Mine monitoring data reconstruction and disaster early warning method and system

Abstract

The present application provides a kind of mine monitoring data reconstruction and disaster early warning method and system, the present application method includes using dynamic time window, read real-time monitoring data;Calculate data missing rate;Based on Lagrange interpolation algorithm and Prophet model short-term prediction algorithm, intelligent reconstruction monitoring data set;Based on monitoring data set, using Prophet model time series data decomposition algorithm, intelligent reconstruction data evolution trend;Using improved anti-speed method, realize the advance warning of potential geological disaster in time dimension.The mine monitoring data reconstruction and disaster early warning system provided by the present application completes the visualization of original monitoring data, reconstructed monitoring data, reconstructed trend data and advance warning result, realizes one-key review of the stability condition of monitoring area by mine personnel anytime and anywhere.

Description

Technical Field

[0001] This invention relates to the field of monitoring and early warning technology for geological disasters in mining engineering, and more specifically, to a method and system for mine monitoring data reconstruction and disaster early warning. Background Technology

[0002] The ever-changing mining environment poses a severe challenge to the stable operation of sensors. Interruptions in sensing data and periodic disturbances caused by the external environment and mining operations seriously interfere with the stability and effectiveness of monitoring data, affecting the accuracy of early warning of disasters. Therefore, it is essential to develop a method and system for mine monitoring data reconstruction and disaster early warning, enabling automated completion of monitoring data, intelligent reconstruction of evolution trends, early warning of disasters, and convenient viewing of related calculation results. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a method and system for mine monitoring data reconstruction and disaster early warning. The technical means employed in this invention are as follows:

[0004] A method for mine monitoring data reconstruction and disaster early warning includes:

[0005] A dynamic time window is used to read real-time monitoring data;

[0006] Calculate the missing data rate;

[0007] Intelligent reconstruction of monitoring datasets based on Lagrange interpolation algorithm and Prophet model short-term prediction algorithm;

[0008] Based on the monitoring dataset, the Prophet model time series data decomposition algorithm is used to intelligently reconstruct the data evolution trend;

[0009] An improved inverse velocity method is used to achieve early warning of potential geological hazards in the time dimension.

[0010] Furthermore, the adoption of a dynamic time window to read real-time monitoring data includes:

[0011] The default dynamic time window length is 24 hours. When the time interval between the warning time and the current time is less than 30 days, the dynamic time window length will automatically change to 1 / 10 of the time interval between the warning time and the current time.

[0012] Furthermore, the calculation of the missing data rate includes:

[0013] The theoretical amount of data to be collected is obtained by dividing the dynamic time window length by the sampling period of the sensing device;

[0014] The data missing rate is obtained by dividing the actual amount of data collected by the theoretical amount of data collected, and a threshold for the data missing rate is set between 20% and 40%.

[0015] Furthermore, the intelligent reconstruction of the monitoring dataset based on the Lagrange interpolation algorithm and the Prophet model short-term prediction algorithm includes:

[0016] Combine the start time of the dynamic time window End time Start time of data collection End time Data collection volume and data sampling interval By deducing the theoretical start time of the monitoring data, With end time As shown in the following formula:

[0017]

[0018] Based on the acquisition interval, a theoretical sampling time vector is developed. Reconstruction;

[0019] When the missing rate is 0%, it means that the monitoring dataset is complete, and no intelligent reconstruction of the monitoring dataset is performed.

[0020] When the number of missing data points is less than the threshold, it indicates that there is little missing data and the existing data can characterize the evolutionary pattern. Therefore, a cubic polynomial is used as the Lagrange interpolation basis function. The data collected within the time window is used as the initial point, and the reconstructed theoretical sampling time vector is used as the interpolation point to complete the monitoring dataset. Reconstruction;

[0021] When the number of missing data values ​​exceeds a threshold, it indicates a large amount of missing data in the monitoring data. Interpolation methods are insufficient to characterize the data's evolution trend. Therefore, data from the first 30 days of the sampling time window are used as learning samples for short-term prediction, and the sampling interval is set accordingly. Theoretical sampling time vector As input, the Prophet model short-term prediction algorithm is used to implement the monitoring dataset. The reconstruction.

[0022] Furthermore, the intelligent reconstruction of data evolution trends based on the monitoring dataset and using the Prophet model time series data decomposition algorithm includes:

[0023] The Prophet model time series data decomposition algorithm uses an additive model, assuming the monitoring dataset... It is a time series composed of a trend term, a periodic term, and an error term, and the specific relationship is shown in the following formula;

[0024]

[0025] in, This is a trend item used to reflect the monitoring dataset. Non-periodic changes; It is a periodic term, consisting of three terms: daily periodic term, weekly periodic term, and monthly periodic term; The error term represents unpredictable fluctuations in the model and follows a Gaussian distribution.

[0026] Based on the above principles, the Prophet model is used to analyze the monitoring dataset. Reconstruct the data to obtain the trend item. .

[0027] Furthermore, the improved inverse velocity method for achieving advanced early warning of potential geological hazards in the time dimension includes:

[0028] Read all reconstruction trend data up to the current moment Assuming the data volume is The velocity vector of the trend data is calculated using the numerical difference method. The calculation formula is as follows:

[0029]

[0030] in, The difference window length is a custom value.

[0031] Calculate the velocity vector average with standard deviation Determine the current Whether the Laida criterion is satisfied is shown in the following formula;

[0032]

[0033] when When kurtosis = 0, it indicates that the current monitoring data does not conform to a normal distribution and is therefore outlier. Thus, its kurtosis needs to be further calculated. With skewness , and when When =1, directly set the kurtosis to 1. and skewness =0;

[0034]

[0035]

[0036] For any index When satisfied ,and , This indicates that the real-time acquired data has been in an abnormal state and extreme values ​​have appeared, suggesting that the rock mass is in an accelerated creep stage. This is the turning point for entering the accelerated creep stage. If the conditions are not met, the rock mass will remain in the stable creep stage, and no disasters or accidents will occur in the short term. At the same time, no further steps will be taken.

[0037] Extracting the velocity vector Chinese index All subsequent data are used to construct the inverse velocity vector by taking the reciprocal. and corresponding time vector Its length is ;

[0038] Considering that the importance of inverse velocity values ​​for disaster early warning gradually decreases over time (i.e., the closer the inverse velocity is to the current time, the higher its importance for early warning), an arithmetic sequence is used to generate a sequence with the first term being 0, and which is then compared with the inverse velocity vector. Equal-length weight vector The tolerance calculation formula is as follows:

[0039]

[0040] Based on the above weights, it is assumed that the evolution of the anti-velocity during the acceleration phase satisfies the linear characterization formula proposed by Fukuzono. ,in Given the potential disaster occurrence time, a weighted linear fitting method with the following formula as the optimization objective is used to calculate the potential disaster occurrence time. To estimate and achieve early warning of disasters;

[0041]

[0042] Considering the disaster forecast results at different times during the creep acceleration phase There are differences, at different times This constitutes the disaster forecast vector. Furthermore, considering the significance of different forecast results, and that more recent forecasts are more reliable, a statistical method for forecast results at different times was designed, as shown in the following formula, ultimately yielding a dynamically updated disaster forecast interval. ;

[0043] .

[0044] This invention also provides a mine monitoring data reconstruction and disaster early warning system based on the above-mentioned mine monitoring data reconstruction and disaster early warning method, comprising:

[0045] Using Python's Flask module as the backend and the APScheduler module to generate a dynamic time window (default 24h), the method for reconstructing mine monitoring data and issuing disaster warnings is automatically executed. UniApp is used as the frontend to visualize the original monitoring data, reconstructed monitoring data, reconstructed trend data, and disaster warning results. When the warning time is less than one month from the current time, the warning will be pushed via SMS and email, and the dynamic time window period will be automatically adjusted to 1 / 10 of the time interval between the warning time and the current time.

[0046] Compared with the prior art, the present invention has the following advantages:

[0047] 1. The mine monitoring data reconstruction and disaster early warning method provided by this invention combines the monitoring data missing rate to carry out intelligent reconstruction of the monitoring dataset through multi-scheme collaboration, thereby ensuring data integrity.

[0048] 2. The mine monitoring data reconstruction and disaster early warning method provided by this invention adopts the Prophet model time series data decomposition algorithm to realize the intelligent extraction of the evolution trend of monitoring data, and provides a more effective data source for early warning of disasters.

[0049] 3. The mine monitoring data reconstruction and disaster early warning method provided by this invention takes into account the correlation between monitoring data collection time and disaster early warning, and improves the traditional inverse velocity method disaster prediction model, thereby realizing early warning of disasters.

[0050] 4. The mine monitoring data reconstruction and disaster early warning system provided by this invention has completed the visualization of original monitoring data, reconstructed monitoring data, reconstructed trend data and early warning results, and enabled mine personnel to view the stability status of the monitoring area anytime and anywhere with one click.

[0051] Based on the above reasons, this invention can be widely promoted in fields such as monitoring and early warning of geological disasters in mining engineering. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is a diagram of the mine monitoring data reconstruction and disaster early warning method and system architecture of the present invention.

[0054] Figure 2The diagram shows the short-term prediction effect of the Prophet model short-term prediction algorithm provided in the embodiment of the present invention.

[0055] Figure 3 The image shows the intelligent reconstruction result of the monitoring dataset provided in this embodiment of the invention.

[0056] Figure 4 The image shows the reconstruction result of the monitoring data evolution trend provided in the embodiments of the present invention.

[0057] Figure 5 This is a screenshot of the mobile APP software monitoring data query page provided in an embodiment of the present invention.

[0058] Figure 6 This is a screenshot of the data processing result query page of the mobile APP software provided in an embodiment of the present invention.

[0059] Figure 7 The displacement monitoring curve is provided for an embodiment of the present invention.

[0060] Figure 8 The evolution curve of GPS real-time monitoring data provided in the embodiments of the present invention.

[0061] Figure 9 This is a diagram showing the results of identifying the accelerated creep stage according to an embodiment of the present invention.

[0062] Figure 10 The diagram shows the results of the reverse velocity calculation provided in the embodiment of the present invention. Detailed Implementation

[0063] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0064] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0065] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0066] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0067] In the description of this invention, it should be understood that the orientation or positional relationship indicated by directional terms such as "front, back, up, down, left, right", "horizontal, vertical, horizontal" and "top, bottom" is generally based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing this invention and simplifying the description. Unless otherwise stated, these directional terms do not indicate or imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on the scope of protection of this invention. The directional terms "inner" and "outer" refer to the inner and outer contours relative to the outline of each component itself.

[0068] For ease of description, spatial relative terms such as "above," "over," "on the upper surface of," "above," etc., are used herein to describe the spatial positional relationship of a device or feature as shown in the figures to other devices or features. It should be understood that spatial relative terms are intended to encompass different orientations in use or operation besides the orientation of the device as described in the figures. For example, if the device in the figures is inverted, a device described as "above" or "above" other devices or structures would subsequently be positioned as "below" or "under" other devices or structures. Thus, the exemplary term "above" can include both "above" and "below." The device may also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatial relative descriptions used herein will be interpreted accordingly.

[0069] Furthermore, it should be noted that the use of terms such as "first" and "second" to define components is merely for the purpose of distinguishing the corresponding components. Unless otherwise stated, the above terms have no special meaning and therefore should not be construed as limiting the scope of protection of this invention.

[0070] Figure 1 shows the system architecture diagram of a mine monitoring data reconstruction and disaster early warning method according to the present invention. The present invention provides a mine monitoring data reconstruction and disaster early warning method, including:

[0071] S1. Use a dynamic time window to read real-time monitoring data;

[0072] S2. Calculate the data missing rate;

[0073] S3. Intelligent reconstruction of monitoring datasets based on Lagrange interpolation algorithm and Prophet model short-term prediction algorithm;

[0074] S4. Based on the monitoring dataset, the Prophet model time series data decomposition algorithm is used to intelligently reconstruct the data evolution trend;

[0075] S5. An improved inverse velocity method is adopted to achieve advanced early warning of potential geological disasters in the time dimension.

[0076] In a specific implementation, as a preferred embodiment of the present invention, the dynamic time window in step S1 is specifically as follows:

[0077] The default dynamic time window length is 24 hours. When the time interval between the warning time and the current time is less than 30 days, the dynamic time window length will automatically change to 1 / 10 of the time interval between the warning time and the current time.

[0078] In a specific implementation, as a preferred embodiment of the present invention, step S2, calculating the data missing rate, includes:

[0079] S21. Divide the sampling period of the sensing device by the length of the dynamic time window to obtain the theoretical amount of data to be collected;

[0080] S22. Divide the actual amount of collected data by the theoretical amount of collected data to obtain the data missing rate, and set the data missing rate threshold between 20% and 40%.

[0081] In a specific implementation, as a preferred embodiment of the present invention, step S3 involves intelligently reconstructing the monitoring dataset based on the Lagrange interpolation algorithm and the Prophet model short-term prediction algorithm, including:

[0082] S31, Combine the start time of the dynamic time window End time Start time of data collection End time Data collection volume and data sampling interval By deducing the theoretical start time of the monitoring data, With end time As shown in the following formula:

[0083]

[0084] S32. Combine the acquisition interval to develop the theoretical sampling time vector. Reconstruction;

[0085] S33. When the missing rate is 0%, it means that the monitoring dataset is complete, and no intelligent reconstruction of the monitoring dataset is performed.

[0086] S34. When the number of missing data points is less than the threshold, it indicates that there is little missing data and the existing data can characterize the evolutionary pattern. Therefore, a cubic polynomial is used as the Lagrange interpolation basis function. The data collected within the time window is used as the initial point, and the reconstructed theoretical sampling time vector is used as the interpolation point to complete the monitoring dataset. Reconstruction;

[0087] S35. When the number of missing data values ​​exceeds the threshold, it indicates that there are a large number of missing monitoring data. Interpolation cannot characterize the evolution trend of the data. Therefore, data from the first 30 days of the sampling time window is used as the learning sample for short-term prediction, and the sampling interval is set accordingly. Theoretical sampling time vector As input, the Prophet model short-term prediction algorithm is used to implement the monitoring dataset. The reconstruction.

[0088] In a specific implementation, as a preferred embodiment of the present invention, step S4 employs the Prophet model time series data decomposition algorithm to intelligently reconstruct the data evolution trend, including:

[0089] The Prophet model time series data decomposition algorithm uses an additive model, assuming the monitoring dataset... It is a time series composed of a trend term, a periodic term, and an error term, and the specific relationship is shown in the following formula;

[0090]

[0091] in, This is a trend item used to reflect the monitoring dataset. Non-periodic changes; It is a periodic term, consisting of three terms: daily periodic term, weekly periodic term, and monthly periodic term; The error term represents unpredictable fluctuations in the model and follows a Gaussian distribution.

[0092] Based on the above principles, the Prophet model is used to analyze the monitoring dataset. Reconstruct the data to obtain the trend item. .

[0093] In a specific implementation, as a preferred embodiment of the present invention, step S5 employs an improved inverse velocity method to achieve advanced early warning of potential geological hazards in the time dimension, including:

[0094] S51. Read all reconstruction trend data up to the current moment. Assuming the data volume is The velocity vector of the trend data is calculated using the numerical difference method. The calculation formula is as follows:

[0095]

[0096] in, The difference window length is a custom value. When the data collection period is long, try to select a relatively large value, which can remove abnormal fluctuations to some extent.

[0097] S52, Calculate the velocity vector average with standard deviation Determine the current Whether the Laida criterion is satisfied is shown in the following formula;

[0098]

[0099] S53, when When kurtosis = 0, it indicates that the current monitoring data does not conform to a normal distribution and is therefore outlier. Thus, its kurtosis needs to be further calculated. With skewness , and when When =1, directly set the kurtosis to 1. and skewness =0;

[0100]

[0101]

[0102] S54, For any index When satisfied ,and , This indicates that the real-time acquired data has been in an abnormal state and extreme values ​​have appeared, suggesting that the rock mass is in an accelerated creep stage. This is the turning point for entering the accelerated creep stage. If the conditions are not met, the rock mass will remain in the stable creep stage, and no disasters or accidents will occur in the short term. At the same time, no further steps will be taken.

[0103] S55, Extracting the velocity vector Chinese index All subsequent data are used to construct the inverse velocity vector by taking the reciprocal. and corresponding time vector Its length is ;

[0104] S56. Considering that the importance of the inverse velocity value for disaster early warning gradually decreases over time, i.e., the closer the inverse velocity is to the current time, the higher its importance for early warning, an arithmetic sequence is used to generate a value with the first term being 0, and which is then compared with the inverse velocity vector. Equal-length weight vector The tolerance calculation formula is as follows:

[0105]

[0106] S57. Based on the above weights, assume that the evolution of the anti-velocity during the acceleration phase satisfies the linear characterization formula proposed by Fukuzono. ,in Given the potential disaster occurrence time, a weighted linear fitting method with the following formula as the optimization objective is used to calculate the potential disaster occurrence time. To estimate and achieve early warning of disasters;

[0107]

[0108] S58. Considering the disaster forecast results at different times during the creep acceleration phase There are differences, at different times This constitutes the disaster forecast vector. Furthermore, considering the significance of different forecast results, and that more recent forecasts are more reliable, a statistical method for forecast results at different times was designed, as shown in the following formula, ultimately yielding a dynamically updated disaster forecast interval. ;

[0109] .

[0110] This invention also provides a mine monitoring data reconstruction and disaster early warning system based on the above-mentioned mine monitoring data reconstruction and disaster early warning method, comprising:

[0111] Using Python's Flask module as the backend and the APScheduler module to generate a dynamic time window (default 24h), the method for reconstructing mine monitoring data and issuing disaster warnings is automatically executed. UniApp is used as the frontend to visualize the original monitoring data, reconstructed monitoring data, reconstructed trend data, and disaster warning results. When the warning time is less than one month from the current time, the warning will be pushed via SMS and email, and the dynamic time window period will be automatically adjusted to 1 / 10 of the time interval between the warning time and the current time.

[0112] Example 1

[0113] In this embodiment, the JTM-V7000 series multi-point displacement meter of JinTuMu is used as an environmental sensing device to carry out the reconstruction of monitoring data and long-term early warning.

[0114] The JTM-V7000 series multi-point displacement gauges from Jintumu were used to monitor surface deformation in open-pit mines in real time. Based on historical monitoring data, the growth pattern, uncertainty index, and flexibility of the Prophet model's short-term prediction algorithm were autonomously adjusted. When the prediction results based on historical data showed a high degree of fit with historical monitoring data, such as... Figure 2 As shown, this can be used as a benchmark to carry out subsequent automatic data reconstruction work;

[0115] By employing a dynamic time window to read real-time monitoring data and inferring the data missing rate, an intelligent reconstruction of the monitoring dataset based on the Lagrange interpolation algorithm and the Prophet model short-term prediction algorithm was completed. The reconstruction results are as follows: Figure 3 As shown;

[0116] With an evolution period of 1 day, the Prophet model time series data decomposition algorithm was used to reconstruct the evolution trend of the reconstructed dataset, resulting in... Figure 4 The monitoring data trend reconstruction results are shown below;

[0117] An improved inverse velocity method was used to analyze the reconstructed evolutionary trend data, and the evolutionary rate curve was obtained as follows: Figure 5 As shown, through steps S52 and S53, no acceleration phase inflection point that meets the conditions was found. At the same time, observing the curve, it can be seen that although the speed fluctuates greatly, after the fluctuation, the speed will remain stable and there is no significant acceleration phenomenon.

[0118] A mine monitoring data reconstruction and disaster early warning system is used to visualize the entire calculation process. Figure 6 The real-time monitoring data from the sensing devices was displayed. Figure 7 The page displays the data processing process after clicking the "View Data" button. Users can click the "Original Data," "Dataset Reconstruction," and "Trend Reconstruction" buttons to view the corresponding monitoring data processing results.

[0119] Example 2

[0120] Since the creep acceleration stage was not detected during the actual field monitoring in Example 1, this example uses the monitoring data of the landslide at Dagushan Open-pit Mine as input to explain the algorithm for disaster prediction effect.

[0121] Figure 8 The evolution curve of GPS real-time monitoring data is shown. Combined with the actual situation of the mine, the monitoring started on May 13, 2017. The landslide disaster actually occurred on May 27, 2018. Converting the date into a numerical index, that is, treating May 13, 2017 as 1, the disaster occurred 373 days later.

[0122] Combining steps S51 to S54 above, the creep acceleration stage is identified. Figure 9 The results of identifying the creep acceleration phase are shown. As can be seen from the figure, there were three creep acceleration phases before the landslide disaster occurred.

[0123] Figure 10The table below shows the reverse velocity curves for each acceleration stage calculated based on step S55. Based on these results, combined with steps S56 and S57, the possible time of landslide disasters is further calculated, as shown in the table below. During the first creep acceleration stage, the slope rock mass begins stress adjustment, with a disaster prediction time of approximately 340-350 days. However, after adjustment, the rock mass returns to equilibrium, and no landslide occurs; therefore, the early warning is ineffective. The second creep acceleration stage occurs closer to the landslide disaster, representing a brief stress adjustment before the disaster. The disaster prediction time is basically consistent with the landslide disaster occurrence time. The final creep acceleration stage occurs before the landslide, with a prediction time of approximately 388 days, later than the actual disaster occurrence time. This is because the deformation rate does not reach infinity before the disaster, and external rainfall causes the disaster to occur earlier. Therefore, this method can provide some basis for early warning in the time dimension of disasters to a certain extent, but it cannot achieve completely accurate early warning.

[0124] Table 1 Disaster Forecast Time

[0125]

[0126] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for reconstructing mine monitoring data and providing disaster early warning, characterized in that, include: A dynamic time window is used to read real-time monitoring data; Calculate the missing data rate; Intelligent reconstruction of monitoring datasets based on Lagrange interpolation algorithm and Prophet model short-term prediction algorithm; Based on the monitoring dataset, the Prophet model time series data decomposition algorithm is used to intelligently reconstruct the data evolution trend, including: The Prophet model time series data decomposition algorithm uses an additive model, assuming the monitoring dataset... It is a time series composed of a trend term, a periodic term, and an error term, and the specific relationship is shown in the following formula; in, This is a trend item used to reflect the monitoring dataset. Non-periodic changes; It is a periodic term, consisting of three terms: daily periodic term, weekly periodic term, and monthly periodic term; The error term represents unpredictable fluctuations in the model and follows a Gaussian distribution. Based on the above principles, the Prophet model is used to analyze the monitoring dataset. Reconstruct the data to obtain the trend item. ; An improved inverse velocity method is used to achieve early warning of potential geological hazards in the time dimension.

2. The method for mine monitoring data reconstruction and disaster early warning according to claim 1, characterized in that, The method of using a dynamic time window to read real-time monitoring data includes: The default dynamic time window length is 24 hours. When the time interval between the warning time and the current time is less than 30 days, the dynamic time window length will automatically change to 1 / 10 of the time interval between the warning time and the current time.

3. The method for mine monitoring data reconstruction and disaster early warning according to claim 1, characterized in that, The calculation of the missing data rate includes: The theoretical amount of data to be collected is obtained by dividing the dynamic time window length by the sampling period of the sensing device; The data missing rate is obtained by dividing the actual amount of data collected by the theoretical amount of data collected, and a threshold for the data missing rate is set between 20% and 40%.

4. The method for mine monitoring data reconstruction and disaster early warning according to claim 1, characterized in that, The short-term prediction algorithm based on Lagrange interpolation and the Prophet model intelligently reconstructs the monitoring dataset, including: Combine the start time of the dynamic time window End time Start time of data collection End time Data collection volume and data sampling interval By deducing the theoretical start time of the monitoring data, With end time As shown in the following formula: Based on the acquisition interval, a theoretical sampling time vector is developed. Reconstruction; When the missing rate is 0%, it means that the monitoring dataset is complete, and no intelligent reconstruction of the monitoring dataset is performed. When the number of missing data points is less than the threshold, it indicates that there is little missing data and the existing data can characterize the evolutionary pattern. Therefore, a cubic polynomial is used as the Lagrange interpolation basis function. The data collected within the time window is used as the initial point, and the reconstructed theoretical sampling time vector is used as the interpolation point to complete the monitoring dataset. Reconstruction; When the number of missing data values ​​exceeds a threshold, it indicates a large amount of missing data in the monitoring data. Interpolation methods are insufficient to characterize the data's evolution trend. Therefore, data from the first 30 days of the sampling time window are used as learning samples for short-term prediction, and the sampling interval is set accordingly. Theoretical sampling time vector As input, the Prophet model short-term prediction algorithm is used to implement the monitoring dataset. The reconstruction.

5. The method for mine monitoring data reconstruction and disaster early warning according to claim 1, characterized in that, The improved inverse velocity method is used to achieve advanced early warning of potential geological hazards in the time dimension, including: Read all reconstruction trend data up to the current moment Assuming the data volume is The velocity vector of the trend data is calculated using the numerical difference method. The calculation formula is as follows: in, The difference window length is a custom value. Calculate the velocity vector average with standard deviation Determine the current Whether the Laida criterion is satisfied is shown in the following formula; when When kurtosis = 0, it indicates that the current monitoring data does not conform to a normal distribution and is therefore outlier. Thus, its kurtosis needs to be further calculated. With skewness , and when When =1, directly set the kurtosis to 1. and skewness =0; For any index When satisfied ,and , This indicates that the real-time acquired data has been in an abnormal state and extreme values ​​have appeared, suggesting that the rock mass is in an accelerated creep stage. This is the turning point for entering the accelerated creep stage. If the conditions are not met, the rock mass will remain in the stable creep stage, and no disasters or accidents will occur in the short term. At the same time, no further steps will be taken. Extracting the velocity vector Chinese index All subsequent data are used to construct the inverse velocity vector by taking the reciprocal. and corresponding time vector Its length is ; Considering that the importance of inverse velocity values ​​for disaster early warning gradually decreases over time (i.e., the closer the inverse velocity is to the current time, the higher its importance for early warning), an arithmetic sequence is used to generate values ​​with the first term being 0, and which are then correlated with the inverse velocity vector. Equal-length weight vectors The tolerance calculation formula is as follows: Based on the above weights, it is assumed that the evolution of the anti-velocity during the acceleration phase satisfies the linear characterization formula proposed by Fukuzono. ,in Given the potential disaster occurrence time, a weighted linear fitting method with the following formula as the optimization objective is used to calculate the potential disaster occurrence time. To estimate and achieve early warning of disasters; Considering the disaster forecast results at different times during the creep acceleration phase There are differences, at different times This constitutes the disaster forecast vector. Furthermore, considering the significance of different forecast results, and that more recent forecasts are more reliable, a statistical method for forecast results at different times was designed, as shown in the following formula, ultimately yielding a dynamically updated disaster forecast interval. ; 。 6. A mine monitoring data reconstruction and disaster early warning system based on the mine monitoring data reconstruction and disaster early warning method according to any one of claims 1-5, characterized in that, include: Using Python's Flask module as the backend and the APScheduler module to generate a dynamic time window (default 24h), the method for reconstructing mine monitoring data and issuing disaster warnings is automatically executed. UniApp is used as the frontend to visualize the original monitoring data, reconstructed monitoring data, reconstructed trend data, and disaster warning results. When the warning time is less than one month from the current time, the warning will be pushed via SMS and email, and the dynamic time window period will be automatically adjusted to 1 / 10 of the time interval between the warning time and the current time.

Images