out in this area, which belongs to the dangerous range area. During this period, it is necessary to strengthen the support in advance or take corresponding pressure - relief measures, such as borehole pressure - relief, pressure - relief blasting or hydraulic fracturing and other methods to reduce the danger. When the information entropy value is greater than 3.20, large - magnitude events often occur, and at this time the rock strata are approaching failure, belonging to the extremely dangerous range area. When the early warning is issued, it is necessary to evacuate the dangerous area in advance.

[0051] In this application, the calculated information entropy value of microseismic data shows the following law. Before the generation of large cracks, it is formed by the aggregation of a series of small cracks. When the small cracks aggregate with each other, they will produce the same or similar magnitude responses, so the information entropy value will decrease. By observing the magnitude characteristics when the information entropy value is low, to judge whether large - magnitude events will occur later and whether the rock will be damaged. Specifically, the third low point of the information entropy value that continuously drops below the average line is used as the instability early - warning point, where the average line is the line where the average value of the calculated information entropy value is located.

[0052] As Figure 5 、 Figure 6 shown, in the two clustering regions of the first main roadway ( Figure 5 ) and the second main roadway ( Figure 6 ), the information entropy values calculated according to the magnitude data also have larger magnitudes at the highest point and relatively high - point positions. On the 14th day in the first main roadway, the information entropy value reached the highest value of 3.44, and at the same time, large magnitudes appeared.

[0053] The information entropy values below the average line of the information entropy passed through 2.66 on the 1st day, 2.62 on the 5th day, and 2.58 on the 8th day, three continuously decreasing information entropy values, and finally rose to 3.44. (Among them, the three continuously decreasing information entropy values are shown as 1 - 1, 1 - 2, 1 - 3 in the figure), resulting in large - magnitude events. Similarly, on the 116th day, a high information entropy value of 3.34 appeared. Before that, on the 76th day, 2.60, on the 95th day, 2.51, and on the 109th day, 2.48. (Among them, the three continuously decreasing information entropy values are shown as 2 - 1, 2 - 2, 2 - 3 in the figure), the information entropy also passed through three consecutive decreases below the average line and then rose until large - magnitude events occurred.

[0054] The image of the information entropy of the second main roadway changing with time also has such a pattern. Therefore, when there are 3 consecutive low information entropy values below the average value line, a warning is issued when the 3rd consecutive low information entropy value appears.

[0055] In this application, the degree of rock stratum chaos in the underground mining area is expressed by the information entropy value, the stability degree of the rock stratum in the underground mining area is evaluated, and the rock fracture situation in the rock stratum instability is quantified, so as to achieve the effect of accurately warning the rock instability and improve the warning accuracy of the rock stratum instability in the underground mining area. Moreover, using the information entropy value as the judgment index for rock stratum instability is simpler and easier to use than multi-index judgment, saving time and effort. There is no need to analyze and process the complex relationships between multi-indices, and it is easier to understand and explain, which is conducive to decision-makers making decisions more easily and effectively avoiding the influence brought by multi-index confusion or omission.

[0056] At the same time, using the method of cluster analysis, the microseismic data are grouped according to the spatial coordinates, the influence of discrete events is removed, and the damage response of rock instability is accurately reflected in a more concentrated damage area, so as to discover the spatial distribution law and spatial characteristics of microseismic events, and capture the local damage information in the microseismic signal of microseismic events more accurately, further improving the prediction range and accuracy of rock stratum instability in the underground mining area.

[0057] The above are only the preferred embodiments of the present application and are not used to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A dynamic early warning method for rock stratum instability based on microseismic spatio-temporal information entropy value, which is used for early warning of rock stratum instability in underground mining areas, and is characterized in that, Including: Step S101: Based on the K-means clustering method, cluster the microseismic monitoring data of the microseismic events in the underground mining area according to spatial coordinates; Step S102: Based on the clustering results, calculate the information entropy value of each type of the microseismic monitoring data; specifically, extract the time and magnitude information of each type of microseismic data among the K types of microseismic data obtained by the K-means clustering method, and sort the magnitude information corresponding to each type of microseismic data according to the time series; based on the preset slice length, perform slice division on the sorted magnitude information corresponding to each type of microseismic data; calculate the slice information entropy value of each group of slices corresponding to each type of microseismic data; Step S103: According to the information entropy value of the microseismic monitoring data, draw a graph of the change of the information entropy of the underground mining area over time, so as to perform dynamic warning on the instability of the rock stratum in the underground mining area according to the graph of the change of the information entropy over time; specifically, take the third low point of the information entropy value that continuously drops below the average line in the graph of the change of the information entropy over time as the instability warning point.

2. The dynamic early warning method for strata instability based on the microseismic spatio-temporal information entropy value according to claim 1, characterized in that Step S101 includes: Based on the silhouette coefficient method, calculate the clustering K value of the microseismic monitoring data of the microseismic events; where K is a positive integer; Based on the K-means clustering method, cluster the microseismic monitoring data into K types of microseismic data according to spatial coordinates.

3. The dynamic early warning method for strata instability based on the microseismic spatio-temporal information entropy value according to claim 1, characterized in that, Based on the dichotomy method, determine the preset slice length.

4. The dynamic early warning method for strata instability based on the microseismic spatio-temporal information entropy value according to claim 1, wherein The calculation of the slice information entropy value of each group of slices corresponding to each type of microseismic data is specifically as follows: According to the formula: Calculate the slice information entropy value H of each group of slices; Among them, n is the number of sliced groups obtained by slicing and dividing the sorted magnitude information corresponding to k j types of microseismic data, k j ∈K, j is a positive integer; A b is the b-th sliced group among the obtained n groups of sliced groups, b ∈ n, and b is a positive integer; a i is the i-th magnitude information in the b-th sliced group A b , i ∈ m, m is the number of magnitude information in the b-th sliced group A b ; P(a i ) is the frequency of occurrence of the magnitude information a b in the b-th sliced group A i .

5. The dynamic early warning method for strata instability based on the microseismic spatio-temporal information entropy value according to claim 1, characterized in that, In step S103, in response to the information entropy value in the graph of the change of the information entropy over time being between [2.28, 2.60], perform dynamic warning on the instability of the rock stratum in the underground mining area.

Images