A smart mine geological disaster early warning method and device

Abstract

The present invention relates to the field of disaster early warning, and more specifically, to a smart mine geological disaster early warning method and device. The method comprises: collecting vibration data of the mine strata above each support device, screening effective support devices; selecting an effective support device as a reference device, marking all other effective support devices as target devices, calculating the similarity between the target device and the reference device, fitting a linear function of the distance and similarity between all target devices and the reference device, calculating the sum of the distances from all points to the linear function and normalizing it as the effectiveness evaluation of the reference device; taking the product of the pressure change and vibration frequency change of the reference device at two adjacent moments and the effectiveness evaluation as the geological disaster risk coefficient, and issuing a warning indication when the geological disaster risk coefficient exceeds a preset risk threshold. The present invention can improve the efficiency and accuracy of mine monitoring and enhance the reliability of geological disaster early warning.

Description

Technical Field

[0001] The present invention relates to the field of disaster early warning, and more specifically, to a smart mine geological disaster early warning method and device. Background Art

[0002] Mining geological disasters refer to the disaster phenomena caused by large-scale shaft and tunnel excavation, destruction of rock and soil structure, and the resulting significant changes in the geological environment, hydrogeological conditions and natural ecosystem of the mining area during the mining process.

[0003] As a key load-bearing component of the mining engineering system, the location of the support equipment is of special importance. It not only undertakes the structural function of stabilizing the surrounding rock, but also becomes a front-end monitoring node for sensing dynamic changes in the stratum due to its direct contact with the rock stratum.

[0004] Existing mine safety monitoring systems integrate high-precision sensor arrays (including pressure sensors, displacement sensors, and vibration sensors) on support equipment to enable real-time dynamic monitoring of the surrounding rock's stress state, deformation characteristics, and dynamic response. These sensors convert geomechanical parameters such as rock formation pressure changes, structural deformation, and vibration waves into standard electrical output signals. After processing by the data acquisition system, these signals form a continuous time-series monitoring data set. For example, when ground subsidence occurs in a mining area, sensors on support equipment can accurately record key parameters such as roof pressure gradient changes and support body displacement rates. This real-time monitoring data, analyzed and processed by specialized analysis software, provides a quantitative, scientific basis for disaster warning decisions.

[0005] However, during the mining process, ground subsidence may occur in different areas. Since the vibration signal will attenuate, reflect and refract during the propagation process, it is difficult for a single monitoring device to fully and accurately obtain the vibration information of the mine stratum, and it is impossible to effectively judge the monitoring effect of each support equipment on the stratum changes. Summary of the Invention

[0006] In order to solve the above-mentioned technical problem of being unable to effectively judge the monitoring effect of each support equipment on the formation changes, the present invention provides solutions in the following aspects.

[0007] In a first aspect, a smart mine geological disaster early warning method includes:

[0008] Collecting vibration data of the mine strata above each support device and screening effective support devices based on the vibration data; each effective support device corresponds to a three-dimensional data set, which includes a correlation index, delay time, and effective monitoring segment between the effective support device and other support devices;

[0009] One of the valid support devices is selected as the reference device, and all other valid support devices are marked as target devices. The similarity between each target device and the reference device is calculated based on the three-dimensional data set. A straight line function is fitted to the distance between all target devices and the reference device and the similarity. The sum of the distances from all points to the straight line function is calculated and normalized as the effectiveness evaluation of the reference device.

[0010] The product of the pressure change and vibration frequency change of the reference equipment at two adjacent moments and the effectiveness evaluation of the reference equipment is used as the geological disaster risk coefficient. When the geological disaster risk coefficient is greater than the preset risk threshold, an early warning indication is issued.

[0011] The present invention firstly screens effective support equipment to accurately identify the position of support equipment that is more sensitive and effective in monitoring stratum vibration conditions, and further comprehensively considers various relationship factors between effective support equipment. The correlation index reflects the degree of mutual influence between equipment when monitoring stratum vibration, the delay time reflects the time difference of signal propagation, and the effective monitoring section clarifies the range that the equipment can accurately monitor. Through the comprehensive analysis of these factors, the effectiveness and importance of effective support equipment in the entire monitoring network can be more comprehensively and accurately evaluated. Finally, the dynamic changes of the stratum and the effectiveness of the support equipment are comprehensively considered to calculate the geological disaster risk coefficient, which can more accurately reflect the possibility of geological disasters, avoid the misjudgment or omission caused by single-factor early warning, and improve the accuracy and reliability of the early warning.

[0012] Preferably, screening out effective support equipment based on vibration data includes:

[0013] Constructing a traversal range parameter, wherein the traversal range parameter includes a delay time for vibration data transmission and an effective monitoring section, calculating a Pearson correlation coefficient between two adjacent support devices based on the traversal range parameter, using the Pearson correlation coefficient and a preset time adjustment factor as a traversal constraint value between the two support devices, using the Pearson correlation coefficient corresponding to the maximum traversal constraint value as a correlation index between the two support devices, and the corresponding traversal range parameter is the optimal traversal range parameter;

[0014] When the correlation index and the optimal traversal range both meet the preset screening conditions, the corresponding support equipment is the effective support equipment.

[0015] Constructing traversal range parameters that include the delay time for vibration data transmission and the effective monitoring section can comprehensively consider the effective time range for signal transmission and monitoring. The delay time reflects the time difference between the signal being transmitted from the stratum to the support equipment, and the effective monitoring section clarifies the time interval during which the equipment can accurately monitor stratum vibrations. By traversing and calculating these parameters, the monitoring range that best suits the current support equipment combination can be found, ensuring that the vibration data obtained can accurately reflect the actual changes in the stratum; by screening out effective support equipment, monitoring resources can be concentrated on these key devices, avoiding the monitoring of a large number of ineffective devices, thereby improving monitoring efficiency. Effective support equipment can more accurately obtain key information on stratum vibrations, provide more reliable data support for geological disaster warnings, help to promptly identify potential geological disaster risks, and ensure safe production in mines.

[0016] Preferably, the screening conditions are:

[0017] The support equipment simultaneously meets the conditions that the correlation index is greater than the correlation threshold, the delay time is greater than the delay threshold, and the effective monitoring section is greater than the effective monitoring threshold.

[0018] In mine stratum vibration monitoring, the correlation between devices means that they can complement each other and work together to capture information about stratum vibration. When a device has a monitoring blind spot or signal anomaly, other related devices can provide effective supplementary data. A higher correlation index indicates that the vibration signals detected by these support devices may come from the same or similar geological activity sources. In a complex mining environment, signal transmission may be interfered with by various factors, such as electromagnetic interference and geological structural obstructions. Setting a delay threshold can screen out support equipment with relatively stable signal transmission and less interference, thereby improving the reliability of monitoring data. Stratum vibration is a dynamic process that requires sufficient time to capture its complete characteristics and patterns. A longer effective monitoring section can obtain richer vibration data, which helps to more accurately analyze the activity of the stratum and improve the accuracy of geological disaster warnings.

[0019] Preferably, the calculating of the similarity between each target device and the reference device based on the three-dimensional data set is obtained by calculating the cosine similarity of the three-dimensional data set between the target device and the reference device.

[0020] The correlation index in a three-dimensional data set reflects the degree of signal correlation between devices, the latency reflects the timeliness of signal transmission, and the effective monitoring period indicates the time range within which the device can accurately monitor. Cosine similarity calculations consider all three dimensions simultaneously, rather than focusing solely on a single dimension. This avoids misjudgments due to idiosyncrasies in one dimension (e.g., a device may have a similar correlation index to a reference device but a long latency or a short effective monitoring period).

[0021] Preferably, the fitting of the straight-line function of the distances and the similarities between all target devices and reference devices comprises:

[0022] A mapping function is constructed, whose horizontal coordinate is the actual distance from each target device to the reference device, and whose vertical coordinate is the similarity between the target device and the reference device. Then, a linear function is obtained by fitting using the least squares method.

[0023] By constructing a mapping function with the actual distance from the target device to the reference device as the horizontal axis and the similarity as the vertical axis, and using the least squares method to fit the straight line function, the potential relationship between the distance and similarity between the target device and the reference device can be presented in a quantitative form; based on the straight line function, we can predict the similarity between the target device and the reference device at different distances, thereby judging the reliability of the target device in the monitoring data.

[0024] Preferably, the pressure change is the ratio of the pressure value of the effective support equipment corresponding to the current sampling moment to the pressure value of the effective support equipment at a moment before the current sampling moment.

[0025] Preferably, the vibration frequency change is the ratio of the difference between the vibration frequency value of the reference device corresponding to the current sampling moment and a preset vibration frequency threshold to the vibration frequency threshold.

[0026] Preferably, the pressure change is the ratio of the difference between the pressure value of the reference device at the current sampling moment and the pressure value at the moment before the current sampling moment to the time difference between the current sampling moment and the moment before the current sampling moment of the reference device as the pressure change.

[0027] Preferably, the vibration frequency change is the absolute difference between the vibration frequency value at the current sampling moment and the vibration frequency threshold.

[0028] In the second aspect, a smart mine geological disaster early warning device includes: a processor and a memory, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, any one of the smart mine geological disaster early warning methods is implemented.

[0029] The beneficial effects of the present invention are:

[0030] The present invention first identifies support equipment with practical value. It then calculates the similarity between each target device and a reference device, fitting a linear function of distance and similarity to evaluate the effectiveness of the reference device. The effectiveness evaluation is then combined with the pressure and vibration frequency changes of the reference device at two adjacent moments to calculate the geological hazard risk factor. This comprehensive approach, which considers multiple factors, can more comprehensively reflect the potential risk of geological hazards in mines, enabling the early warning system to more accurately determine the likelihood of a geological disaster. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a method flow chart of steps S1 to S3 in a smart mine geological disaster early warning method in an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0033] Reference Figure 1 A smart mine geological disaster early warning method includes steps S1 to S3, which are as follows:

[0034] S1: Collect vibration data of the mine strata above each support device, and screen out effective support devices based on the vibration data; the effective support device corresponds to a three-dimensional data group, and the three-dimensional data group includes a correlation index, delay time and effective monitoring section between it and other support devices.

[0035] In one embodiment, a vibrating wire sensor is installed atop support equipment (equipment used to support the mine's internal structure) to record real-time vibration data from the mine's strata above. The duration and frequency of the sensor's acquisition are appropriately configured based on specific monitoring requirements. It should be noted that each support device installed in the mine has a unique number assigned based on the order in which it was installed.

[0036] In addition, the pressure and vibration frequency of the support equipment need to be collected in real time.

[0037] All the data collected above are further denoised.

[0038] It should be noted that when an earthquake occurs, the seismic waves first reach the areas near the epicenter. These areas will first experience strong vibrations, forming a vibration center.

[0039] When the vibration data from a particular support device suddenly changes (mutates), and the surrounding support devices also exhibit similar vibration changes, it is suspected that this may be caused by ground settlement (i.e., ground settlement has occurred near the location of the support device). This is because when the ground settles, the vibration propagates outward from the center of the settlement. The farther away, the weaker the vibration and the longer the propagation time. Furthermore, as the distance between two support devices increases, the vibration energy decreases, and the effective monitoring period becomes shorter.

[0040] To more accurately determine whether these vibrations are caused by ground settlement, the correlation index between support devices is calculated to identify the vibration pattern of the support devices. If the correlation index between multiple support devices is high, it can be inferred that the upper stratum where these devices are located may be experiencing a relatively consistent vibration state. This can then be analyzed to determine whether this vibration pattern is consistent with the characteristics of ground settlement.

[0041] Furthermore, if the correlation index between support equipment closer to the vibration center decreases with that farther away, the gradient of this correlation index can be used to roughly estimate the extent of the ground settlement's impact. For example, as the correlation index between support equipment decreases from 0.8 to below 0.3 outward from the vibration center, this indicates that the impact of ground settlement on support equipment is weakening within this distance range, allowing a preliminary assessment of the primary impact area of ​​ground settlement.

[0042] Specifically, the traversal range parameters are first determined: the start and end times of the support equipment vibration data (which constitute a valid monitoring segment), and the delay time of the vibration signal between support equipment (i.e., the propagation time difference). The start and end times define the time window for vibration data analysis, ensuring that the complete vibration signal pattern is captured, while the delay time ensures that the propagation characteristics of the vibration signal are matched. Therefore, by traversing these three parameters, the optimal time window and delay time combination can be found, resulting in the most accurate correlation index.

[0043] Furthermore, for two support devices with adjacent installation positions, the Pearson correlation coefficient between the two support devices installed adjacently is calculated based on the traversal range parameter, the Pearson correlation coefficient and the preset time adjustment factor are used as the traversal constraint value between the two support devices, and the Pearson correlation coefficient corresponding to the maximum traversal constraint value is used as the correlation index between the two support devices, and the corresponding traversal range parameter is the optimal traversal range parameter.

[0044] For example, Support equipment and For example, the effective monitoring section in the traversal range parameter is The delay time is , then the Pearson correlation coefficient between the two devices satisfies the relationship:

[0045]

[0046] Where, For the The first support equipment and the The Pearson correlation coefficient between the support equipment, For the The support equipment starts from From moment to moment A collection of vibration data at a certain moment, For the The support equipment starts from From moment to moment A collection of vibration data at a certain moment.

[0047] Since the vibration source is the same, the vibration changes reaching each support equipment are relatively consistent. The larger the Pearson correlation coefficient calculated above, the greater the From moment to moment moment, Support equipment and The vibration data of each support equipment showed similar changes.

[0048] In addition, considering the delay time This will affect the Pearson coefficient between the two devices. For example, a short delay may enhance the Pearson coefficient, while a long delay may weaken the Pearson coefficient. Therefore, a time adjustment factor is introduced to limit the delay time. To avoid delays Too large will cause the Pearson coefficient to be distorted.

[0049] Then the above time adjustment factor is:

[0050]

[0051] Where, is the time adjustment factor, is the delay time, is the hyperbolic tangent function.

[0052] Then the product of the time adjustment factor and the Pearson correlation coefficient calculated above is used as the first The first support equipment and the The traversal constraint value of each support device, and then traverse 、 、 All possible values ​​of , find the best one that maximizes the traversal constraint value 、 、 The Pearson correlation coefficient corresponding to the maximum traversal constraint value is taken as the first The first support equipment and the The correlation index of each support equipment.

[0053] According to the above operation, the correlation index between all adjacent support devices and the corresponding optimal delay time and effective monitoring section can be calculated.

[0054] Furthermore, screening conditions are set: the correlation index is greater than the correlation threshold, the delay time is greater than the delay threshold, and the effective monitoring section is greater than the effective monitoring threshold. Only when these three conditions are met at the same time, the corresponding support equipment is marked as effective support equipment.

[0055] The gradient of the correlation index between effective support devices can be used to roughly estimate the extent of ground settlement impact. For example, the correlation index between support devices decreases from the center of vibration outward, and this change can be used to preliminarily determine the primary impact area of ​​ground settlement. Each effective support device corresponds to an optimal delay time and an optimal effective monitoring section. These parameters ensure that the complete pattern of the vibration signal is captured and the propagation characteristics of the vibration signal are matched, thereby more accurately analyzing the vibration data.

[0056] The association threshold, delay threshold, and effective monitoring threshold are initially set to 0.7, 3, and 30 seconds, respectively. In practice, they are dynamically adjusted and optimized based on the mine's actual conditions (historical monitoring data, stratum physical properties, and vibration propagation characteristics).

[0057] S2: Select one of the valid support devices as the reference device, mark all other valid support devices as target devices, calculate the similarity between each target device and the reference device based on the three-dimensional data group, fit the distance between all target devices and the reference device and the straight line function of the similarity, calculate the sum of the distances from all points to the straight line function and normalize it as the effectiveness evaluation of the reference device.

[0058] Based on the effective support devices ultimately obtained in step S1, one of the effective support devices is selected as the reference device, and all other effective support devices are marked as target devices. Each target device then corresponds to a three-dimensional data set relative to the reference device, including a correlation index, delay time, and effective monitoring segment. This three-dimensional data set can be used to describe the interactions between the effective support devices and the vibration propagation characteristics. By comparing the three-dimensional data sets of the target device and the reference device, the validity of the vibration data of the reference device can be verified. If the three-dimensional data sets of the two devices are highly similar, it indicates that the vibration data is valid.

[0059] In one embodiment, the similarity between the three-dimensional data sets of the target device and the reference device is obtained by calculating the cosine similarity of the two three-dimensional data sets. The calculation formula of the cosine similarity is prior art and will not be described in detail here.

[0060] It is important to note that the similarity of the 3D data sets decreases as the distance between effective support devices increases. This is because the vibration energy decays with increasing distance, resulting in a lower correlation index, longer time delay, and a smaller effective monitoring segment.

[0061] Therefore, a mapping function is constructed to describe the relationship between the location distance and similarity of each target device to the reference device, so as to intuitively show the quantitative relationship between the two.

[0062] In one embodiment, it is first necessary to determine the location distance between each target device and the reference device, and then construct a mapping function based on these location distances and the similarity between each target device and the reference device. In this mapping function, the horizontal axis represents the location distance, that is, the actual distance between each target device and the reference device; and the vertical axis represents the similarity between the target device and the reference device.

[0063] Furthermore, a least square method or other linear regression methods are used to fit a straight line function to better describe the mapping relationship between location distance and similarity.

[0064] For each target device, the distance between its actual location distance and the similarity value predicted by the straight line function is calculated (i.e., the residual error). These distances are then summed across all devices to obtain a sum that reflects the overall deviation between the actual data and the fitted straight line.

[0065] To eliminate possible scale differences between different data, the sum of the calculated distances is further normalized. Finally, the sum of the normalized distances is used as an indicator for evaluating the validity of the reference device vibration data. A small value indicates that the actual data is close to the fitted line, indicating that the device's vibration data has high validity. Otherwise, the validity is low.

[0066] Then, according to the above operation, the effectiveness evaluation index of all effective equipment vibration data can be obtained.

[0067] S3: The product of the pressure change and vibration frequency change of the reference equipment at two adjacent moments and the effectiveness evaluation of the reference equipment is used as the geological disaster risk coefficient. When the geological disaster risk coefficient is greater than the preset risk threshold, an early warning indication is issued.

[0068] Because geological disasters are a dynamic process, the continuous presence of valid abnormal vibration data indicates a higher risk of subsidence at the reference equipment location. For example, in a mining environment, if support equipment in a certain area continuously detects vibration frequencies exceeding a threshold, and this data has been verified to be reliable, then the area is likely experiencing geological structural instability, such as subsidence or sliding, and prompt action is needed to address it.

[0069] In one embodiment, starting from the first sampling moment, the pressure and vibration frequency values ​​of the effective support equipment are collected. A vibration frequency threshold is then pre-determined based on geological conditions, historical data, and historical experience. For example, for a specific stratum in a specific mining area, based on previous monitoring data and geological research, the vibration frequency should be within a certain range when the stratum is relatively stable. Exceeding this range may pose a risk. The limit of this range is the vibration frequency threshold.

[0070] Furthermore, starting from the second sampling moment, the pressure change and vibration frequency change of the effective support equipment corresponding to the previous moment are calculated, and the product of the pressure change, vibration frequency change and the effectiveness evaluation corresponding to the effective resident equipment is used as the geological hazard risk coefficient of the location of the support equipment at the current moment. Similarly, the geological hazard risk coefficient corresponding to each subsequent sampling moment is calculated.

[0071] Among them, the above-mentioned pressure change is the ratio of the pressure value of the effective support equipment corresponding to the current sampling moment to the pressure value of the effective support equipment at the moment before the current sampling moment; or the ratio of the difference between the pressure value of the effective support equipment at the current sampling moment and the pressure value at the moment before the current sampling moment to the time difference between the current sampling moment and the time before the current sampling moment of the effective support equipment is used as the above-mentioned pressure change.

[0072] The above-mentioned vibration frequency change is the ratio of the difference between the vibration frequency value of the effective support equipment corresponding to the current sampling moment and the vibration frequency threshold to the vibration frequency threshold; or the absolute difference between the vibration frequency value at the current sampling moment and the vibration frequency threshold is calculated as the above-mentioned vibration frequency change.

[0073] In addition, after obtaining the above-mentioned pressure change and vibration frequency change, they also need to be normalized to the minimum and maximum values.

[0074] Furthermore, data on geological disaster risk factors from historical periods when geological disasters occurred and when no geological disasters occurred are collected, and a reasonable risk threshold is determined through statistical analysis (such as calculating the mean, standard deviation, etc.).

[0075] When the calculated geological hazard risk factor exceeds the aforementioned threshold, the corresponding support equipment location is determined to be at a high geological hazard risk. If this high-risk situation is identified, the early warning system is immediately activated. This warning can include audible and visual alarms (emitting strong audible and visual signals in the mine monitoring room to attract the attention of personnel) and the immediate use of communication software to notify relevant personnel (detailed warning information, including the type of potential disaster and the danger zone, is sent to mine managers and on-site workers), ensuring that relevant personnel receive and respond promptly.

[0076] The smart mine geological disaster early warning device includes a processor and a memory, and the memory stores computer program instructions. When the computer program instructions are executed by the processor, the smart mine geological disaster early warning method of the first aspect of the present invention is implemented.

[0077] The smart mine geological disaster early warning device also includes other components familiar to those skilled in the art, such as a communication bus and a communication interface. Their settings and functions are known in the art, so they will not be described here in detail.

[0078] It should be noted that those skilled in the art may make various modifications and improvements without departing from the scope of the present invention, and these modifications and improvements fall within the scope of protection of the present invention. Therefore, the scope of protection of the patent for this invention shall be based on the appended claims.

Claims

1. A smart mine geological disaster early warning method, characterized in that: include: Collect vibration data of the mine strata above each support equipment and select effective support equipment based on the vibration data; The effective support device corresponds to a three-dimensional data set, which includes a correlation index, delay time, and effective monitoring segment between the effective support device and other support devices; one of the effective support devices is selected as a reference device, and all other effective support devices are marked as target devices. The similarity between each target device and the reference device is calculated based on the three-dimensional data set, and a linear function of the distance between all target devices and the reference device and the similarity is fitted. The sum of the distances of all points to the linear function is calculated and normalized as the effectiveness evaluation of the reference device; The product of the pressure change and vibration frequency change of the reference device at two adjacent moments and the effectiveness evaluation of the reference device is used as the geological disaster risk coefficient. When the geological disaster risk coefficient is greater than the preset risk threshold, an early warning indication is issued; The method of screening out effective support equipment based on vibration data includes: constructing a traversal range parameter, the traversal range parameter including the delay time of vibration data transmission and the effective monitoring section, calculating the Pearson correlation coefficient between two adjacent support equipment based on the traversal range parameter, using the Pearson correlation coefficient and a preset time adjustment factor as the traversal constraint value between the two support equipment, using the Pearson correlation coefficient corresponding to the maximum traversal constraint value as the correlation index between the two support equipment, and the corresponding traversal range parameter is the optimal traversal range parameter; when the correlation index and the optimal traversal range both meet the preset screening conditions, the corresponding support equipment is the effective support equipment.

2. The intelligent mine geological disaster early warning method according to claim 1, characterized in that: The screening condition is: the support equipment simultaneously satisfies the conditions that the correlation index is greater than the correlation threshold, the delay time is greater than the delay threshold, and the effective monitoring section is greater than the effective monitoring threshold.

3. The intelligent mine geological disaster early warning method according to claim 2, characterized in that: The similarity between each target device and the reference device is calculated based on the three-dimensional data set by calculating the cosine similarity of the three-dimensional data set between the target device and the reference device.

4. The intelligent mine geological disaster early warning method according to claim 3 is characterized in that: The method of fitting the straight line function of the distance and the similarity between all target devices and reference devices includes: constructing a mapping function, whose horizontal coordinate is the actual distance from each target device to the reference device, and whose vertical coordinate is the similarity corresponding to the target device and the reference device, and then using the least squares method to fit a straight line function.

5. The intelligent mine geological disaster early warning method according to claim 4 is characterized in that: The pressure change is the ratio of the pressure value of the effective support equipment corresponding to the current sampling moment to the pressure value of the effective support equipment at a moment before the current sampling moment.

6. The intelligent mine geological disaster early warning method according to claim 5, characterized in that: The vibration frequency change is the ratio of the difference between the vibration frequency value of the reference device corresponding to the current sampling moment and the preset vibration frequency threshold to the vibration frequency threshold.

7. The intelligent mine geological disaster early warning method according to claim 4, characterized in that: The pressure change is the ratio of the difference between the pressure value of the reference device at the current sampling moment and the pressure value at the moment before the current sampling moment to the time difference between the current sampling moment and the moment before the current sampling moment of the reference device.

8. The intelligent mine geological disaster early warning method according to claim 7, characterized in that: The vibration frequency change is the absolute difference between the vibration frequency value at the current sampling moment and the vibration frequency threshold.

9. A smart mine geological disaster early warning device, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the smart mine geological disaster early warning method according to any one of claims 1 to 8 is implemented.

Images