Risk determination method for overlying strata migration and risk monitoring system for overlying strata migration

Abstract

The invention provides an overlying strata migration risk determination method and an overlying strata migration risk monitoring system. The method comprises the steps that related parameters of a virtual coal mine are obtained, risk assessment is carried out on all the related parameters, multiple assessment results are obtained, data fusion is carried out according to all the assessment results, a comprehensive analysis result is obtained, the fusion mode at least comprises one or more of weighted average, Bayesian fusion, information entropy fusion and fuzzy processing fusion, and the comprehensive analysis result is obtained. The comprehensive analysis result is used for determining whether the virtual coal mine has an overlying strata migration risk; and generating early warning information under the condition that the comprehensive analysis result represents that the overlying strata migration risk occurs in the virtual coal mine. According to the method, the problem that in the prior art, only single monitoring equipment is adopted for researching the overlying strata risk, and dynamic disaster risk early warning cannot be accurately carried out is solved.

Description

Technical Field

[0001] The present application relates to the technical field of coal mine migration monitoring. Specifically, it relates to a method for determining the risk of overlying rock migration, a device, and a risk monitoring system for overlying rock migration. Background Art

[0002] The physical similarity simulation technology for coal mining is an important means to study the physical phenomena and mechanical behaviors in coal mining. It conducts simulation research by constructing a physical model similar to the actual mining conditions.

[0003] However, in the current solutions, only a single monitoring device is used to study the risk of overlying rock, and it is impossible to accurately give an early warning of dynamic disaster risks. Summary of the Invention

[0004] The main purpose of the present application is to provide a method for determining the risk of overlying rock migration, a device, and a risk monitoring system for overlying rock migration, so as to at least solve the problem that in the prior art, only a single monitoring device is used to study the risk of overlying rock, and it is impossible to accurately give an early warning of dynamic disaster risks.

[0005] To achieve the above object, according to one aspect of the present application, a method for determining the risk of overlying rock migration is provided, including: obtaining relevant parameters of a virtual coal mine, where the virtual coal mine is a virtual model that simulates the actual mining conditions and rock stratum mechanical behaviors of a real coal mine, and the relevant parameters include one or more of a subsidence value, a displacement amount, a borehole wall image, a pressure value, a microseismic signal, an infrared image, an acoustic emission signal, the subsidence value is the vertical movement change amount of a rock stratum collected by a dial gauge, the displacement amount is the horizontal movement change amount of a rock stratum collected by a total station, the borehole wall image is an image of a rock stratum collected by a borehole television, the pressure value is the stress value of a rock stratum collected by a pressure sensor, the microseismic signal is a signal of a microseismic event of a rock stratum collected by a microseismic monitor, the infrared image is an image of the temperature distribution of a rock stratum collected by an infrared thermal imager, and the acoustic emission signal is an energy signal of elastic waves released when a rock stratum generates cracks collected by an acoustic emission sensor; performing risk assessment on all the relevant parameters to obtain a plurality of assessment results, where the assessment results and the relevant parameters correspond one by one, and the assessment results characterize whether the virtual coal mine has a risk of overlying rock migration according to the relevant parameters; performing data fusion on all the assessment results to obtain a comprehensive analysis result, where the fusion method includes at least one or more of weighted average, Bayesian fusion, information entropy fusion, and fuzzy processing fusion, and the comprehensive analysis result is used to determine whether the virtual coal mine has a risk of overlying rock migration; and generating a warning message when the comprehensive analysis result characterizes that the virtual coal mine has a risk of overlying rock migration.

[0006] According to another aspect of the present application, a risk determination device for overburden movement is provided, including: an acquisition unit configured to acquire relevant parameters of a virtual coal mine, where the virtual coal mine is a virtual model simulating the actual mining conditions and strata mechanical behavior of a real coal mine, and the relevant parameters include one or more of a subsidence value, a displacement amount, a borehole wall image, a pressure value, a microseismic signal, an infrared image, and an acoustic emission signal. The subsidence value is the vertical movement change amount of the strata collected by a dial gauge, the displacement amount is the horizontal movement change amount of the strata collected by a total station, the borehole wall image is an image of the strata collected by a borehole television, the pressure value is the stress value of the strata collected by a pressure sensor, the microseismic signal is the signal of a microseismic event of the strata collected by a microseismic monitor, the infrared image is an image of the temperature distribution of the strata collected by an infrared thermal imager, and the acoustic emission signal is the energy signal of elastic waves released when cracks occur in the strata collected by an acoustic emission sensor; an evaluation unit configured to perform risk assessment on all the relevant parameters to obtain a plurality of evaluation results, where the evaluation results correspond one-to-one with the relevant parameters, and the evaluation results represent whether the risk of overburden movement occurs in the virtual coal mine is preliminarily determined according to the relevant parameters; a comprehensive analysis unit configured to perform data fusion based on all the evaluation results to obtain a comprehensive analysis result, where the fusion method includes at least one or more of weighted average, Bayesian fusion, information entropy fusion, and fuzzy processing fusion, and the comprehensive analysis result is used to determine whether the risk of overburden movement occurs in the virtual coal mine; a generation unit configured to generate a warning message when the comprehensive analysis result indicates that the risk of overburden movement occurs in the virtual coal mine.

[0007] According to still another aspect of the present application, a risk monitoring system for overburden movement is provided, including: one or more processors, a memory, and one or more programs, where the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include those for executing any one of the risk determination methods for overburden movement.

[0008] Applying the technical solution of the present application, data is acquired through various monitoring means such as a dial gauge, a total station, a borehole television, a microseismic monitor, a thermal infrared imager, a pressure sensor, and an acoustic emission monitor. Compared with the application of traditional single or a few monitoring devices, it can comprehensively acquire overburden movement information from different physical quantities, different spatial positions (vertical and horizontal directions), and different strata depths, avoiding monitoring blind spots, greatly enriching the data dimension, and providing an early warning for coal mine safety production more timely and accurately than a single monitoring means. Description of the Drawings

[0009] The accompanying drawings of the specification, which form a part of this application, are used to provide a further understanding of this application. The schematic embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation to this application. In the drawings:

[0010] Figure 1 shows a hardware structure block diagram of a mobile terminal for a method of determining the risk of overburden movement provided in an embodiment of this application;

[0011] Figure 2 shows a schematic flowchart of a method of determining the risk of overburden movement provided in an embodiment of this application;

[0012] Figure 3 shows a schematic diagram of a virtual coal mine;

[0013] Figure 4 shows a schematic diagram of multiple devices collecting parameters;

[0014] Figure 5 shows a schematic diagram of dial gauge monitoring;

[0015] Figure 6 shows a schematic diagram of total station observation layout;

[0016] Figure 7 shows a schematic diagram of the principle of forward coordinate calculation;

[0017] Figure 8 shows a schematic diagram of total station monitoring;

[0018] Figure 9 shows a schematic diagram of borehole television layout;

[0019] Figure 10 shows a schematic diagram of pressure sensor layout;

[0020] Figure 11 shows a schematic diagram of microseismic monitor layout;

[0021] Figure 12 shows a schematic flowchart of microseismic detection;

[0022] Figure 13 shows a schematic diagram of the working principle of an infrared thermal imager;

[0023] Figure 14 shows a schematic diagram of acoustic emission waveform characteristics;

[0024] Figure 15 shows a schematic diagram of the acoustic emission monitoring position along the strike;

[0025] Figure 16 shows a structure block diagram of a device for determining the risk of overburden movement provided in an embodiment of this application.

[0026] Among them, the above-mentioned drawings include the following reference numerals:

[0027] 102, processor; 104, memory; 106, transmission device; 108, input / output device. Detailed implementation manners

[0028] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other. The following will describe the present application in detail with reference to the drawings and in combination with the embodiments.

[0029] In order to enable those skilled in the art to better understand the solution of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present application.

[0030] It should be noted that the terms "first", "second", etc. in the specification, claims and drawings of the present application are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such used data may be interchanged under appropriate circumstances so as to describe the embodiments of the present application herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily need to be limited to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these process, method, product or device.

[0031] As introduced in the background art, in the prior art, only a single monitoring device is used to study the overlying rock risk, and the dynamic disaster risk warning cannot be accurately carried out. To solve the above problems, the embodiments of the present application provide a method for determining the risk of overlying rock migration, a device and a risk monitoring system for overlying rock migration.

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention.

[0033] The method embodiments provided in the embodiments of the present application can be executed on a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 is a hardware structure block diagram of a mobile terminal for a method of determining the risk of overlying rock migration according to an embodiment of the present invention. As Figure 1 shown, the mobile terminal may include one or more (Figure 1 Only one processor 102 is shown (the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a field programmable gate array FPGA), and a memory 104 for storing data. Among them, the above mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those of ordinary skill in the art can understand that Figure 1 The structure shown is only schematic and does not limit the structure of the above mobile terminal. For example, the mobile terminal may further include more or fewer components than Figure 1 shown, or have a different configuration from Figure 1 shown.

[0034] The memory 104 can be used to store computer programs. For example, software programs and modules of application software, such as the computer program corresponding to the display method of device information in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, implements the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memories. In some instances, the memory 104 may further include a memory remotely provided with respect to the processor 102, and these remote memories can be connected to the mobile terminal through a network. Examples of the above network include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof. The transmission device 106 is used to receive or send data via a network. Specific examples of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one instance, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices through a base station and thus communicate with the Internet. In one instance, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0035] In this embodiment, a method for determining the risk of overburden movement running on a mobile terminal, a computer terminal, or a similar computing device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from here.

[0036] Figure 2 is a flowchart diagram of a method for determining the risk of overburden movement according to an embodiment of the present application. As Figure 2 shown, the method includes the following steps:

[0037] Step S201: Obtain relevant parameters of the virtual coal mine. Herein, the virtual coal mine is a virtual model that simulates the actual mining conditions and strata mechanical behaviors of a real coal mine. The relevant parameters include one or more of the subsidence value, displacement amount, borehole wall image, pressure value, microseismic signal, infrared image, and acoustic emission signal. The subsidence value is the vertical movement change amount of the strata collected by a dial gauge. The displacement amount is the horizontal movement change amount of the strata collected by a total station. The borehole wall image is the image of the strata collected by a borehole television. The pressure value is the stress value of the strata collected by a pressure sensor. The microseismic signal is the signal of microseismic events of the strata collected by a microseismic monitor. The infrared image is the image of the temperature distribution of the strata collected by an infrared thermal imager. The acoustic emission signal is the energy signal of elastic waves released when cracks are generated in the strata collected by an acoustic emission sensor;

[0038] Specifically, the subsidence value refers to the vertical displacement of the strata or the ground surface relative to the original position during the mining process. The vertical displacement change amount of the strata during the mining process is collected by a virtual dial gauge, reflecting the movement condition and stability of the roof.

[0039] The displacement amount includes horizontal movement. The horizontal displacement is the displacement of the strata in the horizontal direction relative to the original position during the mining process. The horizontal displacement of the strata is measured by a virtual total station to understand the lateral movement and overall deformation of the strata.

[0040] The borehole wall image is obtained by a borehole television, which is a non-destructive testing technology used to inspect the fissures, joints, and structural characteristics of the strata inside the borehole. It can capture detailed images inside the strata, helping to analyze the integrity of the strata structure, the degree of fissure development, and the internal damage of the strata. The analysis of the borehole wall image is crucial for predicting the potential damage of rocks and evaluating the stability of the strata.

[0041] The pressure value is usually measured by a pressure sensor, reflecting the stress magnitude borne by the floor or inside the strata. During the mining process, the change in the floor pressure can reveal the bearing capacity of the strata and the stress release condition, which is of great significance for evaluating the floor stability, predicting floor heave, and rock pressure manifestation.

[0042] The microseismic signal is an elastic wave signal released when microcracks are formed and expanded inside the rock under stress. The microseismic monitor can detect these signals and, through positioning technology and data analysis, determine parameters such as the location, frequency, and energy of microseismic events. The monitoring of microseismic signals helps to identify stress concentration areas inside the strata and potential strata failures, playing an important role in warning dynamic disasters such as mine tremors and rock bursts.

[0043] The infrared images are obtained by a thermal infrared imager, which reflects the temperature distribution on the surface or inside of the rock strata. Heat is generated in the rock strata under stress, and temperature anomalies can reveal the stress distribution, crack development, and rock strata failure inside the rock strata. The monitoring of infrared images helps to evaluate the mechanical state of the rock strata through thermal radiation changes and predict the potential failure and disaster risks of the rock strata.

[0044] Acoustic emission is the elastic wave energy released when cracks form or expand inside the rock. An acoustic emission monitor can detect these signals and evaluate the degree and type of rock strata failure by analyzing the frequency, intensity, and pattern of the signals. The monitoring of acoustic emission signals is of great value for real-time detection of rock strata failure and prediction of rock burst and other rock dynamic disasters.

[0045] Step S202: Conduct a risk assessment on all the above-mentioned relevant parameters to obtain multiple assessment results. Among them, the above-mentioned assessment results correspond one by one to the above-mentioned relevant parameters, and the above-mentioned assessment results characterize whether the overlying strata migration risk occurs in the virtual coal mine according to the above-mentioned relevant parameters.

[0046] Specifically, after collecting all the relevant parameters in the virtual coal mine model, the next is the risk assessment stage. This step involves conducting a separate risk analysis on each parameter to determine whether they indicate possible overlying strata migration risks.

[0047] For each monitoring parameter, risk assessment thresholds and criteria are set, such as:

[0048] When the subsidence value exceeds a certain threshold, it may indicate problems with roof stability.

[0049] An abnormal increase in the displacement may mean abnormal deformation of the rock strata.

[0050] An increase in the crack density in the borehole wall image may indicate an increase in rock strata damage.

[0051] A sudden increase in the pressure value may indicate stress concentration and increase the risk of rock strata fracture.

[0052] An increase in the frequency and energy of microseismic events indicates that the stress release and damage inside the rock strata may intensify.

[0053] An abnormal increase in temperature in the infrared image may reflect stress concentration or crack propagation inside the rock strata.

[0054] An increase in acoustic emission signals indicates an intensification of rock strata damage.

[0055] By comparing each parameter with the preset assessment criteria, an assessment result reflecting the risk status of the parameter can be generated. These assessment results are preliminary and are for each monitoring parameter individually.

[0056] Step S203: Perform data fusion based on all the above evaluation results to obtain a comprehensive analysis result. Among them, the fusion methods include at least one or more of weighted average, Bayesian fusion, information entropy fusion, and fuzzy processing fusion. The above comprehensive analysis result is used to determine whether there is a risk of overlying rock movement in the virtual coal mine;

[0057] Specifically, after separately evaluating the risks of each parameter, it is necessary to perform data fusion on all the evaluation results to obtain a comprehensive analysis result. This step is crucial because it can combine the information of multiple parameters and improve the accuracy and reliability of risk assessment.

[0058] Multiple methods can be used for data fusion, including:

[0059] Weighted average: Assign different weights to the evaluation results of each parameter and calculate the weighted average value. The weights can be determined based on the importance and reliability of the parameter for the stability of the rock formation.

[0060] Bayesian fusion: Apply Bayesian theory and comprehensively consider the risk assessment results and prior probabilities of each parameter to calculate the posterior probability, that is, the comprehensive risk.

[0061] Information entropy fusion: Based on information theory, use the information entropy values of the evaluation results of each parameter to fuse data. The information entropy reflects the uncertainty of information, and the uncertainty can be reduced during the fusion process to improve the reliability of decision-making.

[0062] Fuzzy processing fusion: Adopt fuzzy logic methods, regard the evaluation results as fuzzy sets, and fuse information through fuzzy set operations, which is suitable for processing data with high uncertainty.

[0063] Regardless of which data fusion method is used, the ultimate goal is to obtain a comprehensive analysis result that can comprehensively reflect the overlying rock movement risk state in the virtual coal mine model.

[0064] Step S204: Generate a warning message when the above comprehensive analysis result indicates that the virtual coal mine has a risk of overlying rock movement.

[0065] Specifically, generate a warning message according to the comprehensive analysis result. If the comprehensive analysis result shows that there is a risk of overlying rock movement in the virtual coal mine, then the system will trigger a warning. The warning message should include:

[0066] Risk level: Determine the severity of the risk based on the comprehensive analysis result.

[0067] Parameter analysis: Provide specific information on the monitoring parameters that cause the risk, such as which parameters are abnormal and the thresholds reached, etc.

[0068] Location and time: Indicate the specific location where the risk may occur and the predicted time.

[0069] Suggested measures: Based on the results of risk assessment, propose suggested measures to prevent or mitigate the risk of overlying rock movement, such as adjusting mining parameters, increasing support strength, etc.

[0070] In this embodiment, data is obtained through various monitoring means such as dial gauges, total stations, borehole televisions, microseismic monitors, thermal infrared imagers, pressure sensors, and acoustic emission monitors. Compared with the application of traditional single or a few monitoring devices, it can comprehensively obtain overlying rock movement information from different physical quantities, different spatial positions (vertical and horizontal directions), and different rock stratum depths, avoiding monitoring dead angles, greatly enriching the data dimension, and providing an early warning for coal mine safety production more timely and accurately than single monitoring means.

[0071] Specifically, for the solution of this application, the similar theoretical basis includes determining similarity criteria (such as geometric, mechanical, kinematic similarity, etc.) and selecting appropriate similar materials (such as gypsum, sand, iron powder, etc.). The construction of the virtual coal mine should consider the actual mining conditions and set monitoring devices such as strain gauges, displacement sensors, etc., and study the stress distribution law, the law of surrounding rock deformation and failure, etc. Integrate the technology of using devices such as total stations, dial gauges, borehole televisions, microseismic monitors, thermal infrared imagers in physical similarity simulation experiments to monitor the overlying rock movement, comprehensively analyze different monitoring data, accurately judge the law of overlying rock movement and the risk of possible dynamic disasters, and provide timely early warning and decision-making support for coal mine safety production.

[0072] Through the first similarity law (geometric law), the second similarity law (mechanical law), and the third similarity law (kinematic law) of physical similarity simulation, and the basic theory of key strata, the size of the plane strain model frame is designed to be 3m×0.2m×2.0m, and the total simulation height is 452.6m. The geometric similarity ratio of this physical similarity simulation model (i.e., virtual coal mine) is 1:250. According to the similarity criterion, the unit weight similarity ratio is 1:1.5, the stress similarity ratio is 1:375, and the time similarity ratio is 0.063. The model paving size (length×width×height) is 3.0×0.2×1.81m. During the mining process of the working face, comprehensive monitoring of the virtual coal mine is carried out using microseismic, thermal infrared, total station, borehole television, pressure sensor, and dial gauge, etc., to realize the real-time test of comprehensive information index parameters such as "sound-light-electricity" physics, mechanics, and damage of the overlying rock in the working face during the strike mining process. The virtual coal mine is as Figure 3 shown, and the installation positions of multiple devices are as Figure 4 shown, including dial gauges, total stations, borehole televisions, pressure sensors, microseismic monitors, infrared thermal imagers, and acoustic emission sensors. The specific positions of the devices are not limited, and these devices are all connected to the execution entity of this application, and the execution entity can be a controller.

[0073] Of course, the executing entity can also be a processor, a server, or other devices.

[0074] Specifically, the above solution uses a total station to obtain high-precision three-dimensional coordinates and monitor the surface displacement changes of the model; dial gauges are used for precise measurement of local height changes. The combination of the two can accurately grasp the deformation law of the overlying rock surface. High-definition probes are used to capture the images of the borehole wall to monitor the development of rock fractures inside the borehole, providing an intuitive basis for analyzing the internal damage of the overlying rock. Microseismic events inside the rock layer are monitored. By analyzing the frequency, amplitude, and energy of microseismic signals, the crack propagation inside the rock and the rock layer failure are judged, and dynamic disasters such as roof caving are predicted. Based on the temperature changes during rock mass damage and fracture, infrared thermal imaging technology is used to detect the surface temperature field of the rock mass to identify cracks and damaged areas in the rock layer. The stress changes of the floor are monitored by sensors to analyze the influence of overlying strata caving on the floor and master the law of mine pressure manifestation. The acoustic emission signals released during the rock layer failure process are monitored, and the relationship between the signal characteristic parameters (such as the total number of events and energy rate) and the degree of rock layer failure is analyzed to provide a basis for the evaluation of rock layer stability.

[0075] Through the comprehensive application of the above various monitoring technologies, the movement, stress changes, crack development, temperature changes, and acoustic emission signals during the rock layer failure process of the overlying rock can be comprehensively and systematically monitored and analyzed, so as to comprehensively judge the dynamic changes of the overlying rock and the potential dynamic disaster risks, provide more reliable and comprehensive monitoring data, and provide strong support for safety management and decision-making in coal mining.

[0076] In the specific implementation process, when the above relevant parameters include the above subsidence value, a risk assessment is carried out on all the above relevant parameters to obtain multiple assessment results, which can be achieved through the following steps: obtaining the subsidence change amount, where the subsidence change amount is the difference between the initial subsidence value and the current subsidence value, where the initial subsidence value is the subsidence value obtained at the initial moment after the virtual coal mine is built, and the current subsidence value is the subsidence value obtained at the current moment after the virtual coal mine is built; when the subsidence change amount is greater than or equal to the preset subsidence threshold, initially determining that the virtual coal mine has the risk of overlying rock movement and obtaining the first assessment result; when the subsidence change amount is less than the preset subsidence threshold, initially determining that the virtual coal mine has no risk of overlying rock movement and obtaining the second assessment result.

[0077] In this solution, by continuously monitoring the subsidence change amount, the abnormal changes in the vertical displacement of the rock layer can be detected in real time, and potential overlying rock movement risks can be discovered in time. The monitoring of the subsidence change amount takes into account the dynamic changes of the rock layer during the mining process. By comparing the initial subsidence value and the current subsidence value, the displacement situation of the rock layer can be more accurately reflected.

[0078] Specifically, first, the monitoring system needs to record the subsidence value at the initial moment after the virtual coal mine is built, which is usually the state when the rock stratum has not been disturbed by mining. Then, it will continuously monitor the subsidence value of the rock stratum at the current moment. By comparing the difference between the current subsidence value and the initial subsidence value, that is, the subsidence change amount, the vertical displacement change of the rock stratum during the mining process can be intuitively understood.

[0079] Specifically, a subsidence threshold is preset, which is determined based on engineering experience or theoretical analysis of the rock stratum stability and the characteristics of mining pressure manifestation. When the monitored subsidence change amount is greater than or equal to this preset subsidence threshold, it will be preliminarily judged that there is a risk of overlying rock movement in the virtual coal mine model, which means that the vertical displacement of the rock stratum is abnormal, and it may indicate an unstable state of the overlying rock or a potential dynamic disaster risk, such as roof caving, rock stratum rupture, etc. At this time, the first evaluation result generated is "there is a risk of overlying rock movement".

[0080] Specifically, on the contrary, if the subsidence change amount is less than the preset subsidence threshold, it will be preliminarily judged that the vertical displacement of the rock stratum in the virtual coal mine model is within the normal range and does not show abnormal vertical displacement changes, which means that the rock stratum has good stability and there is no obvious risk of overlying rock movement. At this time, the second evaluation result generated is "there is no risk of overlying rock movement".

[0081] Based on historical data, geological conditions and mining depth, a subsidence threshold is preset, which reflects the critical point of the decline in rock stratum stability. For example, for a specific type of coal seam and mining conditions, the subsidence threshold may be set to 10 mm. The calculated subsidence change amount is compared with the subsidence threshold. If the subsidence change amount is greater than or equal to the preset subsidence threshold (for example, the subsidence change amount is 15 mm), it is preliminarily judged that there is a risk of overlying rock movement in the virtual coal mine model, and the first evaluation result (risk occurs) is generated. If the subsidence change amount is less than the preset subsidence threshold (for example, the subsidence change amount is 5 mm), it is preliminarily judged that there is no risk of overlying rock movement in the virtual coal mine model, and the second evaluation result (no risk occurs) is generated.

[0082] Specifically, before the physical similarity model is built and no mining activities are carried out, a dial gauge is used to record the initial subsidence value at the top of the model as the benchmark for subsequent monitoring. During the model mining process, every time a certain mining progress or time interval is reached, the dial gauge is used to measure the subsidence value at the top again, and it is compared with the initial subsidence value to calculate the change amount of the subsidence value.

[0083] Specifically, analyze the change amount of the subsidence value to identify whether there are sudden changes or abnormal increases. Sudden changes or abnormal increases may indicate a change in the stability of the rock formation, which may be a precursor to the initial caving or periodic weighting of the rock formation. Calculate the statistical characteristics of the change amount of the subsidence value, such as the mean value, standard deviation, etc., to identify long-term or short-term subsidence trends. The long-term subsidence trend may reflect the continuous deformation of the rock formation under mining pressure, while the short-term sudden change may indicate local damage or stress release.

[0084] Specifically, by establishing a mathematical model or empirical formula between the change of the subsidence value and the stability of the rock formation, analyze the relationship between the change amount of the subsidence value and the risk of overlying rock movement. For example, if the settlement value exceeds a certain threshold, it may indicate that large-scale caving of the rock formation is about to occur. Combine other monitoring data (such as stress changes, microseismic events, etc.) to cross-verify the change of the subsidence value to improve the accuracy and reliability of risk assessment. For example, if the microseismic monitor simultaneously records an increase in microseismic events, this may further support the deterioration of the stability of the rock formation.

[0085] Specifically, according to historical data and expert judgment, set different thresholds for the change amount of the subsidence value, corresponding to different risk levels. For example, a slight change in the subsidence value may indicate a low risk, while a drastic subsidence may indicate a high risk. When the change amount of the subsidence value exceeds the set threshold, the system should automatically trigger an alarm to remind the monitoring personnel to pay attention to the possible unstable state of the overlying rock and take countermeasures in advance.

[0086] Specifically, the dial gauge monitoring should be carried out continuously, record the change of the subsidence value after each mining, and form time series data. Use time series analysis methods, such as autoregressive integrated moving average model (ARIMA), seasonal decomposition model, etc., to predict the future change trend of the subsidence value and identify possible risks in advance.

[0087] Specifically, use theoretical models of physical similarity simulation, such as stress distribution models, rock formation failure models, etc., to analyze the possible physical mechanisms behind the change of the subsidence value, in order to more deeply understand the reasons for the change of the stability of the rock formation.

[0088] Through the above steps, the risk of overlying rock movement can be effectively evaluated by using the subsidence value detected by the dial gauge. The change of the subsidence value is an intuitive indicator of the stability of the rock formation, which is closely related to mining disturbance and internal stress changes in the rock formation. Therefore, by continuously monitoring and analyzing the change of the subsidence value, the movement state of the overlying rock and the risk of potential dynamic disasters can be timely warned.

[0089] Specifically, the tiny linear movement of the measuring rod caused by the dimension to be measured is amplified through gear transmission and becomes the rotation of the pointer on the dial, so as to measure the size of the dimension to be measured. When the measuring rod moves up or down by 1 mm, the large pointer rotates one full circle through the gear transmission system, and at the same time, the small pointer rotates one grid. Each grid rotation of the large pointer has a reading value of 0.01 mm, and each grid rotation of the small pointer has a reading of 1 mm. First, read the scale line passed by the small pointer (i.e., the integer part in millimeters), then read the scale line passed by the large pointer (i.e., the decimal part), multiply it by 0.01, and then add the two together to obtain the measured value. The difference between each measured value and the initial value is basically the displacement change amount at the top of the model each time.

[0090] Specifically, in this physical similarity simulation experiment, a dial indicator is used to regularly and comprehensively monitor the movement of the top of the overlying rock model during the backfilling process of the model. The selection of the dial indicator is not limited, and it can accurately and conveniently measure narrow or recessed objects, as well as the internal and external diameter dimensions that are difficult to measure with a micrometer. The non-clutch structure is used to achieve automatic reverse of the measuring wire. Due to the design of the gemstone seat type base and the "crystal" watch cover with an O-ring, it can be waterproof and dustproof. The dial indicator is a measuring instrument with relatively high precision, mainly used for calibrating the shape and displacement errors of workpieces. It can only measure relative values and cannot measure absolute values. This dial indicator is used to monitor the subsidence value at the top of the model, and the subsidence value is recorded every time the working face is backfilled by a certain distance; by analyzing the surface displacement and top subsidence of the mined model, the movement law of the overlying rock of the working face can be mastered.

[0091] Specifically, according to the surface size of the strike physical similarity model (300*181 cm), the dial indicators are installed on the top of the strike model. To prevent the dial indicator needle from drilling into the soil layer at the top of the model, a hard abrasive paper sheet is placed under the lower part of the dial indicator needle. Dial indicators are installed within a range of 260 cm along the length of the model from 0 to 300 cm. The spacing is slightly larger near the borehole TV viewing hole, and a total of 14 dial indicators are installed at intervals of 20 cm. The layout of the dial indicators on the strike model is as Figure 5 shown.

[0092] Specifically, according to the working face backfilling plan of the strike physical similarity simulation experiment, when backfilling the W1145 working face of the B4-1 coal seam, the W1123 working face of the B2 coal seam, and the I010102 working face of the B1 coal seam in the strike direction, dial indicators are used to monitor the surface displacement data of the model and draw its displacement change diagram. By analyzing the surface displacement diagram of the model, we can more intuitively and quickly obtain the change situation of the rock layer at the top of the model and draw conclusions. According to the actual situation of the mined model, key detections are carried out on the initial caving and the change of the overlying rock after periodic weighting during the backfilling process of the model, and the change law of the top of the overlying rock model when the overlying layer caves is analyzed. Through the change distribution of the top of the overlying rock model of the model at each stage during the mining process of the model working face, the movement law of the top of the overlying rock model of the model working face during the backfilling process can be mastered.

[0093] Specifically, in the above dial gauge monitoring, the dial gauges are installed on the top of the model at a preset interval of 20 cm to capture the minute displacement changes during the movement of the overlying strata and record the displacement data after the initial caving and periodic weighting of the overlying strata.

[0094] In some embodiments, when the above relevant parameters include the above displacement amount, a risk assessment is performed on all the above relevant parameters to obtain a plurality of assessment results, which can be specifically achieved through the following steps: obtaining a preset displacement threshold; when the above displacement amount is greater than or equal to the above preset displacement threshold, preliminarily determining that there is a risk of overlying strata movement in the above virtual coal mine to obtain a third assessment result; when the above displacement amount is less than the above preset displacement threshold, preliminarily determining that there is no risk of overlying strata movement in the above virtual coal mine to obtain a fourth assessment result.

[0095] In this solution, the displacement amount monitoring can capture the dynamic changes of the rock strata in real time. Once the displacement amount reaches the preset threshold, a preliminary risk assessment result is immediately generated, standardizing the displacement amount change into an index for risk assessment, making the assessment of rock strata stability more quantitative and objective, helping to avoid the deviation of subjective judgment, and improving the accuracy and reliability.

[0096] Specifically, in the design stage of the monitoring system, a reasonable displacement threshold needs to be determined according to geological conditions, mining depth, rock stratum properties, and historical mining data. This threshold, as a benchmark for judging whether the rock stratum displacement reaches the risk level, is set based on an in-depth understanding of rock stratum stability and engineering practice experience. The preset displacement threshold should reflect the dynamic change range of the rock strata, and exceeding this range will be regarded as a signal of the decline in rock stratum stability.

[0097] Specifically, during the mining process of the virtual coal mine, high-precision monitoring equipment such as total stations is used to continuously monitor the displacement amount of the rock strata, including horizontal displacement and vertical displacement. The real-time monitoring of the displacement amount can capture the dynamic response of the rock strata under mining disturbances and provide real-time data for the assessment of rock stratum stability.

[0098] Specifically, when the detected displacement amount reaches or exceeds the preset displacement threshold, it will be preliminarily determined that there is a risk of overlying strata movement in the rock strata of the virtual coal mine model. At this time, the obtained assessment result is the third assessment result, which means that the displacement change of the rock strata may indicate abnormal deformation or potential stability problems of the rock strata, such as rock stratum shear, crack expansion, or rock stratum failure, etc.

[0099] Specifically, if the detected displacement is lower than the preset displacement threshold, it is preliminarily determined that the displacement change of the rock formation is within the normal range, the rock formation in the virtual coal mine model is relatively stable, and there is no risk of overlying rock movement. At this time, the obtained evaluation result is the fourth evaluation result, indicating that the current displacement change of the rock formation does not exceed the safety level, and the mining activities can continue, but still need to remain vigilant and conduct continuous monitoring.

[0100] Based on historical data, geological conditions, mining depth and type, a displacement threshold is preset. For example, for a specific coal seam and mining conditions, the horizontal displacement threshold may be set at 5 mm, and the vertical displacement threshold is set at 10 mm. When the detected displacement reaches or exceeds the preset displacement threshold (for example, the horizontal displacement reaches 8 mm and the vertical displacement reaches 15 mm), the monitoring system preliminarily determines that there is a risk of overlying rock movement in the rock formation of the virtual coal mine model and generates the third evaluation result (risk occurs). If the detected displacement is lower than the preset displacement threshold (for example, the horizontal displacement is 3 mm and the vertical displacement is 8 mm), the monitoring system preliminarily determines that the displacement change of the rock formation is within the safe range, there is no risk of overlying rock movement, and the fourth evaluation result (no risk occurs) is generated.

[0101] Specifically, before the physical similarity simulation experiment starts, use a total station to pre-position and initialize the key points on the model surface, and record the displacement reference data when not affected by mining. Ensure that the accuracy settings of the total station match the experimental requirements, and at the same time adjust the parameters to ensure the most accurate data collection.

[0102] Specifically, according to the experimental plan, the total station regularly or continuously monitors the change of the model surface displacement. Record the displacement data, including but not limited to vertical displacement and horizontal displacement, as well as the time and coordinate information of the displacement.

[0103] Specifically, conduct preliminary processing on the collected displacement data to remove noise and outliers. Calculate the displacement change amount, identify the trend and pattern of the displacement change. Use statistical analysis methods (such as mean, standard deviation, trend line, etc.) to describe the change characteristics of the displacement. Compare with the physical similarity simulation theoretical model to analyze whether the displacement change conforms to the expected mining effect.

[0104] Specifically, based on historical data, theoretical models and geological conditions, set the risk threshold for displacement change. For example, if the displacement change exceeds a specific threshold, it may indicate a decline in the stability of the rock formation. The risk threshold should take into account multiple factors such as geological structure, mining depth, and rock formation type.

[0105] Specifically, when the displacement change exceeds a preset risk threshold, evaluate the risk of dynamic disasters. Combine other monitoring data, such as stress, microseismic, thermal infrared data, etc., for multi-source information fusion to more comprehensively evaluate the risk. If the comprehensive evaluation shows a significant risk, the system should issue a warning to remind the monitoring personnel to take preventive measures.

[0106] Specifically, use time series analysis methods, such as ARIMA models, trend extrapolation, etc., to predict the future change trend of displacement. Trend prediction helps to identify potential risks in advance and gain time for taking response measures.

[0107] Specifically, by comparing the displacement data with the geological theoretical model, deeply understand the geomechanical background of displacement changes. Analyze the relationship between displacement changes and rock layer structure, geological stress field to more accurately predict the risk of overlying rock movement.

[0108] Specifically, according to the results of displacement monitoring, dynamically adjust the monitoring frequency and range of the total station to more effectively capture risk signals. If the data shows a high displacement change rate in a specific area, the monitoring density of this area can be increased.

[0109] Through the above steps, the displacement data detected by the total station is converted into an assessment and warning of the risk of overlying rock movement. Displacement data is direct evidence of the impact of mining on rock layers, which can reveal the movement trend, deformation degree and stability changes of rock layers. Combining in-depth geological theoretical analysis and multi-source information fusion can achieve precise monitoring and warning of dynamic disaster risks, providing strong support for safe decision-making in coal mine mining.

[0110] Specifically, the total station can achieve plane and elevation measurements. It is simple to operate, flexible and convenient, and has strong data storage capabilities. It is a simple, practical and highly accurate observation method. Using the total station as the observation method in similar material simulation, vertical and horizontal displacements can be observed simultaneously; it overcomes the reading errors that occurred when using the theodolite in the past when the workload was too large or the light was dim. The total station polar coordinate method observation meets the requirements of the measurement accuracy of similar simulation tests, can accurately determine the fault activation time, measure the fault activation amount, and reflect the overlying rock movement law. It is an effective observation method in similar material simulation tests.

[0111] The observation principle of establishing a plane rectangular coordinate system by backsight orientation is as follows.

[0112] The first step of the total station polar coordinate method is backsight orientation, that is, according to the coordinates of two known points, use the built-in program of the total station to calculate the coordinate azimuth angle ∠AB of the connection direction between the measuring station A and the backsight point B. The layout of the observation device is as Figure 6, The measuring station A is selected at a position 5 m in front of the test bench, and a total station is set up. The backsight point B is selected at a flat position on the same side of the test bench and has good visibility with the measuring station A. At the same time, both points A and B are marked with reflective sheets. After the starting and ending point coordinates are determined, their azimuth angles are uniquely determined, and the coordinate north direction can be indirectly calculated, thus setting up a plane rectangular coordinate system for the total station.

[0113] Let the coordinates of the measuring station A and the backsight point B be (x A , y A , z A ) and (x B , y B , z B ), and the point b(x B , y B , z A ) is the projection of point B in the plane XOY. The coordinate azimuth angle of the straight line AB is equal to the coordinate azimuth angle of the straight line Ab. Using the principle of coordinate inverse calculation, the coordinate azimuth angle ∠Ab and the horizontal distance D Ab .

[0114]

[0115] The observation principle of plane coordinate calculation is as follows.

[0116] Using the total station to observe the direction, zenith distance (βAD), and inclined distance (SAD) from the measuring station A to the point to be measured D (reflective sheet), the plane coordinates of the point to be measured D(xD, yD, zD) can be obtained through coordinate forward calculation, as Figure 7 shown.

[0117] β A d = 90° - β A d;

[0118] DAd = SAdcosβAd;

[0119] αAd = αAd + βbd;

[0120] x D = xA + DAd c osαAd;

[0121] y D = yA + DAdsinαAd;

[0122] Similarly, the plane coordinates of point D' can be obtained.

[0123] x D' ＝x A + DAd' cosα Ad';

[0124] y D′ = y A + DAd′sinαAd′.

[0125] Specifically, a total station is selected for experimental monitoring. The instrument is equipped with a three-axis compensator, which ensures the angular measurement accuracy of the instrument. Moreover, the instrument has a built-in zero-point correction function, which can correct the index error. The instrument is sensitive and is provided with meteorological correction, earth curvature correction, etc. When observing, the instrument has a large-screen display window, which can display the inclination angle, horizontal angle and distance at the same time. It adopts a double-speed micro-motion screw, can quickly aim at the target, and is flexible and convenient to operate, small and highly efficient.

[0126] Specifically, according to the surface size of the strike physical similarity model (300*181 cm) and the severity of overlying strata movement at different heights of the working face, a total of 17 rows of total station monitoring lines are arranged along the vertical height direction of the model above the coal seam. The overlying strata near the working face are active, and measuring points are arranged within the range of 20 - 280 cm along the horizontal direction of the model. A row of measuring points with an interval of 10 cm is arranged in the middle of B1 coal and B2 coal, with 27 measuring points. Two rows of measuring points with an interval of 10 cm are arranged in the middle of B2 coal and B4-1 coal, with a total of 54 measuring points. 14 rows of measuring points with a row spacing of 10 cm and a column spacing of 10 cm are arranged above B4-1 coal, with 27 measuring points in each row. The total station monitoring origin is arranged at the lower left corner of the model. The arrangement of total station monitoring points for the strike model is as Figure 8 shown.

[0127] Specifically, according to the mining plan of the working face in the strike physical similarity simulation experiment, when mining the B4-1 coal working face, B2 coal seam working face and B1 coal seam working face of the strike model, the total station is used to collect the surface displacement data of the model and draw the displacement nephogram. By analyzing the displacement nephogram of the model surface, we can more intuitively and quickly obtain the caving situation of the overlying strata and draw conclusions. According to the actual situation of the model after mining, the key is to focus on detecting the initial caving and the changes of the overlying strata after periodic weighting during the mining process of the model, and analyze the change law of the overlying strata surface when the overlying strata cave. Through the distribution of the overlying strata deformation on the model surface at each stage during the mining process of the model working face, the movement law of the overlying strata on the model surface during the mining process of the model working face can be mastered.

[0128] Specifically, the above total station monitoring includes establishing a plane rectangular coordinate system, and calculating the plane coordinates of the measuring points on the model surface through direction, zenith distance and inclined distance measurement, so as to realize the accurate monitoring of the model surface displacement.

[0129] In the specific implementation process, when the above-mentioned relevant parameters include the above-mentioned hole wall image, risk assessment is performed on all of the above-mentioned relevant parameters to obtain multiple assessment results, which can be achieved through the following steps: computer vision technology is used to perform image recognition on the above-mentioned hole wall image to obtain a recognition result, wherein the above-mentioned recognition result characterizes whether the rock formation has cracks; when the above-mentioned recognition result characterizes that the rock formation has cracks, it is preliminarily determined that the above-mentioned virtual coal mine has a risk of overburden migration, and a fifth assessment result is obtained; when the above-mentioned recognition result characterizes that the rock formation has no cracks, it is preliminarily determined that the above-mentioned virtual coal mine has no risk of overburden migration, and a sixth assessment result is obtained.

[0130] In this solution, real-time monitoring of borehole wall images and crack identification can promptly detect changes in the internal structure of the rock formation. Computer vision technology can identify cracks in borehole wall images with high precision. Compared with manual observation, computer vision technology can detect cracks more quickly and accurately.

[0131] Specifically, during the simulation experiment, high-definition images of the inside of the rock formation borehole were collected through the borehole television monitoring system. These images showed in detail the structural characteristics of the rock formation, including the distribution, size and shape of the cracks.

[0132] Specifically, the acquired borehole wall image is input into the computer vision system, and the image is analyzed using image processing algorithms (such as edge detection, feature extraction, machine learning models, etc.) to identify the cracks inside the rock formation. This step usually involves the following sub-steps:

[0133] Image preprocessing: Perform preprocessing operations such as denoising, contrast enhancement, and image correction on the hole wall image to improve the accuracy of crack detection.

[0134] Crack feature extraction: Computer vision technology is used to extract crack features, such as the position, length, width and direction of the crack, from the preprocessed hole wall image.

[0135] Fracture identification and classification: Based on the fracture characteristics, machine learning or deep learning models are used to identify and classify fractures to determine whether there are fractures inside the rock formation, and further analyze the type (such as tension and shear fractures) and scale of the fractures.

[0136] Specifically, when the borehole wall image recognition results show that there are cracks inside the rock formation and the characteristics of the cracks indicate that the stability of the rock formation is reduced, the monitoring system preliminarily determines that there is a risk of overburden migration in the virtual coal mine model and generates the fifth assessment result, indicating that the development of cracks inside the rock formation may indicate abnormal deformation of the rock formation or potential stability problems, such as rock shear, crack expansion or rock formation destruction.

[0137] Specifically, if the recognition result of the borehole wall image shows that there are no obvious fractures inside the rock formation, or the characteristics and scale of the fractures are within the safety range, the monitoring system initially judges that the stability of the rock formation is good, there is no risk of overlying strata movement, and a sixth evaluation result is generated, indicating that the current internal structural changes of the rock formation do not exceed the safety level, and the mining activities can continue, but still need to remain vigilant and conduct continuous monitoring.

[0138] Specifically, before the experiment starts, use borehole television to collect initial images of the inside of the rock formation that is not affected by mining as a reference baseline. During the experiment, regularly or continuously monitor the development of fractures inside the rock formation and collect high-definition image data of the inside of the rock formation.

[0139] Specifically, preprocess the collected images, including steps such as denoising, enhancing contrast, and image correction, to improve the accuracy of subsequent analysis.

[0140] Specifically, use image processing and computer vision technologies, such as edge detection and pattern recognition algorithms, to automatically identify the characteristics such as the location, length, width, and direction of fractures inside the rock formation. For complex images, it may be necessary to combine manual annotation and machine learning algorithms to improve the accuracy of fracture recognition.

[0141] Specifically, compare the fracture images at different stages and analyze the development trend of fractures. The expansion, convergence, and bifurcation of fractures may indicate changes in the internal stress distribution of the rock formation and a decline in the stability of the rock formation. Statistically analyze the changes in the number, size, and distribution of fractures to identify potential failure modes.

[0142] Specifically, in addition to fractures, borehole television images can also reveal other changes in the internal structure of the rock formation, such as changes in bedding and the exposure of weak planes. These changes may indicate local failure or overall instability of the rock formation.

[0143] Specifically, based on the analysis of fractures and structural changes, set a risk threshold. For example, when the average length or width of fractures exceeds a certain threshold, or the number of fractures increases significantly in a short period of time, it may indicate a decline in the stability of the rock formation and a risk of dynamic disasters.

[0144] Specifically, by comparing the fracture images in the experimental stage with the reference images, and the dynamic development of fractures with the set risk threshold, evaluate whether there is a risk of rock formation stability. If the fracture development exceeds the risk threshold, or significant changes in the rock formation structure occur, a warning should be issued to alert the monitoring personnel of possible dynamic disasters.

[0145] Specifically, conduct correlation analysis and data fusion on the borehole television image data and other monitoring technologies (such as total station displacement data, microseismic monitoring data, etc.). Combine multi-source monitoring data to evaluate the stability of the rock formation from different dimensions and improve the accuracy and reliability of dynamic disaster risk warning.

[0146] Specifically, more advanced image analysis techniques, such as deep learning neural network models, are used to conduct a more in-depth analysis of the fracture images. These models can identify the complex patterns of fractures, predict the future development trends of fractures, and give early warnings of possible dynamic disaster risks.

[0147] Specifically, according to the development of fractures inside the rock strata, the monitoring frequency and depth of the borehole television are dynamically adjusted. For areas where fractures develop rapidly, the monitoring density is increased to ensure that detailed information on the changes in the rock strata can be captured in a timely manner.

[0148] Through the above steps, the borehole television image data is converted into an assessment and early warning signal for the overlying rock movement risk. The development of fractures and the changes in the internal structure of the rock strata are direct indicators of the rock strata stability and dynamic disaster risk, which can help the monitoring personnel identify and take measures to prevent possible disasters in a timely manner. Combining multi-sensor data fusion analysis, a more comprehensive and accurate rock strata stability monitoring system can be constructed.

[0149] Specifically, the borehole television uses a high-definition probe with a built-in camera to capture the hole wall image, and the image data is transmitted to the controller and computer via a cable. The main types of data collected include in-hole videos and stitchable image data. The in-hole videos can be used to observe the hole wall damage conditions at different stages by playing back the videos; the stitchable image data can be digitally synthesized through high-definition intelligent software to obtain a digital core image. The core can be freely rotated and observed at any angle, or the core can be unfolded into a 360° unfolded view. Based on the interpretation of these two types of borehole television data, the internal rock stratum damage conditions and fracture development conditions of the model overlying rock can be studied, so as to obtain the fracture range and fracture height of the model overlying rock.

[0150] Specifically, the 4D ultra-high-definition full-intelligent in-hole television adopts advanced image acquisition and processing techniques, with high system integration and large memory. The test window of this system is 360° full-hole observation (vertical holes, horizontal holes, inclined holes, downward and upward angle holes). It solves the technical problem that it is difficult to quantitatively observe and determine relevant geological parameters in horizontal holes and inclined holes by previous borehole televisions. The system realizes high intelligence, automatically realizes the lifting and lowering of the probe and the real-time unfolding and stitching of the full-hole video image during the test process. During the acquisition state, the real-time moving video of the probe in the hole can be seen. The instrument is easy to operate, stable and reliable. At the same time, the digital core image displayed by the ultra-clear television realizes the accurate monitoring of the number and orientation of fractures in the borehole.

[0151] Specifically, according to the requirements of the physical trend simulation experiment, two identical drill holes are arranged in the physical similarity experiment model, which are evenly arranged along a model frame with a length of 300 cm and a height of 181 cm. The diameter of each drill hole is 50 mm. The two drill holes divide the experimental model into three parts. The hole numbers from left to right are: 1# drill hole and 2# drill hole. The 1# and 2# drill holes are each 100 cm away from the left and right boundaries of the model frame, and the hole spacing between adjacent two drill holes is 100 cm. The peep holes of the borehole television for the trend are arranged as Figure 9 shown.

[0152] Specifically, the trend similarity simulation borehole television monitoring is completed in an indoor laboratory. After the physical material model is arranged and before the experiment, the installation and debugging of the relevant borehole television equipment are carried out first. The parameters are adjusted and the distribution law of the overlying strata above the coal seam before excavation is observed. The excavation process is completed through geometric similarity. According to the working face mining plan, the B4-1 coal seam I010408 working face, B2 coal seam I010206 working face and B1 coal seam I010102 working face of the trend model are mined successively. During the excavation process, the movement law of the rock strata is monitored by borehole television under the condition of synchronous mining disturbance. The experimental excavation method is manual successive excavation, which helps to reduce external interference and can accurately simulate the deformation and failure characteristics of the rock under the real in-situ stress conditions and the mining disturbance during the actual excavation process. According to the actual mining situation of the model, the change law of the overlying strata inside the drill hole after the initial caving and periodic weighting during the model mining process is detected. For the image acquisition and processing technology, the vertical core of the 360° full-hole observation and the change of the overlying strata in the plane expansion of the figure are carried out, and the azimuth angle of the fitting measuring points for the fissures within the stage is determined, and the change law of the number of fissures is monitored.

[0153] Specifically, in the above-mentioned borehole television monitoring, a 4D ultra-high-definition full-intelligent in-hole television is used to collect the hole wall images, analyze the internal fissure characteristics of the rock strata, and determine the internal broken range and fissure height of the overlying strata.

[0154] In some embodiments, when the above-mentioned relevant parameters include the above-mentioned pressure value, a risk assessment is carried out on all the above-mentioned relevant parameters to obtain a plurality of assessment results, which can be specifically realized through the following steps: obtaining a preset pressure threshold; when the above-mentioned pressure value is greater than or equal to the above-mentioned preset pressure threshold, preliminarily determining that the above-mentioned virtual coal mine has a risk of overlying strata movement to obtain a seventh assessment result; when the above-mentioned pressure value is less than the above-mentioned preset pressure threshold, preliminarily determining that the above-mentioned virtual coal mine has no risk of overlying strata movement to obtain an eighth assessment result.

[0155] In this solution, by real-time monitoring the internal pressure of the rock strata, the abnormal change of the rock strata pressure can be quickly identified, providing immediate information for the early warning of the risk of overlying strata movement and further improving the accuracy.

[0156] Specifically, a preset pressure threshold is set according to the physical properties of the rock strata, geological conditions, mining depth, and historical data. This threshold serves as the criterion for judging whether there is a risk of overlying strata movement and usually reflects the maximum safe pressure that the rock strata can withstand. The setting of the preset pressure threshold needs to comprehensively consider factors such as the strength, elastic modulus, Poisson's ratio of the rock strata, and mining disturbances.

[0157] Specifically, high-precision pressure sensors are used to monitor the rock strata in real time. The sensors can be buried at key positions inside the rock strata, such as the contact areas between the coal seam and the roof and floor, as well as the areas with concentrated internal stress of the rock strata. By monitoring the pressure changes inside the rock strata, the dynamic changes in the stability of the rock strata can be captured, especially the abnormal increase in pressure under mining disturbances.

[0158] Specifically, when the pressure value is greater than or equal to the preset pressure threshold, the monitoring system preliminarily judges that there may be a risk of overlying strata movement in the virtual coal mine model. This usually means that the stress state inside the rock strata has reached the critical point, and the rock strata may be about to deform or break, generating the seventh evaluation result.

[0159] Specifically, when the pressure value is less than the preset pressure threshold, the monitoring system preliminarily judges that the rock strata in the virtual coal mine model are in good stability and there is no risk of overlying strata movement. This indicates that the pressure change inside the rock strata is within the safe range, and the rock strata are still in a stable state, generating the eighth evaluation result.

[0160] In a certain physical similarity simulation experiment, the rock strata pressure monitoring of the virtual coal mine model was used to evaluate the risk of overlying strata movement. The rock strata characteristics of the model are similar to those of an actual coal mine, including a hard roof and a soft floor, and the mining depth is set at 300 meters. According to the rock strata properties, mining depth, and historical mining experience, the preset pressure threshold is set at 30 MPa. Once the rock strata pressure reaches or exceeds 30 MPa, it will be preliminarily judged that there is a risk of overlying strata movement. During the simulated mining process, the pressure data inside the rock strata are automatically read every 5 minutes. In a certain monitoring, the recorded pressure value inside the rock strata was 32 MPa, exceeding the preset pressure threshold. It was preliminarily judged that there was a risk of overlying strata movement, generating the seventh evaluation result. At another monitoring point, the pressure value inside the rock strata was 27 MPa, lower than the preset pressure threshold. It was preliminarily judged that there was no risk of overlying strata movement, generating the eighth evaluation result.

[0161] Specifically, before the experiment starts or before mining, the pressure sensors are initialized to record the baseline data of the mine pressure when not affected by mining. Mine pressure data are collected regularly or continuously to ensure the real-time and integrity of the data.

[0162] Specifically, the collected mine pressure data are cleaned to remove noise and outliers. Filtering techniques, such as low-pass filtering or high-pass filtering, are used to enhance the smoothness of the data and remove the noise caused by non-mine pressure.

[0163] Specifically, analyze the trend of mine pressure data over time to identify whether there are obvious upward or downward trends. Evaluate the volatility of mine pressure by calculating statistical measures such as the average value, standard deviation, maximum value, and minimum value of mine pressure.

[0164] Specifically, detect the peaks in the mine pressure data, especially abnormally high peaks, which may indicate that the rock formation is about to fail or stress is being released. If mine pressure peaks occur frequently within a short period, this may indicate that the stability of the rock formation is deteriorating.

[0165] Specifically, identify the outliers in the mine pressure data, including sudden large fluctuations or values outside the range of historical data. Outliers may indicate the instability of local rock formations or potential geological structure problems.

[0166] Specifically, set the threshold for mine pressure risk based on historical mine pressure data, rock formation characteristics, mining depth, and geological conditions. When the mine pressure data exceeds the preset threshold, the system should trigger an alarm to indicate that there may be a risk of dynamic disasters. Threshold setting needs to consider various factors, such as the bearing capacity of the rock formation and the expected impact of mining disturbances.

[0167] Specifically, monitor the mine pressure data in real time. When abnormal mine pressure data is detected, conduct a detailed analysis immediately. Dynamically adjust the monitoring frequency or location of the pressure sensors according to the changes in mine pressure to more accurately capture the changes in the state of the rock formation.

[0168] Specifically, combine the data from other monitoring technologies (such as borehole television, microseismic monitoring, displacement measurement, etc.) for multi-source information fusion analysis. Comprehensively consider the physical changes and mechanical behaviors of the rock formation, and improve the accuracy and comprehensiveness of risk assessment through the correlation analysis between data.

[0169] Specifically, input the mine pressure data into a geomechanics model to predict the stress distribution and possible failure modes of the rock formation. The geomechanics model can be based on finite element analysis, discrete element analysis, or other numerical simulation technologies to provide theoretical support for the assessment of rock formation stability.

[0170] Specifically, submit the analysis results of the mine pressure data to geological and mining engineering experts for manual review and professional assessment. The judgment of experts can be based on experience and theoretical knowledge to provide deeper insights into risk assessment.

[0171] Through the above steps, the mine pressure data detected by the pressure sensors can be converted into an assessment of the risk of overlying strata movement, helping decision-makers take timely measures to prevent potential dynamic disasters and ensure the safe progress of coal mine mining. Mine pressure data is an important indicator of rock formation stability, and its changes directly reflect the internal stress state of the rock formation. Therefore, accurate monitoring and in-depth analysis of mine pressure data are the keys to assessing the risk of overlying strata movement.

[0172] Specifically, the main parameters of the experimentally customized pressure sensor are shown in Table 1. The sensor range is 0 - 100 kg, and the comprehensive accuracy is ≤0.05% F.S. It has the advantages of convenient installation and high measurement accuracy. The piezoelectric element of the unidirectional force sensor uses an xy-cut quartz crystal. Utilizing its longitudinal piezoelectric effect, the force-electricity conversion is achieved through 11d. Two piezoelectric wafers are stacked together along the electrical axis direction and connected in parallel. The middle is a sheet-shaped electrode (negative electrode), which collects negative charges. The base and the force-transmitting cover form the positive electrode, and the insulating sleeve isolates the positive and negative electrodes to achieve the pressure monitoring effect.

[0173] Table 1

[0174] Model CL-YB-114B Range 0~100Kg Comprehensive accuracy ≤0.05%F.S Supply voltage DC5~18V Output sensitivity 2.0±0.1mV / V Zero output ≤3%F.S Temperature zero drift ≤0.03%F.S / 10℃ Temperature sensitivity drift ≤0.03%F.S / 10℃ Creep ≤0.05%F.S / 30min Insulation resistance 2000MΩ / 50VDC Operating temperature -20～60℃ Safe overload 120%F.S Loading method Tension / Compression Material Alloy steel

[0175] Specifically, in this physical similarity simulation experiment, a floor pressure sensor is used to monitor the change of the floor pressure of the working face in real time during the model mining process. The rated working resistance of the sensor is 32.0 MPa. The weight acting on the sensor through the overlying strata is used to monitor the pressure change during the mining of the working face, analyze the change law of the floor pressure of the working face when the overlying strata collapse, and understand the law of the mine pressure appearance of the model working face during the mining process through the distribution of the bottom pressure of the model working face during the mining process.

[0176] Specifically, the layout schematic diagram of the strike model stress gauges is as Figure 10 shown. A total of 60 stress gauges are arranged at the same interval. According to the actual mining sequence of the mine strike model, in the physical material model, according to the mining plan, the floor pressure sensors arranged in the rock stratum measure the floor pressure. To ensure that the monitoring data of the floor pressure sensor is carried out from beginning to end, the floor pressure sensor is put into the designed model for construction after zeroing. Through the monitoring results of the channel data acquisition software, 3 groups of initial data of the floor pressure sensor are recorded, and their average value is taken. When the floor is significantly pressured or collapses, 3 groups of force values of the support sensors are recorded again and their average value is taken as the floor pressure value for pressure analysis.

[0177] Specifically, the strike similarity simulation and borehole television monitoring are completed in the indoor laboratory. Before the physical material model is arranged and the experiment is carried out, the initial in-situ rock stress is monitored first. According to the working face mining plan, the B4-1 coal seam I010408 working face, B2 coal seam I010206 working face and B1 coal seam I010102 working face of the strike model are mined successively. During the excavation process, the distribution law of the mine pressure of the rock strata where the floor stress is monitored under the condition of synchronous mining disturbance is carried out. According to the actual situation of the post-mining model, the stress changes of the floor during the initial caving and periodic weighting in the process of model mining are detected emphatically, and the change law of the floor pressure of the working face when the overlying strata collapse is analyzed. Through the distribution of the bottom pressure during the mining process of the model working face, the law of mine pressure manifestation during the mining process of the model working face is mastered.

[0178] In the specific implementation process, in the case of the above-mentioned relevant parameters including the above-mentioned microseismic signals, risk assessment is carried out on all the above-mentioned relevant parameters to obtain multiple assessment results, which can be realized through the following steps: feature recognition is carried out on the above-mentioned microseismic signals to obtain the microseismic feature parameters of the above-mentioned microseismic signals, where the above-mentioned microseismic feature parameters include one or more of position, frequency, amplitude, and duration; a microseismic recognition model is constructed, where the above-mentioned microseismic recognition model is trained using multiple groups of training data, and each group of training data in the above-mentioned multiple groups of training data includes historical microseismic signals obtained within a historical time period and the corresponding historical ninth assessment results of the above-mentioned historical microseismic signals; the above-mentioned microseismic signals are input into the above-mentioned microseismic recognition model to obtain the ninth assessment result corresponding to the above-mentioned microseismic signals, where the above-mentioned ninth assessment result is used to characterize whether the overlying strata migration risk of the above-mentioned virtual coal mine has occurred according to the above-mentioned microseismic signals.

[0179] In this solution, through the trained microseismic recognition model, the overlying strata migration risk can be predicted with high precision based on the characteristic parameters of the microseismic signals. Compared with the traditional simple judgment based on thresholds, the model prediction can consider the complexity and multi-dimensional characteristics of the microseismic signals, reducing false alarms and missed alarms.

[0180] Specifically, microseismic signals contain rich rock dynamics information, such as the position, frequency, amplitude, and duration of microseisms. These characteristic parameters are directly related to the stress state inside the rock strata and the degree of rock strata damage. Through computer algorithms, such as wavelet analysis, Fourier transform, deep learning and other methods, feature recognition is carried out on the microseismic signals to extract the key parameters reflecting the stability of the rock strata.

[0181] Specifically, a microseismic recognition model is constructed through machine learning or deep learning techniques. The purpose of model training is to find the correlation law between the microseismic feature parameters and the overlying strata migration risk, and to be able to predict the preliminary assessment result of the overlying strata migration risk based on the characteristics of the current microseismic signals. This model needs to go through the following key steps:

[0182] Data Preparation: Collect historical microseismic signal data obtained within a historical time period, and at the same time record the corresponding historical evaluation results (the ninth evaluation result) as labels for training the model.

[0183] Feature Engineering: Preprocess the historical microseismic data, such as denoising, feature extraction, data standardization, etc., to improve the model training effect.

[0184] Model Training: Use the historical microseismic signal data and the corresponding labels (historical evaluation results) to train the model, and adjust the model parameters so that it can predict the overlying strata migration risk from the microseismic signals.

[0185] Model Validation: Evaluate the accuracy, reliability, and generalization ability of the model through the reserved validation set data to ensure that the model can make accurate risk assessments on unknown data.

[0186] Specifically, input the real-time monitored microseismic signals into the trained microseismic recognition model. The model will predict the preliminary evaluation result of the overlying strata migration risk according to the microseismic characteristic parameters (such as location, frequency, amplitude, duration). If the model predicts the existence of overlying strata migration risk, that is, generate the ninth evaluation result, indicating that according to the current microseismic signals, it is preliminarily judged that there may be overlying strata migration risk in the virtual coal mine model. If the model predicts that there is no overlying strata migration risk, that is, generate a risk-free evaluation result.

[0187] Specifically, the microseismic monitor continuously collects the microseismic signals in the mine. The preprocessing stage includes data cleaning, noise removal, and possible signal enhancement to ensure the accuracy and availability of the data.

[0188] Specifically, identify microseismic events by analyzing the characteristic parameters of microseismic signals, such as frequency, amplitude, duration, etc. Use the time difference and waveform characteristics of multi-channel microseismic signals, and adopt inversion technology to locate the precise position of microseismic events, which helps to understand the distribution and change of stress in the rock strata.

[0189] Specifically, conduct parameter analysis on the detected microseismic events, including the frequency, energy, dominant frequency, spectral characteristics, etc. of the events. These parameters can reveal the stress state and failure behavior inside the rock strata. For example, the increase in the frequency and energy of microseismic events may indicate a decrease in the stability of the rock strata.

[0190] Specifically, conduct trend analysis on the time series data of microseismic events to identify whether the frequency and energy of the events show an upward trend. Trend analysis can predict the possible unstable state of the rock strata in advance and help predict potential failures or dynamic disasters.

[0191] Specifically, based on historical data, mine geological conditions, and mining situations, set the risk thresholds for microseismic events. For example, when the event energy exceeds a certain level, the event frequency increases significantly, or multiple events occur within a short period, it may indicate an increase in risk.

[0192] Specifically, classify microseismic events through machine learning or pattern recognition techniques to distinguish natural microseisms from those induced by mining. Microseisms induced by mining are more likely to be related to the risk of strata movement and require special attention.

[0193] Specifically, when the monitored microseismic signals meet the preset risk thresholds, the system should immediately trigger an alarm. The alarm signals should include key information such as the location, time, and energy magnitude of the microseismic events for mine engineers and safety personnel to take necessary countermeasures.

[0194] Specifically, conduct integrated analysis by fusing microseismic monitoring data with data from other monitoring techniques (such as stress measurement, displacement monitoring, borehole television image analysis, etc.) to comprehensively evaluate the stability of the strata. The mutual verification of data from different monitoring techniques can improve the accuracy and reliability of risk assessment.

[0195] Specifically, input the microseismic monitoring data into a geomechanical model to verify the prediction results of the model. At the same time, the microseismic data can serve as an important input for model calibration and optimization, enabling the model to more accurately simulate the stress state and failure behavior of the strata.

[0196] Specifically, an expert system or manual analysis of the correlation between microseismic signals and risk assessment, combined with the experience and judgment of mine engineers, evaluates the potential impact of mining activities on the stability of the strata and guides the adjustment of the mining plan or the strengthening of safety measures.

[0197] Through the above steps, the microseismic signals detected by the microseismic monitor can be transformed into an assessment of the stability of the strata and the risk of dynamic disasters, helping the mine management to take timely measures to prevent potential strata failures and dynamic disasters and ensuring the safe progress of coal mining. Microseismic signals are a direct reflection of the internal stress changes and failure activities in the strata, and their analysis and utilization have important value in the safety monitoring of coal mining.

[0198] Specifically, tests have shown that as the rock is gradually pressurized, its internal micro-defects are fractured, expanded, or closed, and at this time, acoustic emissions with very low energy levels are generated. When the cracks expand to a certain scale and the loading strength of the rock approaches half of its failure strength, large-scale crack penetration begins to occur and acoustic emissions with larger energy levels are generated, which are called "microseisms" or MS. When the pressure is closer to the ultimate strength of the rock, the number of microseismic events increases until the rock fails. Each microseismic signal contains rich information about the internal state changes of the rock mass. Processing and analyzing the received microseismic signals can be used as a basis for evaluating the stability of the rock mass. Therefore, this characteristic of rock mass microseisms can be utilized to monitor the stability of the rock mass, thereby predicting major dynamic disaster phenomena in mines such as roof caving, mine water inrush, and rock bursts.

[0199] Specifically, in this experiment, a new generation of microseismic monitoring instrument (SOS microseismic monitoring instrument) was used. The SOS microseismic monitoring system can realize long-distance (up to 10 km), real-time, dynamic, and automatic monitoring of mine seismic signals (including rock bursts) in the mine, and record the complete waveforms of the seismic signals. Through data processing, the time, energy, and three-dimensional coordinates in space of vibrations with energy greater than 100 J can be accurately calculated, the vibration type of each vibration can be determined, and the force source of the rock burst can be judged. The fracture information of the overlying strata of the working face can be analyzed, and the movement law of the spatial strata structure and the migration and evolution law of the stress field can be realized. In this similar simulation experiment, microseismic monitoring technology was used to identify the fracture signals of the strata during the advancement of the working face, and record the changes in microseismic signals inside the strata when the overlying strata fracture and the working face is subjected to pressure, so as to better analyze the energy evolution law of overlying strata failure under the influence of mining in the working face.

[0200] Specifically, in this similar simulation experiment, a new generation of microseismic monitoring instrument (SOS microseismic monitoring instrument) designed and manufactured was used. The geometric similarity ratio of the strike model in this physical similarity simulation experiment is 1:250 (model: actual), and it was built based on the Kuangou Coal Mine. Five microseismic probes were paved in the model, and the dimensions are as shown in the model in the figure. The speed probe numbers are 1#, 2#, 3#, 4#, and 5# in sequence, and the layout positions are as follows Figure 11 shown.

[0201] Specifically, the strike similar simulation microseismic monitoring was completed in the indoor laboratory. The excavation process was completed through stress similarity and volume similarity, and the mining was carried out step by step, with the working face advancing along the strike. During the process of advancing a unit length each time, the movement law of the strata under the condition of mining disturbance was monitored synchronously by microseismic monitoring. The experimental excavation method was manual successive excavation, which helps to accurately simulate the deformation and failure characteristics of the strata under real in-situ stress conditions and mining disturbance during the actual excavation process. The specific process is as Figure 12 shown, including the following steps:

[0202] Experimental preparation: Monitor the layout, set parameters, and prepare for recording;

[0203] Start the experiment: Monitor and record the experimental data;

[0204] Data processing: Separate the wave groups, extract the signals, visualize the signals, locate the microseismic events according to the P-waves, calculate the energy magnitude, and record the positions and energies of the microseismic events.

[0205] Specifically, the above microseismic monitoring includes using an SOS microseismic monitor to monitor microseismic events inside the rock formation in real time, and determining the positions, energies, and types of microseismic events through signal processing and analysis to warn of disasters such as roof caving, mine water inrush, and rock burst.

[0206] In some embodiments, when the above relevant parameters include the above infrared image, a risk assessment is performed on all the above relevant parameters to obtain multiple assessment results, which can be specifically achieved through the following steps: Perform image recognition on the above infrared image to determine whether there is an abnormal area in the above infrared image, and obtain an infrared recognition result, where the above abnormal area is a part with a temperature distribution different from other areas; when the above infrared recognition result indicates that there is the above abnormal area in the above infrared image, preliminarily determine that there is a risk of overlying rock movement in the above virtual coal mine to obtain a tenth assessment result; when the above infrared recognition result indicates that there is no such abnormal area in the above infrared image, preliminarily determine that there is no risk of overlying rock movement in the above virtual coal mine to obtain an eleventh assessment result.

[0207] In this solution, a temperature distribution image of the rock formation is obtained in real time through an infrared thermal imager, which can quickly identify the temperature abnormal area and provide real-time temperature information for the warning of overlying rock movement risk. The infrared thermal imager forms an image by detecting the infrared radiation emitted by the object, without direct contact with the rock formation, avoiding the disturbance of the monitoring equipment to the rock formation and ensuring the accuracy and reliability of the monitoring results.

[0208] Specifically, during the experiment, a high-precision infrared thermal imager is used to perform real-time infrared imaging on the rock formation and its surface to obtain the surface temperature distribution image of the rock formation. These images can reflect the differences in heat conduction inside the rock formation because the cracks, weak planes, and damages inside the rock formation will directly affect its heat conduction properties.

[0209] Specifically, preprocess the obtained infrared images, including denoising, image enhancement, calibration, etc., to improve the accuracy of subsequent image recognition.

[0210] Specifically, computer vision and image processing technologies, such as threshold segmentation, edge detection, machine learning, or deep learning models, are used to analyze the processed infrared images to identify regions with abnormal temperature distributions in the images. The abnormal regions are usually parts where the temperature is significantly higher or lower than the surrounding areas, which may indicate fractures or damages inside the rock formation.

[0211] Specifically, the above analysis results are converted into infrared recognition results, that is, to determine whether there are abnormal regions in the infrared images. If an abnormal region is recognized, an infrared recognition result indicating the existence of an abnormal region is generated; if no abnormal region is recognized, an infrared recognition result indicating the non-existence of an abnormal region is generated.

[0212] Specifically, when the infrared recognition result shows that there is an abnormal region, the monitoring system preliminarily judges that there may be a risk of overlying rock movement in the virtual coal mine model and generates the tenth evaluation result. This indicates that the temperature distribution of the rock formation has become abnormal, which may indicate internal damage or crack development in the rock formation, and thus lead to overlying rock movement.

[0213] Specifically, when the infrared recognition result shows that there are no abnormal regions in the infrared images, the monitoring system preliminarily judges that the rock formation in the virtual coal mine model is in good stability and there is no risk of overlying rock movement, and generates the eleventh evaluation result. This indicates that the temperature distribution of the rock formation is normal and no obvious signs of damage are found inside the rock formation.

[0214] Specifically, before the start of the experiment and during the experiment, the thermal infrared imager regularly or continuously monitors the temperature distribution on the surface or inside the rock formation to obtain infrared image data.

[0215] Specifically, the collected infrared images are preprocessed, including image enhancement, denoising, calibration, etc., to ensure that the images are clear and facilitate subsequent analysis.

[0216] Specifically, the thermal infrared imager can display the temperature differences in different regions. Through comparative analysis, the temperature change trend and abnormal hot spots are identified. The temperature increase may be related to stress concentration, crack propagation, or rock formation damage inside the rock formation.

[0217] Specifically, features are extracted from the thermal infrared images, such as the position, size, shape, and temperature gradient of the hot spots. These features help to understand the heat distribution and flow characteristics inside the rock formation.

[0218] Specifically, based on information such as historical data, geological conditions, and mining depth, warning thresholds for temperature changes are set. When the temperature change in a certain region exceeds the preset threshold, or new abnormal hot spots appear, a warning should be triggered to indicate the possible risk of rock formation instability or dynamic disasters.

[0219] Specifically, a time series analysis of temperature changes was performed to identify long-term trends and short-term fluctuations in temperature changes. Long-term rising trends may indicate continued destruction of the rock formation, while short-term fluctuations may be related to local stress release.

[0220] Specifically, the thermal infrared imaging data is combined with data from other monitoring technologies (such as stress, displacement, microseismic data, etc.) to verify each other and improve the accuracy and comprehensiveness of risk assessment. For example, combined with microseismic monitoring data, it is analyzed whether the temperature anomaly area is consistent with the high-incidence area of ​​microseismic events.

[0221] Specifically, thermal infrared image data is used to verify geomechanical models. Temperature changes can reveal the mechanical behavior of rock formations, such as plastic deformation and fracture behavior, which helps optimize and calibrate the model.

[0222] Specifically, the infrared image data and temperature change analysis results are submitted to geological and mining experts for review and further analysis. Combined with the experts' geological knowledge and experience, the rock stability risk is assessed and corresponding prevention and response measures are formulated.

[0223] Specifically, the monitoring frequency and range of the thermal infrared imager are dynamically adjusted according to the temperature changes of the rock formation. For areas with drastic temperature changes, the monitoring density is increased to ensure that changes in the rock formation state are captured in a timely manner.

[0224] Through the above steps, the temperature changes and infrared images detected by the thermal infrared imager can be converted into assessment and early warning signals for the risk of overburden migration. Temperature anomalies and their changing trends are important indicators of rock stability, which can reveal the stress state and destructive behavior inside the rock formation. Combined with multi-sensor data fusion analysis, a more comprehensive and accurate rock stability monitoring system can be constructed to provide a scientific basis for safety decisions in coal mining.

[0225] Specifically, infrared rays are electromagnetic waves between visible light and microwaves, with a wavelength range of 0.77 to 1000 μm and a frequency of 3×10¹¹ to 4×10¹⁴ Hz. From the electromagnetic radiation spectrum, it can be seen that the visible light commonly seen by people occupies a very small part, while infrared rays occupy a quite large part. In scientific research, the wavelength range of 0.77 to 3 μm is called the near-infrared region; the wavelength range of 3 to 30 μm is the mid-infrared region; above 30 μm is called far-infrared (also known as long-wave). In nature, any object with a temperature higher than absolute zero (-273 °C) is an infrared radiation source and has a radiation phenomenon. Infrared non-destructive testing is based on the transfer of heat and heat flow of an object. When there are cracks or other defects inside the object, it will change the heat conduction of the object, causing differences or uneven changes in the surface temperature distribution of the object. By using these differential or uneven change images, the defect location of the object can be visually detected. Detecting the distribution image of the infrared radiation energy formed by the surface temperature of each part of the object with an infrared thermal imager is the thermal image, which is a detection technology that visually shows the integrity and continuity of materials and structures and the discontinuous defects in their joints. It is a non-contact non-destructive testing technology, that is, continuously scanning the object to be measured non-contact up and down and left and right, so as to achieve the detection of relevant indicators. The working principle of the thermal infrared imager is as Figure 13 shown.

[0226] Specifically, during the process of rock mass instability and failure, internal damage and fracture occur in the rock mass, and energy dissipation is generated through the force-heat coupling effect. In this process, the joints and weak surfaces of the rock are often damaged. Microscopically, it is the fracture of the crystal lattice. The fracture of the crystal lattice will cause the electrons inside the molecule to undergo energy level transitions and then generate electromagnetic radiation. The essential reason for the instability and failure of the rock mass is that the microcracks generated by the fracture of the crystal lattice and the breaking of crystal bonds converge into nuclei and expand into macroscopic cracks. Infrared thermal imaging technology can qualitatively and quantitatively characterize the changes in infrared radiation during the stress process of the rock mass. Therefore, an infrared thermal imager is used to detect the whole process of the physical similarity simulation experiment.

[0227] Specifically, this experiment was carried out indoors. An infrared imager was used during the infrared detection process. This instrument has strong adaptability, a small and compact shape, a thermal sensitivity (NETD) < 0.05 °C, an infrared resolution of 640×480 pixels, and a 5-megapixel industrial performance digital camera, which can provide high-definition experimental image quality; advanced display output options, and detailed information can be obtained by video streaming (USB and HDMI) to a PC or a high-resolution monitor. This device is mainly composed of an infrared thermal imager, an image processing system, a data transmission line, etc., and can detect the distribution changes of the surface temperature field of an object caused by various reasons and generate infrared image files in a specific format.

[0228] Specifically, one day before the experiment starts, all the detection instruments related to the experiment need to be placed in the laboratory and installed and debugged. The clocks of each device should be set to the same time to facilitate the time comparison of various data during the later experiment data processing. Keep the detection instruments, physical model, and laboratory environment temperature consistent to ensure that the temperature measured by the infrared thermal imager is the temperature increment caused by the damage and fracture of the physical finite element plate. Before the test, the infrared imager needs to be initialized and parameters such as emissivity, reflected temperature, ambient temperature, and relative humidity are set. During the experiment, the emissivity of the model rock mass in the simulation experiment is set to 0.92, the background temperature is 18.9 °C, and the detection angle of view is 14.6°×18.2°. During the physical model experiment, the infrared imager is placed 2000 mm directly in front of the physical model, and an imaging area of 500×500 mm2 can be detected. While the infrared imager is working, a camera is also placed in front of the physical model to record the entire experiment process.

[0229] Specifically, in the above thermal infrared radiation monitoring, an infrared thermal imager is used to monitor the temperature change of the rock formation surface in real time. By analyzing the temperature field, the location of the rock formation damage and fracture is located, and the overlying rock failure is warned.

[0230] In the specific implementation process, in the case of the above relevant parameters including the above acoustic emission signals, a risk assessment is carried out on all the above relevant parameters to obtain multiple assessment results, which can be achieved through the following steps: extract the characteristic data of the above acoustic emission signals, where the above characteristic data includes acoustic emission frequency and / or energy rate; in the case where the above acoustic emission frequency is greater than or equal to the preset frequency threshold, and / or, in the case where the above energy rate is greater than or equal to the preset energy rate threshold, preliminarily determine that the above virtual coal mine has a risk of overlying rock movement, and obtain the twelfth assessment result; in the case where the above acoustic emission frequency is less than the above preset frequency threshold and the above energy rate is less than the above preset energy rate threshold, preliminarily determine that the above virtual coal mine has no risk of overlying rock movement, and obtain the thirteenth assessment result.

[0231] In this scheme, the real-time monitoring of acoustic emission signals can immediately capture the occurrence of rock formation damage or fracture. The setting of the preset frequency threshold and energy rate threshold enables the monitoring results of acoustic emission signals to be converted into specific risk assessment levels, realizing the quantification and standardization of risk assessment, and improving the accuracy and reliability of early warning.

[0232] Specifically, during the experiment, an acoustic emission monitoring system is used to collect the acoustic emission signals generated inside the rock formation in real time. These signals contain the dynamic information of rock formation damage and fracture. Through signal processing techniques such as spectral analysis, wavelet transform, and pattern recognition, the characteristic data of acoustic emission signals are extracted, including acoustic emission frequency and energy rate. The acoustic emission frequency reflects the frequency of rock formation damage events, while the energy rate represents the magnitude of the energy released by damage events.

[0233] Specifically, preset thresholds for acoustic emission frequency and energy rate are set, and these thresholds are usually determined based on historical data, physical properties of rock formations, mining depth, and geological conditions. The preset frequency threshold and preset energy rate threshold respectively represent the critical values when the internal damage events of the rock formation reach a certain frequency or the energy release reaches a certain scale. Exceeding these thresholds may indicate a decline in the stability of the rock formation.

[0234] Specifically, if the acoustic emission frequency monitored in the experiment reaches or exceeds the preset frequency threshold, and / or the acoustic emission energy rate reaches or exceeds the preset energy rate threshold, the monitoring system will preliminarily judge that there may be a risk of overlying rock movement in the virtual coal mine model and generate the twelfth evaluation result. This indicates that the internal damage events of the rock formation are frequent or the energy release is strong, which may indicate that the rock formation is about to occur or has already occurred damage, thus leading to overlying rock movement.

[0235] Specifically, on the contrary, if the acoustic emission frequency is lower than the preset frequency threshold and the acoustic emission energy rate is lower than the preset energy rate threshold, the monitoring system preliminarily judges that the rock formation in the virtual coal mine model is in good stability and there is no risk of overlying rock movement, and generates the thirteenth evaluation result. This means that there are fewer internal damage events of the rock formation and the energy release is weak, and the rock formation is in a stable state.

[0236] Based on the analysis of the previous experimental data and rock formation characteristics, the preset threshold for acoustic emission frequency is set at 100 Hz, and the preset threshold for acoustic emission energy rate is set at 50 dB as the standard for preliminarily judging the risk of overlying rock movement. During the experiment, the acoustic emission monitoring sensors are buried in the key parts of the rock formation to collect acoustic emission signals in real time. The acoustic emission data, including frequency and energy rate information, are automatically read every 10 minutes. In a certain monitoring, the recorded acoustic emission frequency is 120 Hz and the energy rate is 60 dB, both of which exceed the preset thresholds. Immediately, it is judged that there is a risk of overlying rock movement and the twelfth evaluation result is generated. At another monitoring point, the acoustic emission frequency is 80 Hz and the energy rate is 45 dB, both of which are lower than the preset thresholds. It is judged that the rock formation is in good stability and there is no risk of overlying rock movement, and the thirteenth evaluation result is generated.

[0237] Specifically, before the experiment or mining, the acoustic emission monitor is initialized and monitoring parameters such as sensitivity and sampling frequency are set. The acoustic emission activities in the rock formation are monitored regularly or continuously to collect acoustic emission signal data. The collected acoustic emission signals are preprocessed, including filtering, removing background noise and false signals, to ensure the data quality.

[0238] Specifically, the characteristic parameters of the signal, such as signal amplitude, frequency, duration, etc., are analyzed to identify acoustic emission events. Using threshold detection, a threshold for signal amplitude is set, and signals exceeding this threshold are considered acoustic emission events.

[0239] Specifically, analyze the changing trends of the frequency and energy of acoustic emission events over time. Identify the patterns of acoustic emission activities, such as the distribution, clustering, and periodic characteristics of events. Be vigilant about sudden changes or abnormal increases in patterns, which may indicate the acceleration of strata failure or stress concentration.

[0240] Specifically, calculate the statistical parameters of acoustic emission activities, such as the total number of events, the highest energy event, the average energy, etc. Conduct a risk assessment on the statistical parameters and set thresholds. For example, a sharp increase in the total number of events or the occurrence of high-energy events may indicate an impending large-scale failure of the strata. Combine the stress state of the strata, the mining depth, and other monitoring data for multi-parameter correlation analysis to improve the accuracy of risk assessment.

[0241] Specifically, analyze the waveform characteristics of acoustic emission signals, such as the rise time, decay time, etc. These characteristics can reflect the mechanism and type of strata failure. Detect the spectral characteristics of the signals, analyze the proportion of different frequency components in the signals, and identify the patterns and scope of strata failure.

[0242] Specifically, establish a real-time warning system. When the acoustic emission parameters exceed the preset thresholds, immediately trigger a warning to indicate the possible risk of strata failure or dynamic disasters. The warning should include detailed information about the event, such as time, location, energy magnitude, etc., for quick positioning and response.

[0243] Specifically, conduct a fusion analysis of the acoustic emission monitoring data with the data of other monitoring technologies, such as stress measurement, displacement monitoring, microseismic monitoring, etc., to obtain more comprehensive information about the strata state. Comprehensive analysis helps to mutually verify the monitoring results and enhance the reliability of risk assessment.

[0244] Specifically, input the acoustic emission data into a geomechanics model to predict the stress distribution and failure behavior of the strata. The model prediction results can serve as a theoretical basis for risk assessment, guiding the adjustment of mining plans and the deployment of safety measures.

[0245] Specifically, submit the analysis results of acoustic emission signals to geological and mining engineering experts for review. The experts conduct manual analysis in combination with on-site conditions, mining plans, and historical data to provide professional advice for risk assessment. According to the experts' advice and the risk assessment results, adjust the mining plan and take preventive measures to reduce the risk of dynamic disasters.

[0246] Through the above steps, the acoustic emission signals detected by the acoustic emission monitor can be transformed into an assessment and warning of the risk of overlying strata movement, helping decision-makers take timely measures to prevent potential strata failure and dynamic disasters, and ensuring the safe progress of coal mine mining. Acoustic emission signals are a direct reflection of the internal stress state and failure activities of the strata, and their analysis and utilization have important value in the early warning of mine geological disasters.

[0247] Specifically, during the movement of overlying strata, the increase in the degree of strata failure is accompanied by corresponding characteristics of acoustic emission signals. Moreover, when failure is approaching, the types and intensities of information increase sharply. To comprehensively reflect the acoustic emission characteristics of strata failure, appropriate parameters must be selected. The characteristic parameters of acoustic emission signals include total events, large events, energy rate, spectrum, waveform, and time difference between multi-channel signals. In this experiment, the variation characteristics of total events and energy rate of rock mass with the advancement are mainly analyzed. The total events are the cumulative total number of acoustic emission events detected by the instrument per unit time, reflecting the acoustic emission frequency, which is an important indicator of the occurrence of failure in rock mass materials; the energy rate is the relative cumulative value of the acoustic emission energy detected by the instrument per unit time, reflecting the acoustic emission energy, which is the total number of acoustic emission events detected by the instrument per unit time, reflecting the acoustic emission frequency, which is an important indicator of the occurrence of failure in rock mass materials.

[0248] Specifically, in this physical similarity simulation experiment, acoustic emission is used to monitor the magnitude of the energy released and the total number of events occurring during the failure of overlying strata. Among the characteristics of the acoustic emission waveforms monitored by the acoustic emission monitor, the x-axis represents time (unit: μs), and the y-axis represents amplitude (unit: dB). Before the experiment, an amplitude threshold needs to be set so that the acoustic emission signal just appears, that is, the lowest amplitude. During the experiment, the instrument only records the acoustic emission events with amplitudes greater than this lowest amplitude. The area of the graph formed by the amplitude and time greater than the lowest amplitude within the monitored time period is the energy for this period. The real-time monitoring of acoustic emission in the laboratory is as Figure 14 shown.

[0249] Specifically, by burying 2 acoustic emission sensors above the coal seam and 1 acoustic emission sensor below the coal seam, the dynamic damage signals of the coal seam are monitored, numbered counterclockwise as 1#, 2#, and 3# in sequence. Figure 15 The marked points are the monitoring positions of strike acoustic emission. The specific installation is as Figure 15 shown. The excavation process is completed through stress similarity and three-dimensional similarity, with gradual mining. The 45 working face advances along the strike, with each advance of 4 cm and an excavation height of 3 cm, and a total of 60 excavation steps. The 23 working face advances along the strike, with each advance of 4.8 cm and an excavation height of 4 cm, and a total of 75 excavation steps. During the process of advancing a unit length each time, acoustic emission monitoring is synchronously carried out to study the movement law of strata under the condition of mining disturbance. The experimental excavation method is layered and gradual excavation, which helps to accurately simulate the deformation and failure characteristics of strata under mining disturbance under real in-situ stress conditions.

[0250] Specifically, in the above acoustic emission signal monitoring, the acoustic emission monitor is used to capture the acoustic emission signals during the strata failure process in real time, analyze the signal characteristic parameters (such as total events and energy rate), and evaluate the stability of the rock mass.

[0251] Data fusion is performed based on all the above evaluation results to obtain a comprehensive analysis result, including the following fusion methods:

[0252] Weighted average fusion principle: Different weights are given to the risk assessment results of each sensor. These weights can be determined based on the accuracy of the sensor, the reliability of historical data, or expert knowledge. The final risk assessment is the average of all weighted results. Analysis: Evaluate the criticality of each monitoring technology for the risk of overburden movement, and give higher weights to the data of sensors with high precision or high reliability, so as to obtain a more accurate comprehensive evaluation result.

[0253] Bayesian network fusion principle: Based on Bayesian probability theory, a network is constructed to represent the conditional probability relationship between different sensor data, so as to conduct risk assessment under uncertainty. Analysis: Considering the uncertainty of data and the interdependence between sensors, the Bayesian network can update and refine the risk assessment, and can provide effective risk prediction even in the case of partial data loss.

[0254] Information theory fusion principle: Use information theory methods such as information entropy and mutual information to analyze the contribution degree of sensor data to risk assessment, and conduct data fusion according to the contribution degree. Analysis: Evaluate the "information value" of each monitoring data for risk assessment, and give priority to fusing data with higher information value, which can make more effective use of sensor resources and reduce the processing of redundant information.

[0255] Fuzzy logic fusion principle: Use fuzzy logic to handle the uncertainty of data. After fuzzy processing of sensor data, fuzzy set theory is used for fusion analysis. Analysis: In the assessment of the degree of rock formation damage, fuzzy logic can handle the fuzziness and uncertainty of data and provide a risk assessment closer to the actual situation.

[0256] The above data fusion in step S203 specifically includes the following sub-steps:

[0257] a) S203-1, data preprocessing: Preprocess the obtained subsidence value, displacement, borehole wall image, pressure value, microseismic signal, infrared image, acoustic emission signal, including noise reduction, signal enhancement, image clarity improvement, etc., to ensure data quality and reduce the interference of external factors.

[0258] b) S203-2, parameter standardization: Convert all monitoring data into a unified evaluation standard or scale to facilitate comparison and fusion between different physical quantities. For example, convert the displacement, pressure value, energy of microseismic signal, etc. into a standardized relative change rate or level.

[0259] c) S203-3, Weight Assignment: According to the importance and reliability in the evaluation of rock stratum stability, corresponding weights are set for each monitoring parameter. For example, the frequency and energy of microseismic events may be assigned higher weights because they directly reflect the stress release and fracture behavior within the rock stratum.

[0260] d) S203-4, Selection of Data Fusion Method: According to the characteristics of the monitoring data and the fusion requirements, a suitable data fusion method is selected. In this solution, the fusion method includes at least one or more of weighted average, Bayesian fusion, information entropy fusion, and fuzzy processing fusion.

[0261] i. Weighted Average: Multiply the standardized parameter values by their corresponding weights, and then calculate the weighted average value as part of the comprehensive analysis result.

[0262] ii. Bayesian Fusion: Based on the prior probability and posterior probability, calculate the probability values of the contribution of each monitoring parameter to the rock stratum stability through the Bayesian formula, and consider all parameters comprehensively to obtain the final risk assessment result.

[0263] iii. Information Entropy Fusion: Calculate the uncertainty (information entropy) of each monitoring parameter, use the information entropy as the weight, and perform weighted fusion of the parameter values to improve the reliability of decision-making.

[0264] iv. Fuzzy Processing Fusion: Adopt the fuzzy logic method, regard the risk assessment results of each monitoring parameter as fuzzy sets, and fuse the data through the operations of fuzzy sets, which is especially suitable for processing monitoring data with high uncertainty.

[0265] e) S203-5, Generation of Comprehensive Analysis Result: According to the fusion method selected in S203-4, fuse all the standardized monitoring parameter values to generate a comprehensive analysis result representing the risk of overlying strata movement. This result should include the risk level of rock stratum stability and the specific risk sources.

[0266] f) S203-6, Result Verification and Adjustment: Compare and verify the comprehensive analysis result with the monitoring data of actual rock stratum movement, and adjust the weights and other parameters in the fusion method when necessary to optimize the accuracy of risk assessment.

[0267] In the S203-3 weight assignment step, the determination of weights can be based on factors such as historical data, geological conditions, mining depth, etc., or can be dynamically adjusted through expert evaluation or machine learning methods.

[0268] In the S203-4 data fusion method selection step, considering the non-numerical characteristics of infrared images and borehole wall images, image feature extraction can be performed first to convert the images into numerical features (such as fracture density, proportion of temperature anomaly areas, etc.), and then fusion analysis is carried out.

[0269] In the S203-5 comprehensive analysis result generation step, the fusion result should adopt an intuitive representation form, such as a risk map, a risk level list, etc., to facilitate coal mine management personnel to quickly understand the rock stratum risk distribution.

[0270] In the S203-6 result verification and adjustment step, the verification of the fusion result and the adjustment of the method should be carried out regularly or when significant changes occur in the monitored data during the actual mining process, to ensure that the monitoring system is always in the best state and provide reliable early warning information for safe production.

[0271] The data fusion method described in detail in this solution can make the monitoring system more intelligent and reliable. By integrating data from multiple monitoring means, the movement state and potential risks of overlying strata can be comprehensively evaluated, the timeliness and accuracy of early warning can be improved, which is of great significance for preventing dynamic disasters in coal mine mining.

[0272] In summary, in the physical similarity simulation of overlying strata movement monitoring, the seven monitoring means of total station, dial gauge, borehole camera, microseismic monitor, thermal infrared imager, pressure sensor and acoustic emission monitor are integrated, and the data acquisition is synchronized. According to the working face mining plan, each monitoring means works synchronously at model advancement or at specific time intervals to provide data in the same time period for comprehensive analysis. The spatial layout is coordinated. Vertically, equipment is arranged at different heights above the coal seam according to the model height and rock stratum characteristics to form a three-dimensional monitoring network; horizontally, measuring points and sensors are set according to the model surface and working face conditions to fully cover the horizontal range of the model. The data analysis is comprehensive. Multi-parameter correlation analysis is carried out to explore the internal relationship of overlying strata failure and laws such as force-heat coupling; the movement trends of different means are compared to mutually confirm the results, ensure reliability, effectively warn of the risk of dynamic disasters, and contribute to the safe production of coal mines.

[0273] In summary, the solution of this application has many significant advantages and positive effects. In terms of monitoring comprehensiveness, it integrates multiple monitoring means such as total station, dial gauge, borehole camera, microseismic monitor, thermal infrared imager, pressure sensor and acoustic emission monitor. Compared with the application of traditional single or a few monitoring devices, it can comprehensively obtain overlying strata movement information from different physical quantities, different spatial positions (vertical and horizontal directions) and different rock stratum depths, avoiding monitoring dead angles, greatly enriching the data dimension, and enabling a more accurate grasp of the overlying strata movement law. For example, the total station and the dial gauge respectively monitor the surface and top displacement of the model from different precisions and measurement ranges, the borehole camera penetrates into the overlying strata to observe the crack development, and the microseismic monitor reflects the rock stratum failure situation from the perspective of energy release. The combination of multiple means realizes an all-round three-dimensional monitoring.

[0274] In terms of the accuracy of disaster early warning, by comprehensively analyzing various monitoring data, the risk of potential dynamic disasters can be judged more accurately. The data obtained by various monitoring means are mutually verified and supplemented. For example, the increase in microseismic events combined with the rapid development of fractures observed by borehole television and the sudden change in stress monitored by pressure sensors can detect signs of overlying rock instability in advance. Compared with a single monitoring means, it can provide early warning for coal mine safety production more timely and accurately, effectively reduce the incidence of coal mine accidents, ensure the safety of personnel and equipment, improve the efficiency of coal mine extraction, reduce economic losses and production downtime caused by disasters, and has important economic and safety benefits.

[0275] The embodiment of the present application also provides a device for determining the risk of overlying rock migration. It should be noted that the device for determining the risk of overlying rock migration in the embodiment of the present application can be used to execute the method for determining the risk of overlying rock migration provided by the embodiment of the present application. The device is used to implement the above embodiments and preferred implementation manners, and those that have been described will not be repeated. As used hereinafter, the term "module" can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0276] The following introduces the device for determining the risk of overlying rock migration provided by the embodiment of the present application.

[0277] Figure 16 It is a structural block diagram of a device for determining the risk of overlying rock migration according to an embodiment of the present application. As Figure 16 shown, the device includes:

[0278] An acquisition unit 10, configured to acquire relevant parameters of a virtual coal mine, where the virtual coal mine is a virtual model that simulates the actual mining conditions and rock mass mechanical behavior of a real coal mine, and the relevant parameters include one or more of a subsidence value, a displacement amount, a borehole wall image, a pressure value, a microseismic signal, an infrared image, and an acoustic emission signal. The subsidence value is the vertical movement change amount of the rock formation collected by a dial gauge, the displacement amount is the horizontal movement change amount of the rock formation collected by a total station, the borehole wall image is an image of the rock formation collected by a borehole television, the pressure value is the stress value of the rock formation collected by a pressure sensor, the microseismic signal is the signal of microseismic events of the rock formation collected by a microseismic monitor, the infrared image is an image of the temperature distribution of the rock formation collected by an infrared thermal imager, and the acoustic emission signal is the energy signal of elastic waves released when cracks are generated in the rock formation collected by an acoustic emission sensor;

[0279] An evaluation unit 20 is configured to perform risk assessment on all the above-mentioned relevant parameters to obtain a plurality of evaluation results. Among them, the above-mentioned evaluation results correspond to the above-mentioned relevant parameters one by one, and the above-mentioned evaluation results characterize whether the overlying strata migration risk occurs in the virtual coal mine according to the above-mentioned relevant parameters;

[0280] A comprehensive analysis unit 30 is configured to perform data fusion according to all the above-mentioned evaluation results to obtain a comprehensive analysis result. Among them, the fusion method includes at least one or more of weighted average, Bayesian fusion, information entropy fusion, and fuzzy processing fusion. The above-mentioned comprehensive analysis result is used to determine whether the overlying strata migration risk occurs in the virtual coal mine;

[0281] A generation unit 40 is configured to generate a warning message when the above-mentioned comprehensive analysis result indicates that the overlying strata migration risk occurs in the virtual coal mine.

[0282] In this embodiment, data is obtained through a variety of monitoring means such as dial gauges, total stations, borehole televisions, microseismic monitors, thermal infrared imagers, pressure sensors, and acoustic emission monitors. Compared with the application of traditional single or a few monitoring devices, it can comprehensively obtain overlying strata migration information from different physical quantities, different spatial positions (vertical and horizontal directions), and different strata depths, avoiding monitoring blind spots, greatly enriching the data dimension, and being able to provide an early warning for coal mine safety production more timely and accurately than a single monitoring means.

[0283] In the specific implementation process, when the above-mentioned relevant parameters include the above-mentioned subsidence value, the evaluation unit includes a first acquisition module, a first determination module, and a second determination module. The first acquisition module is configured to acquire the subsidence change amount, where the above-mentioned subsidence change amount is the difference between the initial subsidence value and the current subsidence value. Among them, the above-mentioned initial subsidence value is the subsidence value obtained at the initial moment after the virtual coal mine is built, and the above-mentioned current subsidence value is the subsidence value obtained at the current moment after the virtual coal mine is built; the first determination module is configured to preliminarily determine that the overlying strata migration risk occurs in the virtual coal mine and obtain a first evaluation result when the above-mentioned subsidence change amount is greater than or equal to a preset subsidence threshold; the second determination module is configured to preliminarily determine that the overlying strata migration risk does not occur in the virtual coal mine and obtain a second evaluation result when the above-mentioned subsidence change amount is less than the preset subsidence threshold.

[0284] In this solution, by continuously monitoring the subsidence change amount, the abnormal change of the vertical displacement of the rock stratum can be detected in real time, and potential overlying strata migration risks can be discovered in time. The monitoring of the subsidence change amount takes into account the dynamic changes of the rock stratum during the mining process. By comparing the initial subsidence value and the current subsidence value, the displacement situation of the rock stratum can be reflected more accurately.

[0285] In some embodiments, when the above-mentioned relevant parameters include the above-mentioned displacement amount, the evaluation unit includes a second acquisition module, a third determination module, and a fourth determination module. The second acquisition module is used to acquire a preset displacement threshold; the third determination module is used to preliminarily determine that there is a risk of overlying rock migration in the above-mentioned virtual coal mine and obtain a third evaluation result when the above-mentioned displacement amount is greater than or equal to the above-mentioned preset displacement threshold; the fourth determination module is used to preliminarily determine that there is no risk of overlying rock migration in the above-mentioned virtual coal mine and obtain a fourth evaluation result when the above-mentioned displacement amount is less than the above-mentioned preset displacement threshold.

[0286] In this solution, displacement monitoring can capture the dynamic changes of the rock formation in real time. Once the displacement amount reaches the preset threshold, a preliminary risk assessment result is immediately generated, standardizing the displacement amount change into an index for risk assessment, making the assessment of rock formation stability more quantitative and objective, helping to avoid the deviation of subjective judgment, and improving accuracy and reliability.

[0287] In the specific implementation process, when the above-mentioned relevant parameters include the above-mentioned borehole wall image, the evaluation unit includes a first recognition module, a fifth determination module, and a sixth determination module. The first recognition module is used to perform image recognition on the above-mentioned borehole wall image using computer vision technology to obtain a recognition result, where the above-mentioned recognition result indicates whether there are cracks in the rock formation; the fifth determination module is used to preliminarily determine that there is a risk of overlying rock migration in the above-mentioned virtual coal mine and obtain a fifth evaluation result when the above-mentioned recognition result indicates that there are cracks in the rock formation; the sixth determination module is used to preliminarily determine that there is no risk of overlying rock migration in the above-mentioned virtual coal mine and obtain a sixth evaluation result when the above-mentioned recognition result indicates that there are no cracks in the rock formation.

[0288] In this solution, the real-time monitoring of the borehole wall image and crack recognition can timely detect the changes in the internal structure of the rock formation. Computer vision technology can accurately identify cracks in the borehole wall image. Compared with manual observation, computer vision technology can detect cracks more quickly and accurately.

[0289] In some embodiments, when the above-mentioned relevant parameters include the above-mentioned pressure value, the evaluation unit includes a third acquisition module, a seventh determination module, and an eighth determination module. The third acquisition module is used to acquire a preset pressure threshold; the seventh determination module is used to preliminarily determine that there is a risk of overlying rock migration in the above-mentioned virtual coal mine and obtain a seventh evaluation result when the above-mentioned pressure value is greater than or equal to the above-mentioned preset pressure threshold; the eighth determination module is used to preliminarily determine that there is no risk of overlying rock migration in the above-mentioned virtual coal mine and obtain an eighth evaluation result when the above-mentioned pressure value is less than the above-mentioned preset pressure threshold.

[0290] In this solution, by real-time monitoring the internal pressure of the rock formation, the abnormal changes in the rock formation pressure can be quickly identified, providing immediate information for the early warning of the risk of overlying rock migration and further improving the accuracy.

[0291] In the specific implementation process, when the above relevant parameters include the above microseismic signals, the evaluation unit includes a second recognition module, a construction module, and a processing module. The second recognition module is used to perform feature recognition on the above microseismic signals to obtain the microseismic feature parameters of the above microseismic signals. Among them, the above microseismic feature parameters include one or more of position, frequency, amplitude, and duration; the construction module is used to construct a microseismic recognition model. Among them, the above microseismic recognition model is trained using multiple sets of training data, and each set of training data in the above multiple sets of training data includes historical microseismic signals obtained within a historical time period and the corresponding historical ninth evaluation results of the above historical microseismic signals; the processing module is used to input the above microseismic signals into the above microseismic recognition model to obtain the ninth evaluation result corresponding to the above microseismic signals. Among them, the above ninth evaluation result is used to characterize whether the above virtual coal mine has a risk of overlying rock movement according to the above microseismic signals.

[0292] In this solution, the microseismic recognition model obtained through training can accurately predict the risk of overlying rock movement based on the characteristic parameters of microseismic signals. Compared with the traditional simple judgment based on thresholds, model prediction can consider the complexity and multi-dimensional characteristics of microseismic signals, reducing false alarms and missed alarms.

[0293] In some embodiments, when the above relevant parameters include the above infrared images, the evaluation unit includes a third recognition module, a ninth determination module, and a tenth determination module. The third recognition module is used to perform image recognition on the above infrared images to determine whether there is an abnormal area in the above infrared images and obtain an infrared recognition result. Among them, the above abnormal area is a part where the temperature distribution is different from other areas; the ninth determination module is used to preliminarily determine that the above virtual coal mine has a risk of overlying rock movement and obtain a tenth evaluation result when the above infrared recognition result indicates that there is the above abnormal area in the above infrared images; the tenth determination module is used to preliminarily determine that the above virtual coal mine has no risk of overlying rock movement and obtain an eleventh evaluation result when the above infrared recognition result indicates that there is no the above abnormal area in the above infrared images.

[0294] In this solution, by using an infrared thermal imager to obtain the temperature distribution image of the rock formation in real time, the temperature abnormal area can be quickly identified, providing real-time temperature information for the early warning of the risk of overlying rock movement. The infrared thermal imager forms an image by detecting the infrared radiation emitted by an object and does not need to be in direct contact with the rock formation, avoiding the disturbance of the monitoring equipment to the rock formation and ensuring the accuracy and reliability of the monitoring results.

[0295] In the specific implementation process, when the above-mentioned relevant parameters include the above-mentioned acoustic emission signal, the evaluation unit includes an extraction module, an eleventh determination module, and a twelfth determination module. The extraction module is used to extract the characteristic data of the above-mentioned acoustic emission signal, where the above-mentioned characteristic data includes acoustic emission frequency and / or energy rate; the eleventh determination module is used to preliminarily determine that there is a risk of overlying rock movement in the above-mentioned virtual coal mine and obtain a twelfth evaluation result when the above-mentioned acoustic emission frequency is greater than or equal to a preset frequency threshold, and / or when the above-mentioned energy rate is greater than or equal to a preset energy rate threshold; the twelfth determination module is used to preliminarily determine that there is no risk of overlying rock movement in the above-mentioned virtual coal mine and obtain a thirteenth evaluation result when the above-mentioned acoustic emission frequency is less than the above-mentioned preset frequency threshold and the above-mentioned energy rate is less than the above-mentioned preset energy rate threshold.

[0296] In this solution, the real-time monitoring of the acoustic emission signal can immediately capture the occurrence of rock formation damage or rupture. The setting of the preset frequency threshold and energy rate threshold enables the monitoring result of the acoustic emission signal to be converted into a specific risk assessment level, realizing the quantification and standardization of risk assessment and improving the accuracy and reliability of early warning.

[0297] The above-mentioned device for determining the risk of overlying rock movement includes a processor and a memory. The above-mentioned acquisition unit, evaluation unit, comprehensive analysis unit, generation unit, etc. are all stored in the memory as program units, and the corresponding functions are realized by the processor executing the above-mentioned program units stored in the memory. The above-mentioned modules are all located in the same processor; or, the above-mentioned each module is located in different processors in any combination form.

[0298] The processor contains a kernel, and the kernel retrieves the corresponding program unit from the memory. One or more kernels can be set, and by adjusting the kernel parameters, the problem that only a single monitoring device is used to study the overlying rock risk in the prior art and the dynamic disaster risk warning cannot be accurately carried out can be solved.

[0299] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM), and the memory includes at least one storage chip.

[0300] The embodiment of the present invention provides a computer-readable storage medium. The above-mentioned computer-readable storage medium includes a stored program, where when the above-mentioned program runs, it controls the device where the above-mentioned computer-readable storage medium is located to execute the above-mentioned method for determining the risk of overlying rock movement.

[0301] The embodiment of the present invention provides a processor. The above-mentioned processor is used to run a program, where when the above-mentioned program runs, it executes the above-mentioned method for determining the risk of overlying rock movement.

[0302] An embodiment of the present invention provides a device, which includes a processor, a memory, and a program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps of at least a method for determining the risk of overlying rock migration. The device herein can be a server, a PC, a PAD, a mobile phone, etc.

[0303] A computer program product includes a non-volatile computer-readable storage medium. The non-volatile computer-readable storage medium stores a computer program. When the computer program is executed by a processor, it implements the steps of the method for determining the risk of overlying rock migration in each embodiment of the present application.

[0304] The present application also provides a risk monitoring system for overlying rock migration, which includes one or more processors, a memory, and one or more programs. Among them, the one or more programs are stored in the memory and are configured to be executed by the one or more processors. The one or more programs include those for executing any one of the above methods for determining the risk of overlying rock migration.

[0305] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. They can be implemented by program codes executable by the computing device. Thus, they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described herein can be executed in a different order, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module for implementation. In this way, the present invention is not limited to any specific combination of hardware and software.

[0306] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0307] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and combinations of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing device generate means for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0308] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including instruction means that implement the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0309] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operational steps are performed on the computer or other programmable device to produce a computer-implemented process, thereby providing steps for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0310] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0311] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM), and / or non-volatile memory such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of computer-readable media.

[0312] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0313] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0314] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. 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 method for determining the risk of overburden migration, characterized in that: include: Obtain relevant parameters of a virtual coal mine, wherein the virtual coal mine is a virtual model that simulates actual mining conditions and mechanical behavior of rock formations in a real coal mine, and the relevant parameters include one or more of a subsidence value, a displacement, a borehole wall image, a pressure value, a microseismic signal, an infrared image, and an acoustic emission signal, wherein the subsidence value is a vertical movement variation of the rock formation acquired by using a dial indicator, the displacement is a horizontal movement variation of the rock formation acquired by using a total station, the borehole wall image is an image of the rock formation acquired by using a borehole television, the pressure value is a stress value of the rock formation acquired by using a pressure sensor, the microseismic signal is a signal of a microseismic event of the rock formation acquired by using a microseismic monitor, the infrared image is an image of the temperature distribution of the rock formation acquired by using an infrared thermal imager, and the acoustic emission signal is an energy signal of elastic waves released when cracks are generated in the rock formation acquired by using an acoustic emission sensor; Performing risk assessment on all the relevant parameters to obtain multiple assessment results, wherein the assessment results correspond to the relevant parameters one by one, and the assessment results represent whether the virtual coal mine has a risk of overburden migration, preliminarily determined based on the relevant parameters; Perform data fusion according to all the evaluation results to obtain a comprehensive analysis result, wherein the fusion method includes at least one or more of weighted average, Bayesian fusion, information entropy fusion, and fuzzy processing fusion, and the comprehensive analysis result is used to determine whether the virtual coal mine has a risk of overburden migration; When the comprehensive analysis result indicates that the virtual coal mine has a risk of overburden migration, early warning information is generated.

2. The method according to claim 1, characterized in that In the case where the relevant parameters include the sinking value, risk assessment is performed on all the relevant parameters to obtain multiple assessment results, including: Obtaining a subsidence change, wherein the subsidence change is the difference between an initial subsidence value and a current subsidence value, wherein the initial subsidence value is a subsidence value obtained at an initial moment after the virtual coal mine is built, and the current subsidence value is a subsidence value obtained at a current moment after the virtual coal mine is built; When the subsidence change is greater than or equal to a preset subsidence threshold, it is preliminarily determined that the virtual coal mine has a risk of overburden migration, and a first assessment result is obtained; When the subsidence change is less than the preset subsidence threshold, it is preliminarily determined that the virtual coal mine has no risk of overburden migration, and a second assessment result is obtained.

3. The method according to claim 1, characterized in that In the case where the relevant parameters include the displacement, risk assessment is performed on all the relevant parameters to obtain multiple assessment results, including: Obtaining a preset displacement threshold; When the displacement is greater than or equal to the preset displacement threshold, it is preliminarily determined that the virtual coal mine has a risk of overburden migration, and a third assessment result is obtained; When the displacement is less than the preset displacement threshold, it is preliminarily determined that there is no risk of overburden migration in the virtual coal mine, and a fourth assessment result is obtained.

4. The method according to claim 1, characterized in that: In the case where the relevant parameters include the hole wall image, risk assessment is performed on all the relevant parameters to obtain multiple assessment results, including: Using computer vision technology to perform image recognition on the hole wall image to obtain a recognition result, wherein the recognition result indicates whether the rock formation has cracks; When the identification result indicates that the rock strata have cracks, it is preliminarily determined that the virtual coal mine has a risk of overburden migration, and a fifth assessment result is obtained; When the identification result indicates that there are no cracks in the rock strata, it is preliminarily determined that there is no risk of overburden migration in the virtual coal mine, and the sixth assessment result is obtained.

5. The method according to claim 1, characterized in that In the case where the relevant parameters include the pressure value, risk assessment is performed on all the relevant parameters to obtain multiple assessment results, including: Get the preset pressure threshold; When the pressure value is greater than or equal to the preset pressure threshold, it is preliminarily determined that the virtual coal mine has a risk of overburden migration, and a seventh assessment result is obtained; When the pressure value is less than the preset pressure threshold, it is preliminarily determined that there is no risk of overburden migration in the virtual coal mine, and an eighth assessment result is obtained.

6. The method according to claim 1, characterized in that In the case where the relevant parameters include the microseismic signal, risk assessment is performed on all the relevant parameters to obtain multiple assessment results, including: Performing feature recognition on the microseismic signal to obtain microseismic feature parameters of the microseismic signal, wherein the microseismic feature parameters include one or more of position, frequency, amplitude, and duration; Constructing a microseismic identification model, wherein the microseismic identification model is obtained by training using multiple sets of training data, each set of training data in the multiple sets of training data includes historical microseismic signals acquired within a historical time period and historical ninth assessment results corresponding to the historical microseismic signals; The microseismic signal is input into the microseismic identification model to obtain a ninth evaluation result corresponding to the microseismic signal, wherein the ninth evaluation result is used to characterize whether the virtual coal mine has a risk of overburden migration, preliminarily determined based on the microseismic signal.

7. The method according to claim 1, characterized in that In the case where the relevant parameters include the infrared image, risk assessment is performed on all the relevant parameters to obtain multiple assessment results, including: Performing image recognition on the infrared image to determine whether there is an abnormal area in the infrared image, and obtaining an infrared recognition result, wherein the abnormal area is a part where the temperature distribution is different from other areas; When the infrared recognition result indicates that the abnormal area exists in the infrared image, it is preliminarily determined that the virtual coal mine has a risk of overburden migration, and a tenth assessment result is obtained; When the infrared identification result indicates that the abnormal area does not exist in the infrared image, it is preliminarily determined that there is no risk of overburden migration in the virtual coal mine, and the eleventh assessment result is obtained.

8. The method according to claim 1, characterized in that In the case where the relevant parameters include the acoustic emission signal, risk assessment is performed on all the relevant parameters to obtain multiple assessment results, including: Extracting characteristic data of the acoustic emission signal, wherein the characteristic data includes acoustic emission frequency and / or energy rate; When the acoustic emission frequency is greater than or equal to a preset frequency threshold, and / or when the energy rate is greater than or equal to a preset energy rate threshold, it is preliminarily determined that the virtual coal mine has a risk of overburden migration, and a twelfth assessment result is obtained; When the acoustic emission frequency is less than the preset frequency threshold and the energy rate is less than the preset energy rate threshold, it is preliminarily determined that there is no risk of overburden migration in the virtual coal mine, and the thirteenth assessment result is obtained.

9. A device for determining the risk of overburden migration, characterized in that: include: an acquisition unit, for acquiring relevant parameters of a virtual coal mine, wherein the virtual coal mine is a virtual model that simulates actual mining conditions and mechanical behaviors of rock formations in a real coal mine, and the relevant parameters include one or more of a subsidence value, a displacement, a borehole wall image, a pressure value, a microseismic signal, an infrared image, and an acoustic emission signal, wherein the subsidence value is a vertical movement variation of the rock formation acquired by using a dial indicator, the displacement is a horizontal movement variation of the rock formation acquired by using a total station, the borehole wall image is an image of the rock formation acquired by using a borehole television, the pressure value is a stress value of the rock formation acquired by using a pressure sensor, the microseismic signal is a signal of a microseismic event of the rock formation acquired by using a microseismic monitor, the infrared image is an image of the temperature distribution of the rock formation acquired by using an infrared thermal imager, and the acoustic emission signal is an energy signal of elastic waves released when cracks are generated in the rock formation acquired by using an acoustic emission sensor; An evaluation unit, configured to perform risk evaluation on all the relevant parameters to obtain a plurality of evaluation results, wherein the evaluation results correspond to the relevant parameters one by one, and the evaluation results represent whether the virtual coal mine has a risk of overburden migration, preliminarily determined based on the relevant parameters; A comprehensive analysis unit, used for performing data fusion according to all the evaluation results to obtain a comprehensive analysis result, wherein the fusion method includes at least one or more of weighted average, Bayesian fusion, information entropy fusion, and fuzzy processing fusion, and the comprehensive analysis result is used to determine whether the virtual coal mine has a risk of overburden migration; A generating unit is used to generate early warning information when the comprehensive analysis result indicates that the virtual coal mine has a risk of overburden migration.

10. A risk monitoring system for overburden migration, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for determining the risk of overburden migration as described in any one of claims 1 to 8.

Images