A metal mine ground pressure monitoring and early warning method and system for deep mining

Abstract

This invention discloses a method and system for monitoring and early warning of ground pressure in deep-mining metal mines. The method includes: dynamically adjusting the layout of monitoring equipment based on initial ground pressure distribution data and mining progress data; collecting stress change information in real time at different mining stages to obtain dynamic trend data of ground pressure distribution; integrating the dynamic trend data of ground pressure distribution and mining progress data using data fusion technology to construct a correlation model between ground pressure changes and mining progress; obtaining ground pressure anomaly distribution data for each area through a comprehensive evaluation of vertical stress, horizontal stress, and change rate based on the correlation model data; adjusting the layout of monitoring equipment based on the ground pressure anomaly distribution data to obtain updated monitoring point data; and obtaining key monitoring strategy data for high-risk areas by optimizing equipment configuration and monitoring frequency. This invention enables accurate monitoring and risk early warning of ground pressure conditions in mining areas, providing strong support for safe mine production.

Description

Technical Field

[0001] This invention belongs to the field of mine monitoring and early warning, and in particular relates to a method and system for monitoring and early warning of ground pressure in deep metal mines. Background Technology

[0002] Deep metal mining is a key area for ensuring national resource security and economic development. Its complex geological environment and high stress conditions pose severe challenges to safe production. Ground pressure disasters, as one of the main risks in deep mining, can lead to tunnel collapse, surrounding rock damage, and even casualties. Therefore, developing efficient ground pressure monitoring and early warning technologies is crucial. Current ground pressure monitoring methods often suffer from insufficient adaptability in complex terrain and deep environments, especially in mountain mines where topographical variations and geological structures make it difficult for a single monitoring scheme to comprehensively cover the variable ground pressure distribution. Furthermore, existing technologies have bottlenecks in data transmission and equipment reliability. Particularly in complex deep environments, data transmission interruptions or decreased equipment accuracy can lead to untimely early warnings, failing to meet the needs of dynamic mining.

[0003] The complexity of terrain and the dynamic nature of deep mining in mountainous areas pose core challenges. The varied terrain and geological structures require monitoring systems to flexibly adapt to stress distribution patterns in different regions. Traditional fixed monitoring networks struggle to dynamically adjust with mining progress, resulting in incomplete data coverage and difficulty in timely capturing ground pressure changes. Consequently, the real-time performance and accuracy of monitoring data are further constrained, especially in deep mines where the complex underground communication environment makes data transmission susceptible to interference, hindering multi-parameter fusion analysis and centralized early warning. These two factors are interconnected: insufficient dynamic adaptability of the monitoring network limits the comprehensiveness of data acquisition, while unstable data transmission weakens comprehensive analysis and early warning capabilities.

[0004] Therefore, how to develop a monitoring network that can dynamically adapt to the complex terrain of mountain mines and the dynamic changes in deep mining, while ensuring the reliable transmission of monitoring data and multi-parameter fusion analysis in complex environments, has become a key issue in deep metal mine ground pressure monitoring and early warning technology. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method and system for monitoring and early warning of ground pressure in deep-mining metal mines. Specifically, a method for monitoring and early warning of ground pressure in deep-mining metal mines includes:

[0006] Data on topographic slope and geological structure features of the mining area are acquired. A weighted analysis method is used to comprehensively evaluate the topographic slope and geological structure features to obtain the priority data for ground pressure monitoring and the preliminary equipment configuration requirements for each area.

[0007] Based on the ground pressure monitoring priority data and preliminary equipment configuration requirements data, regional equipment configuration is carried out, and three-dimensional stress distribution data is obtained;

[0008] Based on the three-dimensional stress distribution data, the variation characteristics of vertical and horizontal stress are extracted to obtain the initial state data of ground pressure distribution in each region;

[0009] Based on the initial state data of ground pressure distribution and the mining progress data, the layout of monitoring equipment is dynamically adjusted, and stress change information at different mining stages is collected in real time to obtain dynamic change trend data of ground pressure distribution.

[0010] Data fusion technology is used to integrate the dynamic trend data of ground pressure distribution and mining progress data to construct a correlation model data between ground pressure changes and mining progress.

[0011] Based on the data from the aforementioned correlation model, the distribution data of ground pressure anomalies in each region are obtained through a comprehensive evaluation of vertical stress, horizontal stress, and rate of change.

[0012] Based on the ground pressure anomaly distribution data, the layout of monitoring equipment is adjusted, updated monitoring point data is obtained, and key monitoring strategy data for high-risk areas is obtained by optimizing equipment configuration and monitoring frequency.

[0013] If the key monitoring strategy data exceeds the preset stress threshold, an early warning response mechanism is triggered to obtain risk warning information data, and the layout of monitoring equipment is dynamically adjusted to obtain the final optimized ground pressure monitoring scheme data.

[0014] Preferably, the process of using a weighted analysis method to comprehensively evaluate the topographic slope data and geological structural feature data to obtain the ground pressure monitoring priority data and preliminary equipment configuration requirements data for each region includes:

[0015] Acquire topographic slope data and geological structure feature data within the mining area. Conduct a comprehensive scan of each area of ​​the mining area through a pre-established data acquisition system to obtain an initial dataset of topographic slope and geological structure distribution.

[0016] Based on the initial dataset of terrain slope and geological structure distribution, a weighted analysis method was used to quantify the terrain slope data and geological structure complexity values ​​to determine the comprehensive evaluation score for each region.

[0017] By comprehensively evaluating the scores, the ground pressure monitoring needs of each region are ranked, and the priority data of ground pressure monitoring in each region is determined. If the comprehensive evaluation score of a certain region is higher than the preset threshold, the priority is marked as high, and the priority ranking result is obtained.

[0018] Preferably, the process of extracting the variation characteristics of vertical and horizontal stress based on the three-dimensional stress distribution data to obtain the initial state data of ground pressure distribution in each region includes:

[0019] By combining ground pressure monitoring data and priority data with equipment configuration schemes, monitoring equipment is deployed in high-priority areas to collect three-dimensional stress distribution data;

[0020] Based on the three-dimensional stress distribution data, the distribution characteristics of vertical stress and horizontal stress are extracted. Data preprocessing techniques are used to denoise and standardize the extracted features to obtain standardized stress feature data.

[0021] For the standardized stress feature data, the support vector machine algorithm is applied to classify the variation characteristics of vertical stress and horizontal stress, determine the degree of anomaly in stress variation in each region, and determine the distribution range of the abnormal region.

[0022] If the classification results show that the stress change in certain areas exceeds the preset threshold, then the three-dimensional stress distribution data of the corresponding areas will be analyzed in depth to obtain the stress change trend in the abnormal areas and obtain the stress change pattern of the high-risk areas.

[0023] Based on the stress change patterns in high-risk areas and combined with priority data, the deployment locations of monitoring equipment are dynamically adjusted to obtain the adjusted equipment configuration scheme and determine the optimized monitoring coverage.

[0024] By optimizing the monitoring coverage, we continuously collect updated three-dimensional stress distribution data, extract new vertical and horizontal stress characteristics, and update the initial state data of ground pressure distribution.

[0025] Preferably, the process of obtaining dynamic trend data of ground pressure distribution includes:

[0026] Using initial ground pressure distribution data and mining progress data, and employing a preset grid division algorithm, the initial layout location of the monitoring equipment is determined, resulting in an initial equipment distribution scheme.

[0027] If the coverage rate of monitoring points in the initial distribution scheme of the equipment is lower than the preset threshold, the equipment positions are adjusted through an iterative optimization algorithm to obtain an optimized monitoring equipment layout.

[0028] Based on the optimized monitoring equipment layout, stress change data for each mining stage are collected in real time to obtain a real-time stress change dataset.

[0029] The real-time stress change dataset is processed using a time series analysis model to extract the dynamic change characteristics of ground pressure distribution and obtain ground pressure distribution change trend data;

[0030] If the stress anomaly value in the ground pressure distribution trend data exceeds the preset threshold, a high-density monitoring point supplement is performed in the anomaly area using a spatial interpolation algorithm to obtain the supplemented monitoring point layout.

[0031] Based on the revised monitoring point layout, stress change data were re-collected, and the real-time stress change dataset was updated.

[0032] The updated real-time stress change dataset was analyzed using a trend analysis model to obtain the final dynamic trend data of ground pressure distribution.

[0033] Preferably, the process of integrating the dynamic trend data of ground pressure distribution and the mining progress data using data fusion technology to construct a correlation model data between ground pressure changes and mining progress includes:

[0034] Data fusion technology is used to preprocess data related to ground pressure distribution and mining progress, extract dynamic changes and trend data, and obtain a preliminary integrated dataset.

[0035] For the initially integrated dataset, the support vector machine algorithm is used to extract features of the potential relationship between ground pressure changes and mining progress, and to determine the set of associated features;

[0036] Based on the aforementioned set of related features, a correlation model between ground pressure changes and mining progress is constructed to obtain the variation patterns and distribution characteristics in the model and to determine the preliminary applicability of the model.

[0037] If the applicability of the initial model is lower than the preset threshold, then the trend data and progress data are integrated a second time to obtain an optimized combination of data features.

[0038] By optimizing the combination of data features, the parameters of the correlation model are readjusted to obtain the updated correlation between ground pressure changes and mining progress.

[0039] Based on the updated correlation model, the distribution and dynamic changes of ground pressure are continuously monitored to obtain real-time data on the changing patterns.

[0040] Based on the real-time change data, the correlation model is dynamically updated to obtain the latest distribution characteristics and trend data, and the final correlation result between ground pressure change and mining progress is determined.

[0041] Preferably, the process of obtaining the distribution data of ground pressure anomalies in each region based on the correlation model data and through a comprehensive evaluation of vertical stress, horizontal stress, and rate of change includes:

[0042] Preprocessing of vertical stress, horizontal stress, and rate of change data yields stress distribution characteristics for each region.

[0043] Based on the stress distribution characteristics, and considering the numerical differences between vertical and horizontal stresses, a comprehensive analysis is conducted to determine whether there is a significant stress imbalance. If the stress difference in a certain area exceeds a preset threshold, then the area is identified as a potential area of ​​ground pressure anomaly.

[0044] Data on potential geopressure anomaly areas are obtained, and combined with the dynamic trend of the rate of change, time series analysis is used to further analyze the stability of stress changes within the anomaly area and obtain the fluctuation characteristics of the rate of change.

[0045] By conducting in-depth analysis of the fluctuation characteristics of the rate of change and combining it with the geographical information of regional division, if the fluctuation amplitude of the rate of change in a certain region is higher than a preset threshold, the region is determined to be a high-risk ground pressure anomaly region.

[0046] Based on the determination of high-risk ground pressure anomaly areas, and in response to the needs of stress distribution and anomaly detection, the support vector machine algorithm is used to classify the anomaly areas and determine the distribution of anomaly levels.

[0047] After obtaining the distribution of anomaly levels, and by adjusting the model parameters, corresponding risk distribution maps are generated for anomaly areas of different levels, thus obtaining the final ground pressure anomaly distribution data.

[0048] Preferably, the process of obtaining key monitoring strategy data for high-risk areas includes:

[0049] Based on the ground pressure anomaly distribution data, the initial distribution of monitoring points is obtained, and data cleaning techniques are used to remove outliers and noise to obtain a preliminarily processed distribution dataset.

[0050] Based on the pre-processed distribution dataset, the regional characteristics of ground pressure anomalies are analyzed, and spatial clustering methods are used to determine the boundaries of high-risk areas, thus obtaining the division results of high-risk areas.

[0051] Based on the results of the high-risk area classification, the layout of the monitoring equipment was adjusted, and the monitoring points were optimized by using preset density thresholds to determine the updated monitoring point data.

[0052] Starting from the updated monitoring point data, analyze the ground pressure anomaly change trend of each point. If the change trend exceeds the preset fluctuation range, increase the monitoring frequency of the corresponding point to obtain the adjusted frequency configuration data.

[0053] Based on the adjusted frequency configuration data and the distribution characteristics of high-risk areas, the equipment configuration scheme is optimized, and the equipment load is balanced through the resource allocation model to obtain the final equipment configuration data.

[0054] Based on the final equipment configuration data, a key monitoring strategy is generated. If the ground pressure anomaly data in a certain area continues to be higher than the preset threshold, monitoring resources are allocated first, and targeted monitoring strategy data is determined.

[0055] By monitoring strategy data, the operating parameters of the monitoring equipment are dynamically updated. If an abnormal change in the distribution of ground pressure is detected, the location layout is recalculated to obtain real-time optimized monitoring scheme data.

[0056] Preferably, the process of obtaining the final optimized ground pressure monitoring scheme data includes:

[0057] Real-time monitoring data is collected and compared with preset stress thresholds to determine whether there are any exceedances, and a preliminary risk assessment result is obtained.

[0058] If the preliminary risk assessment results show that the monitoring data exceeds the stress threshold, the early warning response mechanism will be triggered to generate corresponding risk early warning information data and determine the specific risk level classification.

[0059] Based on risk warning information data and risk level classification, a pre-established decision-making model is used to obtain targeted equipment layout adjustment strategies and output preliminary layout optimization data.

[0060] Based on the preliminary layout optimization data and combined with the historical records of ground pressure monitoring, it is determined whether there are any layout conflicts or redundancies, and the adjusted equipment layout scheme data is obtained.

[0061] If the adjusted equipment layout data meets the preset optimization conditions, the final ground pressure monitoring optimization data will be generated to determine the latest deployment location of the monitoring equipment.

[0062] Based on the final optimized ground pressure monitoring data, the support vector machine algorithm is used to predict and analyze the stress distribution at the monitoring points to obtain the distribution characteristics of potential risk areas.

[0063] Based on the distribution characteristics of potential risk areas, the frequency and scope of monitoring data collection are dynamically updated to obtain continuously optimized ground pressure monitoring strategy data.

[0064] This invention also provides a ground pressure monitoring and early warning system for deep-mining metal mines, comprising:

[0065] The comprehensive evaluation module is used to acquire topographic slope data and geological structure feature data of the mining area, and to comprehensively evaluate the topographic slope data and geological structure feature data using a weighted analysis method to obtain ground pressure monitoring priority data and preliminary equipment configuration requirements data for each area.

[0066] The three-dimensional stress analysis module is used to configure regional equipment and obtain three-dimensional stress distribution data based on the ground pressure monitoring priority data and preliminary equipment configuration requirements data; and to extract the variation characteristics of vertical stress and horizontal stress based on the three-dimensional stress distribution data to obtain the initial state data of ground pressure distribution in each region.

[0067] The dynamic monitoring and adjustment module is used to dynamically adjust the layout of monitoring equipment based on the initial state data of ground pressure distribution and mining progress data, collect stress change information at different mining stages in real time, and obtain dynamic change trend data of ground pressure distribution.

[0068] The correlation model construction module is used to integrate the dynamic change trend data of ground pressure distribution and mining progress data using data fusion technology to construct the correlation model data between ground pressure change and mining progress.

[0069] The anomaly assessment module is used to obtain the distribution data of ground pressure anomalies in each region based on the associated model data through a comprehensive assessment of vertical stress, horizontal stress, and rate of change.

[0070] The monitoring strategy optimization module is used to adjust the layout of monitoring equipment based on the ground pressure anomaly distribution data, obtain updated monitoring point data, and obtain key monitoring strategy data for high-risk areas by optimizing equipment configuration and monitoring frequency.

[0071] The early warning response module is used to trigger an early warning response mechanism if the key monitoring strategy data exceeds a preset stress threshold, obtain risk early warning information data, dynamically adjust the layout of monitoring equipment, and obtain the final ground pressure monitoring optimization scheme data.

[0072] Compared with the prior art, the present invention has the following advantages and technical effects:

[0073] This invention discloses an optimized method for monitoring ground pressure in mining areas. By acquiring topographic slope and geological structure data, a weighted analysis is used to assess the priority of ground pressure monitoring. High-precision equipment is deployed in high-risk areas to obtain three-dimensional stress distribution. The monitoring layout is dynamically adjusted in conjunction with the mining progress, and stress change information is collected in real time to construct a correlation model between ground pressure changes and mining progress. Based on this model, this invention comprehensively assesses the stress state, identifies abnormal distribution areas, optimizes monitoring strategies and equipment configuration, and achieves focused monitoring of high-risk areas. When monitoring data exceeds a preset threshold, an early warning response mechanism is triggered, and the monitoring plan is dynamically adjusted, thereby achieving accurate monitoring and risk warning of the ground pressure state in mining areas, providing strong support for safe mine production. Attached Figure Description

[0074] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0075] Figure 1 This is a schematic diagram of the method flow according to an embodiment of the present invention;

[0076] Figure 2 This is a schematic diagram of the system structure according to an embodiment of the present invention. Detailed Implementation

[0077] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0078] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0079] Example 1

[0080] like Figure 1 As shown, this embodiment provides a method for monitoring and early warning of ground pressure in deep-mining metal mines, including:

[0081] Data on topographic slope and geological structure features of the mining area were obtained. A weighted analysis method was used to comprehensively evaluate the topographic slope data and geological structure features to obtain the priority data for ground pressure monitoring and the preliminary equipment configuration requirements for each area.

[0082] Based on the priority data of ground pressure monitoring and the preliminary equipment configuration requirements, regional equipment configuration is carried out, and three-dimensional stress distribution data is obtained;

[0083] Based on the three-dimensional stress distribution data, the variation characteristics of vertical and horizontal stress are extracted to obtain the initial state data of ground pressure distribution in each region;

[0084] Based on the initial state data of ground pressure distribution and mining progress data, the layout of monitoring equipment is dynamically adjusted, and stress change information at different mining stages is collected in real time to obtain dynamic change trend data of ground pressure distribution.

[0085] Data fusion technology is used to integrate dynamic trend data of ground pressure distribution and mining progress data to construct a correlation model data between ground pressure changes and mining progress.

[0086] Based on the correlation model data, the distribution data of ground pressure anomalies in each region are obtained through a comprehensive evaluation of vertical stress, horizontal stress, and rate of change.

[0087] Based on the ground pressure anomaly distribution data, the layout of monitoring equipment is adjusted, updated monitoring point data is obtained, and key monitoring strategy data for high-risk areas is obtained by optimizing equipment configuration and monitoring frequency.

[0088] If the key monitoring strategy data exceeds the preset stress threshold, the early warning response mechanism is triggered, risk warning information data is obtained, and the layout of monitoring equipment is dynamically adjusted to obtain the final optimized ground pressure monitoring scheme data.

[0089] Furthermore, the process of comprehensively evaluating topographic slope data and geological structural feature data using a weighted analysis method to obtain priority data for ground pressure monitoring and preliminary equipment configuration requirements for each region includes:

[0090] Acquire topographic slope data and geological structure feature data within the mining area. Conduct a comprehensive scan of each area of ​​the mining area through a pre-established data acquisition system to obtain an initial dataset of topographic slope and geological structure distribution.

[0091] Based on the initial dataset of terrain slope and geological structure distribution, a weighted analysis method was used to quantify the terrain slope data and geological structure complexity values ​​to determine the comprehensive evaluation score for each region.

[0092] By comprehensively evaluating the scores, the ground pressure monitoring needs of each region are ranked, and the priority data of ground pressure monitoring in each region is determined. If the comprehensive evaluation score of a certain region is higher than the preset threshold, the priority is marked as high, and the priority ranking result is obtained.

[0093] For example, to acquire slope data for mining areas, a drone equipped with a LiDAR system can perform a high-precision scan of the mining area to generate a 3D terrain model. The scanning resolution can reach 0.1 meters, covering 500 hectares of the mining area, and acquiring slope data ranging from 0° to 45°. The data acquisition system must ensure coverage of all areas of the mining area, especially the slopes, to avoid data gaps. This method can efficiently generate high-precision slope distribution maps, providing a reliable foundation for subsequent analysis.

[0094] In one possible implementation, geological structural feature data can be collected using a combination of ground-penetrating radar (GPR) and drilling verification. GPR scanning can identify faults and folds within 50 meters underground, and the structural complexity can be verified by combining this with drilling data. Assuming the mining area is divided into 10 zones with fault densities ranging from 2 to 10 per square kilometer, and fold complexity categorized as simple, moderate, and complex, this method can accurately describe the distribution of geological structures and improve the accuracy of ground pressure risk assessment.

[0095] Specifically, the weighted analysis method can assign weights to terrain slope and geological structural complexity, such as 40% for slope and 60% for structural complexity. Taking a certain area as an example, with a slope of 20° and a structural complexity of medium, the comprehensive score is calculated to be 80 points, exceeding the preset threshold of 70 points, and thus marked as high priority. This quantitative processing can objectively reflect the regional ground pressure monitoring needs and optimize resource allocation.

[0096] Furthermore, the process of extracting the variation characteristics of vertical and horizontal stress based on three-dimensional stress distribution data to obtain the initial state data of ground pressure distribution in each region includes:

[0097] By combining ground pressure monitoring data and priority data with equipment configuration schemes, monitoring equipment is deployed in high-priority areas to collect three-dimensional stress distribution data;

[0098] Based on three-dimensional stress distribution data, the distribution characteristics of vertical and horizontal stress are extracted. Data preprocessing techniques are used to denoise and standardize the extracted features to obtain standardized stress feature data.

[0099] For the standardized stress feature data, the support vector machine algorithm is applied to classify the variation characteristics of vertical stress and horizontal stress, determine the degree of anomaly in stress variation in each region, and determine the distribution range of the abnormal region.

[0100] If the classification results show that the stress change in certain areas exceeds the preset threshold, then the three-dimensional stress distribution data of the corresponding areas will be analyzed in depth to obtain the stress change trend in the abnormal areas and obtain the stress change pattern of the high-risk areas.

[0101] Based on the stress change patterns in high-risk areas and combined with priority data, the deployment locations of monitoring equipment are dynamically adjusted to obtain the adjusted equipment configuration scheme and determine the optimized monitoring coverage.

[0102] By optimizing the monitoring coverage, we continuously collect updated three-dimensional stress distribution data, extract new vertical and horizontal stress characteristics, and update the initial state data of ground pressure distribution.

[0103] Specifically, in the process of ground pressure monitoring and equipment configuration, the monitoring area is first divided based on priority data. Assuming a mining area is divided into three priority areas, the high priority area is area A, with an area of ​​approximately 5,000 square meters and a ground pressure risk index of 8.5 (out of 10). The data analysis system automatically selects this area to be equipped with high-precision monitoring equipment. Stress sensors with an accuracy of 0.01 MPa are selected, and one sensor is deployed for every 100 square meters, for a total of 50 devices. The data acquisition frequency of the devices is set to once per minute to ensure real-time performance.

[0104] Next, the system automatically generates an equipment configuration scheme, connects the sensor data to the central processing platform, and calculates the three-dimensional stress distribution data through a three-dimensional modeling algorithm (such as finite element analysis). Assuming that the vertical stress at a certain point in area A is 15.2 MPa and the horizontal stress is 10.8 MPa, the data is processed into a three-dimensional stress field map through meshing, with the error controlled within ±0.05 MPa.

[0105] Subsequently, the system used time series analysis algorithms to extract the variation characteristics of vertical and horizontal stress. The analysis showed that the vertical stress fluctuated between 14.8 and 15.5 MPa within 24 hours, while the horizontal stress fluctuated between 10.5 and 11.2 MPa. By comparing with historical data, it was found that the vertical stress variation rate was about 30% higher than that of the horizontal stress, indicating that there may be a vertical geological stress concentration phenomenon in this area.

[0106] Furthermore, the process of obtaining dynamic trend data on ground pressure distribution includes:

[0107] Using initial ground pressure distribution data and mining progress data, and employing a preset grid division algorithm, the initial layout location of the monitoring equipment is determined, resulting in an initial equipment distribution scheme.

[0108] If the coverage rate of monitoring points in the initial equipment distribution plan is lower than the preset threshold, the equipment positions are adjusted through an iterative optimization algorithm to obtain an optimized monitoring equipment layout.

[0109] Based on the optimized layout of monitoring equipment, stress change data at each mining stage are collected in real time to obtain a real-time stress change dataset.

[0110] A time series analysis model was used to process the real-time stress change dataset, extract the dynamic change characteristics of ground pressure distribution, and obtain the trend data of ground pressure distribution change.

[0111] If the stress anomaly value in the ground pressure distribution trend data exceeds the preset threshold, a high-density monitoring point supplement is carried out in the anomaly area through spatial interpolation algorithm to obtain the supplemented monitoring point layout.

[0112] Based on the revised monitoring point layout, stress change data were re-collected, and the real-time stress change dataset was updated.

[0113] The updated real-time stress change dataset was analyzed using a trend analysis model to obtain the final dynamic trend data of ground pressure distribution.

[0114] For example, when determining the layout of monitoring equipment based on initial ground pressure distribution data and mining progress data, the mining area is first spatially divided using a grid partitioning algorithm. Assuming a mining area of ​​10,000 square meters, initial ground pressure distribution data shows that high-risk areas are concentrated in the central region, with a ground pressure value of approximately 12.5 MPa. The grid partitioning algorithm divides the area into 100 grid cells of 100 square meters each, and each cell is assigned a monitoring priority based on the mining progress and ground pressure distribution.

[0115] For example, grids closer to the mining face have higher priority and require high-precision sensors. In the initial layout, one stress sensor was placed every 200 square meters, for a total of 50 sensors, achieving 80% coverage. However, the preset coverage threshold is 90%, so the device positions need to be adjusted through iterative optimization algorithms.

[0116] In one possible implementation, the iterative optimization algorithm redistributes sensor locations based on ground pressure gradients, prioritizing the addition of monitoring points in high-risk grids, ultimately increasing coverage to 92% and generating an optimized device layout scheme.

[0117] Specifically, after the optimized layout scheme is implemented, the system collects stress change data in real time for each mining stage. Assuming that on the 10th day of mining, the stress value of a high-risk grid increases from 12.5 MPa to 13.8 MPa, the data is uploaded to the central platform every minute, forming a real-time stress change dataset. A time series analysis model processes the dataset to extract dynamic change characteristics.

[0118] For example, analysis revealed that stress in a certain area exhibited periodic fluctuations over 72 hours, with the peak occurring 4 hours after mining operations, and the fluctuation amplitude reaching 1.2 MPa. The data on ground pressure distribution trends indicate that the stress in this area is increasing rapidly, posing a potential risk.

[0119] In one embodiment, if an anomaly in stress is detected in the trend data, such as a stress value of 14.5 MPa in a certain grid exceeding a preset threshold of 14.0 MPa, the system identifies the abnormal region and adds monitoring points using a spatial interpolation algorithm. The spatial interpolation algorithm estimates the stress distribution in the abnormal region based on the stress values ​​of the surrounding grids, determining that five additional sensors are needed to cover 1000 square meters around the abnormal region. The supplemented monitoring point layout is then used to re-collect data and update the real-time stress change dataset.

[0120] For example, the new data shows that the stress in the anomalous area has stabilized at 14.2 MPa and has not deteriorated further. The trend analysis model processes the updated dataset to generate the final dynamic trend data of the ground pressure distribution, showing that the stress growth is gradually leveling off, indicating that the supplementary monitoring points have effectively improved data accuracy and risk warning capabilities.

[0121] For example, by analyzing dynamic trend data, the system can further combine mining progress data to predict stress distribution over the next 24 hours, assisting in the formulation of safe mining plans. This method, through multi-level data processing and equipment optimization, ensures the comprehensiveness and real-time nature of ground pressure monitoring, providing a reliable basis for mine safety management.

[0122] Furthermore, the process of integrating dynamic trend data of ground pressure distribution and mining progress data using data fusion technology to construct a correlation model between ground pressure changes and mining progress includes:

[0123] Data fusion technology is used to preprocess data related to ground pressure distribution and mining progress, extract dynamic changes and trend data, and obtain a preliminary integrated dataset.

[0124] For the initially integrated dataset, the support vector machine algorithm is used to extract features of the potential relationship between ground pressure changes and mining progress, and to determine the set of associated features;

[0125] Based on the set of associated features, a correlation model between ground pressure changes and mining progress is constructed to obtain the variation patterns and distribution characteristics in the model and to determine the preliminary applicability of the model.

[0126] If the applicability of the initial model is lower than the preset threshold, then the trend data and progress data are integrated a second time to obtain an optimized combination of data features.

[0127] By optimizing the combination of data features, the parameters of the correlation model are readjusted to obtain the updated correlation between ground pressure changes and mining progress.

[0128] Based on the updated correlation model, the distribution and dynamic changes of ground pressure are continuously monitored to obtain real-time data on the changing patterns.

[0129] Based on real-time change data, the correlation model is dynamically updated to obtain the latest distribution characteristics and trend data, and the final correlation result between ground pressure changes and mining progress is determined.

[0130] Specifically, in order to integrate the dynamic trend data of ground pressure distribution and mining progress data, and to construct the correlation model data between the two, this embodiment uses data fusion technology for automated processing.

[0131] First, assume that ground pressure distribution data is obtained from a sensor network, and the data is recorded hourly. For example, in the first 5 hours of a certain mining area on a certain day, the ground pressure values ​​are 3.2MPa, 3.5MPa, 3.8MPa, 4.0MPa and 4.2MPa, respectively, showing an upward trend.

[0132] Meanwhile, mining progress data is obtained through the mining equipment monitoring system and recorded as the hourly mining volume within the same time period, which are 50 tons, 55 tons, 60 tons, 65 tons and 70 tons respectively, also showing an increasing trend.

[0133] Next, a data preprocessing algorithm is used to standardize the two sets of data. The mean and standard deviation of the ground pressure data and the mining volume data are calculated using the Z-score standardization method. For example, the mean ground pressure is 3.74 MPa and the standard deviation is 0.38 MPa, and the mean mining volume is 60 tons and the standard deviation is 7.07 tons. The data are converted into standardized values ​​by the formula (Z = (X - mean) / standard deviation) to eliminate dimensional differences.

[0134] Then, using time series analysis algorithms, such as sliding window correlation analysis, with a window size of 3 hours, the Pearson correlation coefficient between ground pressure change and mining progress was calculated. The correlation coefficient was 0.95, indicating that the two are highly positively correlated, further verifying the hypothesis that increased mining progress may lead to increased ground pressure.

[0135] Subsequently, a correlation model was constructed using regression analysis. The linear regression algorithm was used to calculate the regression equation as Y = 0.05X + 0.2, where Y is the standardized ground pressure value and X is the standardized mining volume. The significance test p-value of the regression coefficient was 0.01, which is less than 0.05, indicating that the model has statistical significance.

[0136] Finally, the model data is stored in a database, and dynamic charts are generated using visualization tools to monitor the correlation between ground pressure and mining progress in real time, providing data support for subsequent prediction and early warning.

[0137] Through the above process, the relationship between ground pressure changes and mining progress can be quantified, forming a complete data fusion and modeling chain, ensuring the scientific nature and accuracy of the analysis.

[0138] Furthermore, based on the correlation model data, the process of obtaining the distribution data of ground pressure anomalies in each region through a comprehensive evaluation of vertical stress, horizontal stress, and rate of change includes:

[0139] Preprocessing of vertical stress, horizontal stress, and rate of change data yields stress distribution characteristics for each region.

[0140] Based on the stress distribution characteristics, and considering the numerical differences between vertical and horizontal stresses, a comprehensive analysis is conducted to determine whether there is a significant stress imbalance. If the stress difference in a certain area exceeds a preset threshold, the area is identified as a potential area of ​​ground pressure anomaly.

[0141] Data on potential geopressure anomaly areas are obtained, and combined with the dynamic trend of the rate of change, time series analysis is used to further analyze the stability of stress changes within the anomaly area and obtain the fluctuation characteristics of the rate of change.

[0142] By conducting in-depth analysis of the fluctuation characteristics of the rate of change and combining it with the geographical information of regional division, if the fluctuation amplitude of the rate of change in a certain region is higher than the preset threshold, the region is judged as a high-risk ground pressure anomaly region.

[0143] Based on the determination of high-risk ground pressure anomaly areas, and in response to the needs of stress distribution and anomaly detection, the support vector machine algorithm is used to classify the anomaly areas and determine the distribution of anomaly levels.

[0144] After obtaining the distribution of anomaly levels, and by adjusting the model parameters, corresponding risk distribution maps are generated for anomaly areas of different levels, thus obtaining the final ground pressure anomaly distribution data.

[0145] For example, in this embodiment, the acquisition of raw vertical stress, horizontal stress, and rate of change data is achieved through a multi-point stress sensor network deployed within the mining area. Assuming 100 sensor points are deployed in a mining area, recording data hourly, the vertical stress data might show an average value of 5.0 MPa, the horizontal stress of 4.5 MPa, and a rate of change of 0.2 MPa per hour. This data is wirelessly transmitted to a central database, providing a basis for subsequent analysis.

[0146] It should be noted that the sensors need to be calibrated regularly to ensure data accuracy.

[0147] In one possible implementation, the data is preliminarily processed based on a pre-established correlation model, and the stress distribution characteristics can be calculated using a weighted average method.

[0148] For example, for vertical and horizontal stress data of a certain area, a weighted fusion algorithm is used to comprehensively consider the sensor location and geological conditions to obtain the stress concentration coefficient of the area. Assuming it is 1.2, it indicates that there is slight stress concentration.

[0149] It should be noted that the model needs to be updated regularly according to the geological characteristics of the mining area to adapt to changes in different mining stages.

[0150] For example, when judging stress imbalance, a threshold of 0.8 MPa can be set for the difference between vertical and horizontal stress. If the difference in a certain area is 1.0 MPa, exceeding the threshold, it is marked as a potential area of ​​ground pressure anomaly.

[0151] Preferably, a geological fault distribution map can be used to verify whether the area is close to a fault zone, thereby improving the accuracy of the judgment.

[0152] In one possible implementation, an autoregressive model from time series analysis can be used to analyze the stability of stress changes in potential anomalous regions.

[0153] For example, regarding the rate of change data, assuming that the rate of change in a certain area is 0.15MPa, 0.18MPa, 0.20MPa, 0.22MPa and 0.25MPa for 5 consecutive hours, the fluctuation amplitude is calculated to be 0.1MPa through autoregression analysis, which is higher than the preset threshold of 0.08MPa, and is therefore identified as a high-risk ground pressure anomaly area.

[0154] It should be noted that the selection of the time window needs to be adjusted according to the mining speed in order to capture short-term fluctuations.

[0155] For example, when using the support vector machine algorithm to classify anomalous areas, stress values, rates of change, and geological features can be used as input feature vectors. Suppose a region's feature vectors show high stress values ​​and high volatility; the classification result might categorize it as a level three anomaly, indicating a need for priority attention.

[0156] Preferably, cross-validation can be used to optimize algorithm parameters and improve classification accuracy.

[0157] In one possible implementation, when generating the risk distribution map, geographic information system software can be used to overlay anomaly level data onto the mining area map.

[0158] For example, red areas represent level three anomalies, yellow areas represent level two anomalies, and green areas represent safe areas, visually demonstrating the distribution of high-risk areas.

[0159] It should be noted that dynamically updated distribution maps can help on-site managers adjust mining plans in a timely manner and reduce the risk of ground pressure accidents.

[0160] For example, by adjusting model parameters, the model's effectiveness can be verified through backtesting using historical data. Assuming that the prediction accuracy for high-risk areas in a mining area reached 90% over the past month, the threshold settings can be further optimized to improve the model's adaptability to complex geological conditions.

[0161] Preferably, feedback from on-site engineers should be incorporated regularly to ensure the model's usability.

[0162] Furthermore, the process of obtaining key monitoring strategy data for high-risk areas includes:

[0163] Based on the ground pressure anomaly distribution data, the initial distribution of monitoring points is obtained, and data cleaning techniques are used to remove outliers and noise to obtain a pre-processed distribution dataset.

[0164] Based on the pre-processed distribution dataset, the regional characteristics of ground pressure anomalies are analyzed, and spatial clustering methods are used to determine the boundaries of high-risk areas, thus obtaining the division results of high-risk areas.

[0165] Based on the results of the high-risk area classification, the layout of the monitoring equipment was adjusted, and the monitoring points were optimized by using preset density thresholds to determine the updated monitoring point data.

[0166] Starting from the updated monitoring point data, analyze the ground pressure anomaly change trend of each point. If the change trend exceeds the preset fluctuation range, increase the monitoring frequency of the corresponding point to obtain the adjusted frequency configuration data.

[0167] Based on the adjusted frequency configuration data and the distribution characteristics of high-risk areas, the equipment configuration scheme is optimized, and the equipment load is balanced through the resource allocation model to obtain the final equipment configuration data.

[0168] Based on the final equipment configuration data, a key monitoring strategy is generated. If the ground pressure anomaly data in a certain area continues to be higher than the preset threshold, monitoring resources are allocated first, and targeted monitoring strategy data is determined.

[0169] By monitoring strategy data, the operating parameters of the monitoring equipment are dynamically updated. If an abnormal change in the distribution of ground pressure is detected, the location layout is recalculated to obtain real-time optimized monitoring scheme data.

[0170] For example, in this embodiment, when acquiring the initial distribution of monitoring points, ground pressure anomaly distribution data is collected through the sensor network already deployed within the mining area. Assuming a mining area has 200 monitoring points covering the main mining area, the initial data may show abnormally high stress values ​​at some points. Data cleaning techniques are used to remove noise caused by equipment malfunctions or external interference, such as discarding readings that are clearly outside the reasonable range, to ensure the reliability of subsequent analysis.

[0171] For example, in analyzing the regional characteristics of ground pressure anomalies, this embodiment uses spatial clustering to identify the boundaries of high-risk areas. Suppose that cluster analysis reveals that multiple stress values ​​in the northern part of a mining area are consistently higher than the average level and exhibit spatial clustering characteristics; in this case, it is designated as a high-risk area. This method effectively focuses resources and avoids monitoring blind spots.

[0172] For example, when adjusting the layout of monitoring equipment based on the results of high-risk area classification, the distribution of monitoring points is optimized according to a preset density threshold. Assuming the density threshold is 5 points per square kilometer, but only 3 points are deployed in a certain high-risk area, additional points need to be added to improve coverage and ensure that abnormal changes are detected promptly.

[0173] For example, when analyzing the anomaly trend of ground pressure at various points, if the stress value at a certain point increases beyond the preset fluctuation range for several consecutive hours, such as 0.3 MPa per hour, then its monitoring frequency needs to be increased. Assuming the original frequency is once per hour, it can be adjusted to once every 30 minutes to more accurately grasp the dynamic changes.

[0174] For example, when optimizing equipment configuration based on the distribution characteristics of high-risk areas, a resource allocation model can be used to balance equipment load. Suppose a region has 10 monitoring points requiring high-frequency monitoring, but the existing equipment only supports high-frequency operation at 5 points. In this case, the points with the highest stress values ​​will be prioritized for monitoring, while additional equipment will be planned to ensure efficient resource utilization.

[0175] For example, when generating key monitoring strategies, if the ground pressure anomaly data in a certain area consistently exceeds a preset threshold, such as 5.5 MPa, monitoring resources will be allocated preferentially. Anomaly early warning capabilities can be improved by increasing the number of sensors or introducing portable detection equipment to form a multi-layered monitoring network.

[0176] For example, when dynamically updating the operating parameters of monitoring equipment, if a significant change in the distribution of abnormal ground pressure is detected, such as a sudden increase of 20% in the stress value of a certain area, the monitoring point layout will be recalculated. If, after adding new monitoring points, the abnormal area is found to have expanded, the equipment positions can be further adjusted to form a real-time optimized monitoring scheme. This dynamic adjustment can effectively cope with sudden changes in complex geological environments and improve the level of safety assurance.

[0177] Furthermore, the process of obtaining the final optimized ground pressure monitoring data includes:

[0178] Real-time monitoring data is collected and compared with preset stress thresholds to determine whether there are any exceedances, and a preliminary risk assessment result is obtained.

[0179] If the preliminary risk assessment results show that the monitoring data exceeds the stress threshold, the early warning response mechanism will be triggered to generate corresponding risk early warning information data and determine the specific risk level classification.

[0180] Based on risk warning information data and risk level classification, a pre-established decision-making model is used to obtain targeted equipment layout adjustment strategies and output preliminary layout optimization data.

[0181] Based on the preliminary layout optimization data and combined with the historical records of ground pressure monitoring, it is determined whether there are any layout conflicts or redundancies, and the adjusted equipment layout scheme data is obtained.

[0182] If the adjusted equipment layout data meets the preset optimization conditions, the final ground pressure monitoring optimization data will be generated to determine the latest deployment location of the monitoring equipment.

[0183] Based on the final optimized ground pressure monitoring data, the support vector machine algorithm is used to predict and analyze the stress distribution at the monitoring points to obtain the distribution characteristics of potential risk areas.

[0184] Based on the distribution characteristics of potential risk areas, the frequency and scope of monitoring data collection are dynamically updated to obtain continuously optimized ground pressure monitoring strategy data.

[0185] For example, in this embodiment, during the real-time acquisition of monitoring data, sensors deployed at key locations in the mining area collect ground pressure data every 5 minutes, recording the stress value at each monitoring point and comparing it with a preset stress threshold, such as 50 MPa. If the stress value at a monitoring point reaches 55 MPa, significantly exceeding the threshold, the system automatically identifies it as an over-limit situation and generates a preliminary risk assessment result. This method can quickly identify potential problems and provide data support for subsequent processing.

[0186] For example, regarding the triggering of the early warning response mechanism, when monitored data exceeds a threshold, the system generates risk warning information based on the extent of the exceedance and categorizes it into three risk levels. For instance, a stress value exceeding the threshold by less than 10% is considered Level 1 risk, exceeding it by 10% to 20% is Level 2 risk, and exceeding it by more than 20% is Level 3 risk. Taking a stress value of 60 MPa at a monitoring point as an example, exceeding the threshold by 20%, it would be classified as Level 3 risk, and the system would prioritize pushing high-level warning information to ensure timely response from relevant personnel.

[0187] For example, when using a decision model to obtain equipment layout adjustment strategies, based on historical ground pressure anomaly distribution data and the current risk level, it is recommended to increase the density of monitoring equipment in specific areas. Suppose a high-risk area originally had only 3 monitoring points; after analysis, the decision model suggests increasing this to 5, prioritizing coverage of core areas where ground pressure anomalies frequently occur. This adjustment strategy can more accurately capture data changes and improve the comprehensiveness of monitoring.

[0188] For example, the system identifies layout conflicts or redundancy by analyzing whether the devices are too close together or their coverage overlaps. If two monitoring points in a certain area are only 50 meters apart, far below the preset minimum spacing of 100 meters, the system will mark it as redundant and suggest removing one of the points or adjusting its location. This method helps optimize resource allocation and avoid waste.

[0189] For example, when generating the final optimized ground pressure monitoring plan, if the adjusted layout meets the preset conditions for equipment density and coverage, such as an average of no less than 2 monitoring points per square kilometer in each high-risk area, the system will confirm the feasibility of the plan and output the latest equipment deployment locations. This approach ensures the rationality of the monitoring network.

[0190] For example, predictive analysis of the distribution characteristics of potential risk areas uses support vector machine algorithms to simulate stress distribution trends at monitoring points, identifying areas that may experience high stress in the future. For instance, if the prediction shows that the stress value in a certain area may rise from 40 MPa to 52 MPa within the next week, the system will mark it as a potential risk area. This kind of prediction helps to take preventative measures in advance.

[0191] For example, when dynamically updating the frequency and range of monitoring data collection, if a potentially risky area is identified, the system will adjust the collection frequency for that area from once every 5 minutes to once every 2 minutes, while simultaneously expanding the monitoring range to a radius of 500 meters. This dynamic adjustment allows for more timely understanding of ground pressure changes, improving the accuracy of early warnings and the speed of response.

[0192] Example 2

[0193] like Figure 2 As shown, based on the same inventive concept, this embodiment also provides a ground pressure monitoring and early warning system for deep mining metal mines, including:

[0194] The comprehensive evaluation module is used to acquire topographic slope data and geological structure feature data of the mining area. It uses a weighted analysis method to comprehensively evaluate the topographic slope data and geological structure feature data to obtain the ground pressure monitoring priority data and preliminary equipment configuration requirements data for each area.

[0195] The three-dimensional stress analysis module is used to configure regional equipment based on ground pressure monitoring priority data and preliminary equipment configuration requirements, and to obtain three-dimensional stress distribution data; based on the three-dimensional stress distribution data, it extracts the variation characteristics of vertical and horizontal stress to obtain the initial state data of ground pressure distribution in each region;

[0196] The dynamic monitoring and adjustment module is used to dynamically adjust the layout of monitoring equipment based on the initial state data of ground pressure distribution and mining progress data, and to collect stress change information at different mining stages in real time to obtain dynamic change trend data of ground pressure distribution.

[0197] The correlation model building module is used to integrate dynamic change trend data of ground pressure distribution and mining progress data using data fusion technology to build a correlation model data between ground pressure change and mining progress.

[0198] The anomaly assessment module is used to obtain the distribution data of ground pressure anomalies in each region based on the correlation model data and through a comprehensive assessment of vertical stress, horizontal stress and rate of change.

[0199] The monitoring strategy optimization module is used to adjust the layout of monitoring equipment based on the abnormal distribution data of ground pressure, obtain updated monitoring point data, and obtain key monitoring strategy data for high-risk areas by optimizing equipment configuration and monitoring frequency.

[0200] The early warning response module is used to trigger the early warning response mechanism if the key monitoring strategy data exceeds the preset stress threshold, obtain risk early warning information data, dynamically adjust the layout of monitoring equipment, and obtain the final ground pressure monitoring optimization scheme data.

[0201] The ground pressure monitoring and early warning system for deep mining metal mines provided in this embodiment has all the advantages of the ground pressure monitoring and early warning method for deep mining metal mines provided in Embodiment 1.

[0202] Example 3

[0203] This embodiment also discloses a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described in Embodiment 1.

[0204] Example 4

[0205] This embodiment also discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in Embodiment 1.

[0206] Example 5

[0207] This embodiment also discloses a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in Embodiment 1.

[0208] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for monitoring and early warning of ground pressure in a metal mine for deep mining, characterized in that, include: Data on topographic slope and geological structure features of the mining area are acquired. A weighted analysis method is used to comprehensively evaluate the topographic slope and geological structure features to obtain the priority data for ground pressure monitoring and the preliminary equipment configuration requirements for each area. Based on the ground pressure monitoring priority data and preliminary equipment configuration requirements data, regional equipment configuration is carried out, and three-dimensional stress distribution data is obtained; Based on the three-dimensional stress distribution data, the variation characteristics of vertical and horizontal stress are extracted to obtain the initial state data of ground pressure distribution in each region; Based on the initial state data of ground pressure distribution and the mining progress data, the layout of monitoring equipment is dynamically adjusted, and stress change information at different mining stages is collected in real time to obtain dynamic change trend data of ground pressure distribution. Data fusion technology is used to integrate the dynamic trend data of ground pressure distribution and mining progress data to construct a correlation model data between ground pressure changes and mining progress. Based on the data from the aforementioned correlation model, the distribution data of ground pressure anomalies in each region are obtained through a comprehensive evaluation of vertical stress, horizontal stress, and rate of change. Based on the ground pressure anomaly distribution data, the layout of monitoring equipment is adjusted, updated monitoring point data is obtained, and key monitoring strategy data for high-risk areas is obtained by optimizing equipment configuration and monitoring frequency. If the key monitoring strategy data exceeds the preset stress threshold, an early warning response mechanism is triggered to obtain risk early warning information data, and the layout of monitoring equipment is dynamically adjusted to obtain the final optimized ground pressure monitoring scheme data. The process of integrating the dynamic trend data of ground pressure distribution and mining progress data using data fusion technology to construct a correlation model between ground pressure changes and mining progress includes: Data fusion technology is used to preprocess data related to ground pressure distribution and mining progress, extract dynamic changes and trend data, and obtain a preliminary integrated dataset. For the initially integrated dataset, the support vector machine algorithm is used to extract features of the potential relationship between ground pressure changes and mining progress, and to determine the set of associated features; Based on the aforementioned set of related features, a correlation model between ground pressure changes and mining progress is constructed to obtain the variation patterns and distribution characteristics in the model and to determine the preliminary applicability of the model. If the applicability of the initial model is lower than the preset threshold, then the trend data and progress data are integrated a second time to obtain an optimized combination of data features. By optimizing the combination of data features, the parameters of the correlation model are readjusted to obtain the updated correlation between ground pressure changes and mining progress. Based on the updated correlation model, the distribution and dynamic changes of ground pressure are continuously monitored to obtain real-time data on the changing patterns. Based on the real-time change data, the correlation model is dynamically updated to obtain the latest distribution characteristics and trend data, and the final correlation result between ground pressure change and mining progress is determined.

2. The method according to claim 1, characterized in that, The process of using a weighted analysis method to comprehensively evaluate the topographic slope data and geological structural feature data to obtain the priority data for ground pressure monitoring and the preliminary equipment configuration requirements for each region includes: Acquire topographic slope data and geological structure feature data within the mining area. Conduct a comprehensive scan of each area of ​​the mining area through a pre-established data acquisition system to obtain an initial dataset of topographic slope and geological structure distribution. Based on the initial dataset of terrain slope and geological structure distribution, a weighted analysis method was used to quantify the terrain slope data and geological structure complexity values ​​to determine the comprehensive evaluation score for each region. By comprehensively evaluating the scores, the ground pressure monitoring needs of each region are ranked, and the priority data of ground pressure monitoring in each region is determined. If the comprehensive evaluation score of a certain region is higher than the preset threshold, the priority is marked as high, and the priority ranking result is obtained.

3. The method according to claim 1, characterized in that, The process of extracting the variation characteristics of vertical and horizontal stress based on the three-dimensional stress distribution data to obtain the initial state data of ground pressure distribution in each region includes: By combining ground pressure monitoring data and priority data with equipment configuration schemes, monitoring equipment is deployed in high-priority areas to collect three-dimensional stress distribution data; Based on the three-dimensional stress distribution data, the distribution characteristics of vertical stress and horizontal stress are extracted. Data preprocessing techniques are used to denoise and standardize the extracted features to obtain standardized stress feature data. For the standardized stress feature data, the support vector machine algorithm is applied to classify the variation characteristics of vertical stress and horizontal stress, determine the degree of anomaly in stress variation in each region, and determine the distribution range of the abnormal region. If the classification results show that the stress change in certain areas exceeds the preset threshold, then the three-dimensional stress distribution data of the corresponding areas will be analyzed in depth to obtain the stress change trend in the abnormal areas and obtain the stress change pattern of the high-risk areas. Based on the stress change patterns in high-risk areas and combined with priority data, the deployment locations of monitoring equipment are dynamically adjusted to obtain the adjusted equipment configuration scheme and determine the optimized monitoring coverage. By optimizing the monitoring coverage, we continuously collect updated three-dimensional stress distribution data, extract new vertical and horizontal stress characteristics, and update the initial state data of ground pressure distribution.

4. The method according to claim 1, characterized in that, The process of obtaining dynamic trend data on ground pressure distribution includes: Using initial ground pressure distribution data and mining progress data, and employing a preset grid division algorithm, the initial layout location of the monitoring equipment is determined, resulting in an initial equipment distribution scheme. If the coverage rate of monitoring points in the initial distribution scheme of the equipment is lower than the preset threshold, the equipment positions are adjusted through an iterative optimization algorithm to obtain an optimized monitoring equipment layout. Based on the optimized monitoring equipment layout, stress change data for each mining stage are collected in real time to obtain a real-time stress change dataset. The real-time stress change dataset is processed using a time series analysis model to extract the dynamic change characteristics of ground pressure distribution and obtain ground pressure distribution change trend data; If the stress anomaly value in the ground pressure distribution trend data exceeds the preset threshold, a high-density monitoring point supplement is performed in the anomaly area using a spatial interpolation algorithm to obtain the supplemented monitoring point layout. Based on the revised monitoring point layout, stress change data were re-collected, and the real-time stress change dataset was updated. The updated real-time stress change dataset was analyzed using a trend analysis model to obtain the final dynamic trend data of ground pressure distribution.

5. The method according to claim 1, characterized in that, Based on the aforementioned correlation model data, the process of obtaining the distribution data of ground pressure anomalies in each region through a comprehensive evaluation of vertical stress, horizontal stress, and rate of change includes: Preprocessing of vertical stress, horizontal stress, and rate of change data yields stress distribution characteristics for each region. Based on the stress distribution characteristics, and considering the numerical differences between vertical and horizontal stresses, a comprehensive analysis is conducted to determine whether there is a significant stress imbalance. If the stress difference in a certain area exceeds a preset threshold, then the area is identified as a potential area of ​​ground pressure anomaly. Data on potential geopressure anomaly areas are obtained, and combined with the dynamic trend of the rate of change, time series analysis is used to further analyze the stability of stress changes within the anomaly area and obtain the fluctuation characteristics of the rate of change. By conducting in-depth analysis of the fluctuation characteristics of the rate of change and combining it with the geographical information of regional division, if the fluctuation amplitude of the rate of change in a certain region is higher than a preset threshold, the region is determined to be a high-risk ground pressure anomaly region. Based on the determination of high-risk ground pressure anomaly areas, and in response to the needs of stress distribution and anomaly detection, the support vector machine algorithm is used to classify the anomaly areas and determine the distribution of anomaly levels. After obtaining the distribution of anomaly levels, and by adjusting the model parameters, corresponding risk distribution maps are generated for anomaly areas of different levels, thus obtaining the final ground pressure anomaly distribution data.

6. The method according to claim 1, characterized in that, The process of obtaining key monitoring strategy data for high-risk areas includes: Based on the ground pressure anomaly distribution data, the initial distribution of monitoring points is obtained, and data cleaning techniques are used to remove outliers and noise to obtain a preliminarily processed distribution dataset. Based on the pre-processed distribution dataset, the regional characteristics of ground pressure anomalies are analyzed, and spatial clustering methods are used to determine the boundaries of high-risk areas, thus obtaining the division results of high-risk areas. Based on the results of the high-risk area classification, the layout of the monitoring equipment was adjusted, and the monitoring points were optimized by using preset density thresholds to determine the updated monitoring point data. Starting from the updated monitoring point data, analyze the ground pressure anomaly change trend of each point. If the change trend exceeds the preset fluctuation range, increase the monitoring frequency of the corresponding point to obtain the adjusted frequency configuration data. Based on the adjusted frequency configuration data and the distribution characteristics of high-risk areas, the equipment configuration scheme is optimized, and the equipment load is balanced through the resource allocation model to obtain the final equipment configuration data. Based on the final equipment configuration data, a key monitoring strategy is generated. If the ground pressure anomaly data in a certain area continues to be higher than the preset threshold, monitoring resources are allocated first, and targeted monitoring strategy data is determined. By monitoring strategy data, the operating parameters of the monitoring equipment are dynamically updated. If an abnormal change in the distribution of ground pressure is detected, the location layout is recalculated to obtain real-time optimized monitoring scheme data.

7. The method according to claim 1, characterized in that, The process of obtaining the final optimized ground pressure monitoring data includes: Real-time monitoring data is collected and compared with preset stress thresholds to determine whether there are any exceedances, and a preliminary risk assessment result is obtained. If the preliminary risk assessment results show that the monitoring data exceeds the stress threshold, the early warning response mechanism will be triggered to generate corresponding risk early warning information data and determine the specific risk level classification. Based on risk warning information data and risk level classification, a pre-established decision-making model is used to obtain targeted equipment layout adjustment strategies and output preliminary layout optimization data. Based on the preliminary layout optimization data and combined with the historical records of ground pressure monitoring, it is determined whether there are any layout conflicts or redundancies, and the adjusted equipment layout scheme data is obtained. If the adjusted equipment layout data meets the preset optimization conditions, the final ground pressure monitoring optimization data will be generated to determine the latest deployment location of the monitoring equipment. Based on the data from the final optimized ground pressure monitoring scheme, the support vector machine algorithm is used to predict and analyze the stress distribution at the monitoring points, thereby obtaining the distribution characteristics of potential risk areas. Based on the distribution characteristics of potential risk areas, the frequency and scope of monitoring data collection are dynamically updated to obtain continuously optimized ground pressure monitoring strategy data.

8. A metal mine ground pressure monitoring and early warning system for deep mining, characterized in that, include: The comprehensive evaluation module is used to acquire topographic slope data and geological structure feature data of the mining area, and to comprehensively evaluate the topographic slope data and geological structure feature data using a weighted analysis method to obtain ground pressure monitoring priority data and preliminary equipment configuration requirements data for each area. The three-dimensional stress analysis module is used to configure regional equipment and obtain three-dimensional stress distribution data based on the ground pressure monitoring priority data and preliminary equipment configuration requirements data; and to extract the variation characteristics of vertical stress and horizontal stress based on the three-dimensional stress distribution data to obtain the initial state data of ground pressure distribution in each region. The dynamic monitoring and adjustment module is used to dynamically adjust the layout of monitoring equipment based on the initial state data of ground pressure distribution and mining progress data, collect stress change information at different mining stages in real time, and obtain dynamic change trend data of ground pressure distribution. The correlation model construction module is used to integrate the dynamic change trend data of ground pressure distribution and mining progress data using data fusion technology to construct the correlation model data between ground pressure change and mining progress. Specifically, data fusion technology is used to preprocess data related to ground pressure distribution and mining progress, extract dynamic changes and trend data, and obtain a preliminary integrated dataset. For the initially integrated dataset, the support vector machine algorithm is used to extract features of the potential relationship between ground pressure changes and mining progress, and to determine the set of associated features; Based on the aforementioned set of related features, a correlation model between ground pressure changes and mining progress is constructed to obtain the variation patterns and distribution characteristics in the model and to determine the preliminary applicability of the model. If the applicability of the initial model is lower than the preset threshold, then the trend data and progress data are integrated a second time to obtain an optimized combination of data features. By optimizing the combination of data features, the parameters of the correlation model are readjusted to obtain the updated correlation between ground pressure changes and mining progress. Based on the updated correlation model, the distribution and dynamic changes of ground pressure are continuously monitored to obtain real-time data on the changing patterns. Based on the real-time change data, the correlation model is dynamically updated to obtain the latest distribution characteristics and trend data, and the final correlation result between ground pressure change and mining progress is determined. The anomaly assessment module is used to obtain the distribution data of ground pressure anomalies in each region based on the associated model data through a comprehensive assessment of vertical stress, horizontal stress, and rate of change. The monitoring strategy optimization module is used to adjust the layout of monitoring equipment based on the ground pressure anomaly distribution data, obtain updated monitoring point data, and obtain key monitoring strategy data for high-risk areas by optimizing equipment configuration and monitoring frequency. The early warning response module is used to trigger an early warning response mechanism if the key monitoring strategy data exceeds a preset stress threshold, obtain risk early warning information data, dynamically adjust the layout of monitoring equipment, and obtain the final ground pressure monitoring optimization scheme data.

Images