Monitoring System for Mine Rock and Soil Cracking and Local Collapse Control Based on Satellite Remote Sensing

Abstract

This invention discloses a monitoring system for the treatment of cracking and local collapse of rock and soil in mines based on satellite remote sensing. The system includes: a satellite remote sensing unit for comprehensive three-dimensional monitoring of the mining area, acquiring real-world three-dimensional data of the mining area; a data analysis unit for analyzing the real-world three-dimensional data and obtaining analysis results, including illegal production and construction activities in the mine, local potential risk points, and types of safety hazards; a monitoring and early warning unit for generating early warning information based on the analysis results, targeting the current type of safety hazard, including cracking of rock and soil and local collapse; and a treatment and management unit for formulating corresponding mine treatment plans based on the current early warning information, combined with mine design data and development specifications. This system provides scientific and targeted early warning information, giving mine managers valuable time to address potential hazards and thus reducing the accident rate.

Description

Technical Field

[0001] This invention relates to the field of geological monitoring technology, and in particular to a monitoring system for the treatment of cracking and local collapse of rock and soil in mines based on satellite remote sensing. Background Technology

[0002] During mining operations, the stability of the rock and soil mass directly affects production safety and environmental protection. Traditional mine monitoring technologies primarily rely on on-site inspections and visual patrols, methods with numerous limitations. First, on-site inspections struggle to cover the entire mine, especially in high-altitude work areas, complex terrain, and hazardous locations, creating blind spots that are difficult to see or access, making it hard to detect and address safety hazards in a timely manner. Second, manual monitoring is susceptible to the influence of personnel skills and subjective judgment, resulting in inconsistent monitoring effectiveness and significant errors and oversights. Furthermore, the limited frequency and efficiency of on-site inspections fail to meet the rapidly changing monitoring needs of the mine environment, increasing the risk of accidents.

[0003] While traditional remote sensing monitoring methods can provide information on surface changes over a wide area, they still fall short in terms of resolution, accuracy, and real-time performance, making it difficult to accurately identify and analyze minor safety hazards in mines, such as cracks in soil and rock masses and localized collapses. Especially under complex geological conditions, the processing and analysis of traditional remote sensing data requires advanced professional skills and complex algorithms, limiting its widespread adoption and application in grassroots regulatory departments. Furthermore, existing remote sensing technologies struggle to provide a comprehensive, objective, and accurate reconstruction of both the overall appearance and details of a mine, failing to fully meet the needs of safety supervision operations.

[0004] Therefore, there is an urgent need for a monitoring system for the treatment of cracking and local collapse of rock and soil in mines based on satellite remote sensing. Summary of the Invention

[0005] This invention provides a monitoring system for the treatment of cracking and local collapse of rock and soil in mines based on satellite remote sensing, in order to solve the above-mentioned problems existing in the prior art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A satellite remote sensing-based monitoring system for monitoring and managing cracking and localized collapse of rock and soil in mines includes:

[0008] Satellite remote sensing units are used to conduct full-range three-dimensional monitoring of mining areas and acquire real-world three-dimensional data of the mining area;

[0009] The data analysis unit is used to analyze real-world 3D data and obtain analysis results, including illegal production and construction activities in the mine, local potential risk points, and types of safety hazards.

[0010] The monitoring and early warning unit is used to generate early warning information based on the analysis results, which includes rock and soil cracking and local collapse.

[0011] The governance and management unit is used to formulate corresponding mine governance plans based on current early warning information, combined with mine design data and development specifications.

[0012] The satellite remote sensing unit includes:

[0013] The orbit planning module is used to plan the optimal observation orbit for satellites based on the geographical features and monitoring needs of the mining area, so as to ensure comprehensive coverage of the mining area;

[0014] The data acquisition triggering module is used to synchronously trigger the data acquisition process when the satellite passes through the mining area based on the orbit planning results, and to record the acquisition time and location information in real time.

[0015] The data acquisition module is used to acquire lidar data and multispectral data after the data acquisition process is triggered;

[0016] The stereo image synthesis module is used to fuse lidar data and multispectral data based on 3D modeling technology to generate real-world 3D image data of the mining area.

[0017] The data analysis unit includes:

[0018] The data receiving module is used to receive real-scene 3D data and extract the initial analysis object from the real-scene 3D data;

[0019] The initial analysis determination module is used to determine the analysis results for the initial analysis object. The analysis results include illegal production and construction activities in the mine, local potential risk points, and corresponding types of safety hazards.

[0020] The second analysis determination module is used to identify and classify multiple new potential analysis objects based on the analysis results, and determine the analysis results for each new potential analysis object until the number of consecutive local potential risk points identified in the new analysis results reaches a preset number threshold, or the current analysis object is the last analysis object.

[0021] The analysis results report generation module is used to generate a comprehensive analysis results report based on each analysis result and the corresponding safety hazard type, and output the report content, including a list of illegal production and construction activities in the mine, details of local potential risk points, and a classification description of the safety hazard type.

[0022] The monitoring and early warning unit includes:

[0023] The first early warning module is used to determine safety hazards based on the analysis results and obtain the safety hazard determination results;

[0024] The early warning information generation module generates early warning information for the current type of safety hazard based on the safety hazard assessment results. The types of safety hazards include rock and soil cracking and local collapse.

[0025] The governance and management unit includes:

[0026] The early warning assessment module is used to assess the severity of the type of safety hazard based on the current early warning information;

[0027] The governance plan formulation module is used to determine the necessary reinforcement or support measures when the early warning information indicates the potential for cracking in the soil and rock mass, in conjunction with the mine design data.

[0028] When the early warning information indicates a potential for localized collapse, corresponding protection or repair plans should be formulated in accordance with the development specifications.

[0029] The governance scheme optimization module is used to optimize the preliminary governance scheme by comprehensively considering the requirements of mine design data and development specifications. If multiple safety hazards in the early warning information exist at the same time, the governance scheme optimization module will comprehensively formulate integrated governance measures to improve the overall governance effect.

[0030] The governance plan generation module generates corresponding mine governance plans based on the optimized governance measures. The governance plan includes specific implementation steps, required materials, time schedule and expected results.

[0031] Among them, the assessment of safety hazards includes:

[0032] If the illegal production and construction activities of a mine exceed the illegal production threshold but are less than the preset illegal production limit, local potential risk points are identified.

[0033] If a local potential risk point exceeds the potential risk point threshold but is less than the preset risk point limit, a safety hazard judgment result is generated.

[0034] This includes generating early warning information specific to the current type of safety hazard, including:

[0035] Obtain the severity level associated with each hazard type, and use it as the data group to be analyzed;

[0036] Based on the type of hazard, each data group to be analyzed is sorted, and the data corresponding to each hazard type and severity level is stored in the corresponding data group to be analyzed to obtain the hazard data to be analyzed.

[0037] Obtain safety hazard early warning standard data, and extract early warning rule data from the safety hazard early warning standard data;

[0038] After noise removal processing of the early warning rule data, rule feature vectors are constructed, and standard correlation data are obtained from the safety hazard early warning standard data.

[0039] After clustering the rule feature vectors based on the standardized associated data, a training data set is obtained. The initial model is then trained using the training data set through question-and-answer training to obtain the early warning analysis model.

[0040] The data of potential hazards to be analyzed are input one by one into the preset early warning analysis model to generate early warning information of the current safety hazards;

[0041] Generate tiered early warning information based on the type and severity level of safety hazards;

[0042] The tiered early warning information is input into the early warning analysis model and integrated to generate the final monitoring and early warning information data.

[0043] The generation of early warning information for current safety hazards includes:

[0044] The first potential hazard data to be analyzed is input into the early warning analysis model. The first potential hazard data to be analyzed is then matched for similarity in the early warning analysis model. Based on the matching results, the first early warning result is output.

[0045] Starting with the second potential hazard data to be analyzed, the previous warning result is input into the warning analysis model to obtain the corresponding associated warning results. The first warning result and all associated warning results are then integrated to generate warning information for the current type of safety hazard.

[0046] Among them, graded early warning information is generated based on the type and severity level of safety hazards, including:

[0047] When the hazard type is rock and soil cracking and the severity level is low, a primary warning message is triggered;

[0048] When the hazard type is rock and soil cracking and the severity level is high, a medium-level early warning information is triggered;

[0049] When the type of hazard is local collapse and the severity level is low, an intermediate warning message is triggered;

[0050] When the hazard type is local collapse and the severity level is high, an advanced early warning information is triggered.

[0051] The optimization of the initially formulated governance plan includes:

[0052] At the moment of each warning message, identify the types of existing safety hazards and analyze the relationships between the hazards;

[0053] Based on the mutual influence between potential hazards, the preliminary remediation plan was optimized and the details of the remediation measures were adjusted.

[0054] For situations where multiple safety hazards exist simultaneously, comprehensive management measures should be formulated to improve the overall management effectiveness, including:

[0055] When there is a risk of soil and rock cracking and local collapse, the optimized treatment plan includes prioritizing anchor bolt support measures based on the cracking risk and implementing concrete reinforcement schemes based on the collapse risk to ensure the treatment effect.

[0056] Based on the relationships between different potential hazards, and taking into account the severity of the hazards, risk assessment results, and technical requirements for remediation, a joint remediation plan should be developed to avoid redundant remediation and improve resource utilization efficiency.

[0057] The remediation plan is adjusted in real time based on the actual monitoring data and remediation results of each potential hazard to ensure the coordination and effectiveness of various measures.

[0058] In comprehensive management measures, the corresponding support structure, reinforcement technology and protective materials are adjusted according to the spatiotemporal characteristics of different hazards to ensure the feasibility and economy of the management plan.

[0059] The optimized governance plan is based on the following steps:

[0060] Acquire monitoring data for each current hazard, and combine the interactions between hazards to calculate the comprehensive mitigation effect of the current hazard using a pre-set dynamic model;

[0061] Based on the calculated comprehensive treatment effect, the specific parameters of the treatment measures of the anchor bolt support and reinforcement scheme are adjusted to form the final optimized treatment scheme;

[0062] If the actual effect of the optimization plan deviates from the expected effect, the plan will be further optimized based on dynamic monitoring data and real-time feedback to form a closed-loop feedback mechanism to improve the governance effect.

[0063] Compared with the prior art, the present invention has the following advantages:

[0064] The satellite remote sensing-based monitoring system for mine rock and soil cracking and localized collapse includes: a satellite remote sensing unit for comprehensive three-dimensional monitoring of the mining area, acquiring real-world 3D data; a data analysis unit for analyzing the real-world 3D data, obtaining analysis results including illegal production and construction activities, potential local risk points, and types of safety hazards; a monitoring and early warning unit for generating early warning information based on the analysis results, targeting specific safety hazard types such as rock and soil cracking and localized collapse; and a management unit for developing corresponding mine management plans based on the current early warning information, combined with mine design data and development specifications. This system provides scientific and targeted early warning information, giving mine managers valuable time to address potential hazards and thus reducing the accident rate.

[0065] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention.

[0066] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0067] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0068] Figure 1 This is a structural diagram of the monitoring system for the treatment of cracking and local collapse of rock and soil in mines based on satellite remote sensing, as described in this embodiment of the invention.

[0069] Figure 2 This is a structural diagram of a satellite remote sensing unit in an embodiment of the present invention;

[0070] Figure 3 This is a structural diagram of the data analysis unit in an embodiment of the present invention. Detailed Implementation

[0071] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0072] This invention provides a monitoring system for the treatment of cracking and local collapse of rock and soil in mines based on satellite remote sensing, including:

[0073] Satellite remote sensing units are used to conduct full-range three-dimensional monitoring of mining areas and acquire real-world three-dimensional data of the mining area;

[0074] The data analysis unit is used to analyze real-world 3D data and obtain analysis results, including illegal production and construction activities in the mine, local potential risk points, and types of safety hazards.

[0075] The monitoring and early warning unit is used to generate early warning information based on the analysis results, which includes rock and soil cracking and local collapse.

[0076] The governance and management unit is used to formulate corresponding mine governance plans based on current early warning information, combined with mine design data and development specifications.

[0077] The working principle of the above technical solution is as follows: A satellite remote sensing unit uses high-resolution satellite imagery to scan the entire mining area and acquires three-dimensional real-world data of the mining area through stereo imaging. Remote sensing technology can perceive the shape and material of surface objects through the reflection and absorption characteristics of electromagnetic waves. For example, satellite sensors (such as optical imaging or radar sensors) can be used to image the terrain of the mining area from multiple angles, and a high-precision three-dimensional model can be generated through data fusion and stereo reconstruction techniques. Remote sensing technology: a technology that uses electromagnetic waves (such as light, infrared, or microwaves) to perceive and analyze ground objects. Stereo imaging: a technology that generates a three-dimensional effect by taking images of the same area from multiple angles.

[0078] The data analysis unit, based on artificial intelligence (AI) and big data technologies, processes and analyzes 3D data. Algorithms detect topographical changes, abnormal building morphologies, and variations in the mechanical parameters of soil and rock masses in mining areas, thereby identifying violations and potential hazards. For example, by comparing 3D data from different periods, the system can detect a sudden and significant subsidence in a certain area, indicating a geological problem caused by illegal mining. AI algorithms can also classify and identify hazard types, such as soil and rock cracks or landslide risks.

[0079] Based on the results from the analysis unit, the monitoring and early warning unit generates corresponding early warning information through risk models and a rule base. The early warning system transmits the type of hazard, possible consequences, and recommended measures to relevant personnel. For example, if the system detects a crack at the edge of a mine slope, it will determine, based on the risk assessment model, whether the crack's development could lead to a local collapse and generate an early warning. Risk model: A mathematical model used to assess potential risks based on historical data and experience. Rule base: A knowledge base containing information on risk identification and mitigation recommendations.

[0080] Based on early warning information and mine design data, and in accordance with industry standards (such as the "Technical Regulations for Mine Safety Production"), the governance and management unit formulates targeted governance plans. These plans may include geological reinforcement, drainage optimization, and mining adjustments. If a high risk of landslide is identified, the governance and management unit will, based on design data, propose recommendations: install drainage facilities at the slope crest to reduce instability caused by rainwater infiltration, and strengthen the support structures around the slope. Support structures are devices used to support mine slopes or tunnel structures, such as steel mesh or retaining walls.

[0081] The beneficial effects of the above technical solution are as follows: The satellite remote sensing unit enables full-range, uninterrupted monitoring of the mining area, providing accurate and timely data support, especially for mines in remote areas, reducing blind spots in traditional manual inspections. The data analysis unit utilizes artificial intelligence technology to quickly identify illegal activities and potential hazards in the mine, significantly improving the efficiency of hazard investigation while reducing errors in human judgment. The monitoring and early warning unit, based on real-time data and analysis results, provides scientific and targeted early warning information, giving mine managers valuable time to address hazards and thus reducing the accident rate.

[0082] In another embodiment, the satellite remote sensing unit includes:

[0083] The orbit planning module is used to plan the optimal observation orbit for satellites based on the geographical features and monitoring needs of the mining area, so as to ensure comprehensive coverage of the mining area;

[0084] The data acquisition triggering module is used to synchronously trigger the data acquisition process when the satellite passes through the mining area based on the orbit planning results, and to record the acquisition time and location information in real time.

[0085] The data acquisition module is used to acquire lidar data and multispectral data after the data acquisition process is triggered;

[0086] The stereo image synthesis module is used to fuse lidar data and multispectral data based on 3D modeling technology to generate real-world 3D image data of the mining area.

[0087] The optimal observation orbits for the planned satellites include:

[0088] Obtain information on the geographical features and monitoring needs of the mining area;

[0089] Based on geographical features and monitoring needs, multiple first-track planning strategies are generated;

[0090] The first track planning strategy is traversed sequentially. During each traversal, the feature-performance value library corresponding to the first track planning strategy is obtained.

[0091] The monitoring requirement information is broken down into multiple secondary monitoring requirement items;

[0092] Feature extraction is performed on the second monitoring requirement item to obtain multiple third features;

[0093] Based on the feature-performance value library, determine the performance value corresponding to the third feature and associate it with the corresponding second monitoring requirement item;

[0094] The performance values ​​associated with the second monitoring requirement item are summed to obtain the sum of the performance values.

[0095] Obtain the performance value and threshold corresponding to the first track planning strategy traversed. If the sum of the performance value and the threshold is greater than or equal to the performance value and the threshold, the corresponding second monitoring requirement item will be used as the third monitoring requirement item.

[0096] Obtain the adaptation performance model corresponding to the first track planning strategy that has been traversed. Based on the adaptation performance model, according to the third monitoring requirement, perform adaptation performance on the first track planning strategy that has been traversed, obtain the adaptation value, and associate it with the first track planning strategy that has been traversed.

[0097] When the first track planning strategy is completed, the adaptation value associated with the first track planning strategy is accumulated to obtain the adaptation value sum.

[0098] The first track planning strategy with the largest fit value is used as the second track planning strategy.

[0099] Based on the second orbit planning strategy, the optimal observation orbit of the satellite is planned to ensure comprehensive coverage of the mining area, especially for monitoring the cracking and local collapse of the rock and soil in the mine, and the final observation orbit is obtained.

[0100] The working principle of the above technical solution is as follows: The orbit planning module utilizes orbital dynamics calculations and the geographical characteristics of the mining area (such as terrain complexity and area), combined with monitoring requirements (monitoring frequency and monitoring time period), to plan the optimal satellite orbit, ensuring comprehensive coverage of the observation range and efficient use of satellite resources. Assuming a mining area needs to be monitored, and this area is located in a mountainous region and covers a large area, the orbit planning module will calculate the optimal path for a Low Earth Orbit (LEO), allowing the satellite to pass over the area multiple times at the optimal angle. Simultaneously, the module will avoid excessive overlap in observations, conserving satellite resources. Orbital dynamics: the study of the forces and motion of satellites in orbit. Low Earth Orbit (LEO): a satellite orbit relatively close to the Earth's surface (approximately 200-2000 kilometers).

[0101] When the satellite passes through the target area according to the orbital planning results, the data acquisition triggering module synchronously triggers the remote sensing equipment on the satellite to begin collecting data, while recording the acquisition time and the satellite's position coordinates. This ensures spatiotemporal consistency in data acquisition. For example, when the satellite enters the planned orbit and approaches the mining area, the system automatically triggers the lidar and multispectral sensors to collect surface information and records the timestamps and location information of the data. This information is subsequently used for analysis and modeling.

[0102] The data acquisition module utilizes lidar and multispectral imaging technologies to collect surface information of the mining area. LiDAR captures topographic elevation data, while multispectral imaging acquires reflectance information of ground features (such as vegetation and soil type) across different spectral bands. Assuming a mine has significant topographic relief and complex vegetation cover, lidar can generate accurate topographic elevation maps, while multispectral data captures the spectral characteristics of different surface materials, such as the infrared reflectance intensity of green vegetation and the visible light reflectance intensity of rocks. LiDAR: A technology that measures the distance and shape of a target object by emitting laser light and receiving the reflected signal. Multispectral imaging: A technology that acquires target information across multiple spectral bands (such as visible light and infrared light).

[0103] The stereo image synthesis module fuses LiDAR data and multispectral data, using 3D modeling technology to generate realistic 3D images of the mining area. The fused image contains both high-precision 3D topographical structure information and rich surface material and spectral attributes. The system uses algorithms to spatially align LiDAR elevation point cloud data with multispectral ground feature reflectance data, generating a 3D model with realistic material textures. For example, it shows the true shape of the mine slope and whether there is exposed soil or lack of vegetation on the slope surface. Data fusion: Combining data from multiple sources to generate more comprehensive information. Point cloud data: A large number of 3D coordinate points generated by LiDAR scanning, representing the shape of the object's surface.

[0104] Geographical Features and Monitoring Needs Information: First, basic information about the mining area needs to be collected, including topography, climate, soil and rock types, mining methods, etc. (these are "geographical features"), as well as potential problems such as soil and rock cracking and collapse that may occur during mining (these are "monitoring needs"). First Orbit Planning Strategy: Based on the geographical features and monitoring needs, multiple satellite orbit schemes are designed (i.e., "first orbit planning strategies"). These strategies are set according to the geographical characteristics of the mining area and the coverage of satellite orbits. Feature-Performance Value Library and Performance Values: For each orbit planning strategy, there is a corresponding feature-performance value library. This library contains the relationship between different orbit strategies and monitoring needs, and the performance values ​​represent the effectiveness of these orbit strategies for specific monitoring tasks. For example, a certain orbit strategy is particularly suitable for monitoring cracking but not for collapse. Second Monitoring Needs Items: The monitoring needs information is broken down into smaller, more specific needs items. For example, crack monitoring can be refined into smaller needs such as "surface crack width" and "crack development speed." Feature Extraction and Third Feature: For each requirement, features related to these requirements, such as "crack morphology" or "soil subsidence," are extracted through analysis of remote sensing data. These features help determine whether specific monitoring requirements are met. Performance Value Calculation and Threshold Determination: Based on the feature-performance value library, the performance value corresponding to each requirement is calculated and accumulated. If the accumulated value exceeds a certain threshold, the requirement is considered met under the orbital strategy. Fit Value Calculation: Using the fit performance model, the fit value of each orbital strategy under specific requirements is calculated. The fit value represents the actual performance of the orbital strategy; a higher value indicates better adaptability to the monitoring task. Final Orbit Planning: All orbital strategies are traversed, and the orbital strategy with the highest fit value is selected as the final second orbital planning strategy. Based on this strategy, the optimal satellite orbit is designed to ensure comprehensive coverage of the mining area, with special consideration given to the monitoring requirements of cracking and local collapse.

[0105] The beneficial effects of the above technical solutions are as follows: The orbit planning module ensures comprehensive coverage of the target area while reducing redundant observations and improving monitoring efficiency. For mining areas with complex terrain and variable weather conditions, this module can dynamically adjust the observation plan to ensure the timeliness and completeness of data acquisition. The data acquisition triggering module, through an automated triggering mechanism, quickly activates the remote sensing equipment when the satellite reaches the target area, ensuring the accuracy and spatiotemporal consistency of data acquisition. Time and location recording facilitates subsequent data calibration and analysis. The data acquisition module provides multi-dimensional surface information through lidar and multispectral imaging technology. It can not only generate high-precision terrain data but also acquire the spectral characteristics of vegetation, soil, and water bodies in the mining area, providing comprehensive basic data for mine safety and ecological environment assessment. The three-dimensional real-scene image data generated by the stereo image synthesis module contains both terrain structure and real ground material, which can intuitively reflect the current state of the mine. This image is not only suitable for mine monitoring but can also be used for visualization, disaster early warning, and ecological restoration planning.

[0106] In another embodiment, the data analysis unit includes:

[0107] The data receiving module is used to receive real-scene 3D data and extract the initial analysis object from the real-scene 3D data;

[0108] The initial analysis determination module is used to determine the analysis results for the initial analysis object. The analysis results include illegal production and construction activities in the mine, local potential risk points, and corresponding types of safety hazards.

[0109] The second analysis determination module is used to identify and classify multiple new potential analysis objects based on the analysis results, and determine the analysis results for each new potential analysis object until the number of consecutive local potential risk points identified in the new analysis results reaches a preset number threshold, or the current analysis object is the last analysis object.

[0110] The analysis results report generation module is used to generate a comprehensive analysis results report based on each analysis result and the corresponding safety hazard type, and output the report content, including a list of illegal production and construction activities in the mine, details of local potential risk points, and a classification description of the safety hazard type.

[0111] The working principle of the above technical solution is as follows: The data receiving module receives externally input real-scene 3D data. Real-scene 3D data refers to real-scene 3D model data generated through technologies such as remote sensing, laser scanning, and drones, including information such as terrain, landforms, and buildings. The module preprocesses this data to extract initial analysis objects. For example, from a 3D mine scene, key points of the mining area, mining equipment, slope morphology, and other potentially risky targets are extracted as analysis objects. If the 3D model of the mine scene shows a slope angle exceeding the design safety standard, then that area becomes the initial analysis object.

[0112] The initial analysis module analyzes the initial analysis object and generates analysis results, including: illegal production and construction activities: such as mining beyond the planned area or unregistered new facilities. Local potential risk points: such as unstable slope areas or equipment overload operation. Types of safety hazards: classified according to standards, such as slope collapse, subsidence, and equipment failure. If unapproved extended mining activities are detected in a mining area, and the slope angle near that area is close to the critical value, it is marked as a "high-risk area".

[0113] The second analysis and determination module, based on the analysis results, further identifies and classifies new potential analysis objects, and repeats the above analysis process. This module continues to explore areas or objects associated with the initial risk until one of the following conditions is met:

[0114] The number of consecutive occurrences reaches a preset threshold: for example, 5 risk points are identified.

[0115] Object exhaustive: All relevant objects have been analyzed.

[0116] If the initial risk point is a slope, the system may further analyze the surrounding drainage facilities, transportation roads, mining equipment, etc., and discover multiple hidden dangers related to the slope.

[0117] The analysis results report generation module summarizes all analysis results and generates a comprehensive analysis results report in a structured format, including: a list of illegal production activities (e.g., unauthorized equipment installation or area expansion); details of local potential risk points (including location and impact range); and a classification of safety hazard types (e.g., "Slope stability risk: 3 locations," "Equipment operation risk: 2 locations").

[0118] The output report is provided for user review or further decision-making. The report may indicate that the eastern slope of the mine is unstable and poses a landslide risk, or that aging equipment in the southwest is causing transportation disruptions.

[0119] The beneficial effects of the above technical solution are as follows: Based on 3D data, it enables high-precision analysis of mine scenes, which is more efficient than traditional manual inspections. It can automatically identify multiple potential risk points, avoiding overlooking hidden dangers. Automatic classification and gradual expansion of the analysis area ensure comprehensive coverage of mine risks. By setting threshold conditions, the analysis depth can be dynamically adjusted to optimize resource utilization.

[0120] In another embodiment, the monitoring and early warning unit includes:

[0121] The first early warning module is used to determine safety hazards based on the analysis results and obtain the safety hazard determination results;

[0122] The early warning information generation module generates early warning information for the current type of safety hazard based on the safety hazard assessment results. The types of safety hazards include rock and soil cracking and local collapse.

[0123] The working principle of the above technical solution is as follows: The first early warning module, based on the analysis results generated by the previous module, comprehensively judges the safety hazards in the mining scenario and generates a safety hazard judgment result. Safety hazard judgment refers to a detailed analysis of the risk based on the type of hazard, its probability of occurrence, and possible consequences. For example, using a 3D data model combined with mechanical analysis, it determines whether a slope collapse is likely. Hazard judgment criteria are typically based on industry standards, historical data, and real-time monitoring parameters (such as slope inclination angle, crack width, etc.). Assuming that multiple new cracks appear on a slope in a certain area, and the crack width gradually increases, the module, combined with a geomechanical model, judges that there is a high probability of local collapse in this area, generating a "high-risk" judgment result.

[0124] The early warning information generation module generates early warning information based on the judgment results of the first early warning module. Early warning information refers to risk alerts conveyed to managers or relevant personnel, including the type of hazard, its location, scope of impact, and recommended measures. Types of safety hazards mainly include: Rock and soil cracking: Cracks appear in the rock and soil on the surface or slope due to stress, which may further develop into landslides or collapses. Localized collapse: Sudden instability of the rock and soil mass or structure leads to the collapse of a partial area. For example, for the identified rock and soil cracking hazard, the module generates the following early warning information: "Warning: Cracks have appeared on the north slope, with a depth of 1.5 meters. It is recommended to immediately arrange personnel to inspect and take support measures."

[0125] The beneficial effects of the above technical solution are as follows: The module identifies potential hazards and generates early warnings in real time, reducing reaction time before accidents occur and helping managers take swift action. Early intervention keeps hazards within a manageable range, preventing further escalation of accidents. It provides information on hazard type, specific location, and severity, reducing ambiguity and improving early warning accuracy. It provides targeted early warnings specifically for common issues such as "rock and soil cracking" and "local collapse." Automated early warning information generation reduces manual intervention and improves the efficiency of mine safety management. The early warning information is clear and structured, facilitating subsequent analysis and decision-making.

[0126] In another embodiment, the governance management unit includes:

[0127] The early warning assessment module is used to assess the severity of the type of safety hazard based on the current early warning information;

[0128] The governance plan formulation module is used to determine the necessary reinforcement or support measures when the early warning information indicates the potential for cracking in the soil and rock mass, in conjunction with the mine design data.

[0129] When the early warning information indicates a potential for localized collapse, corresponding protection or repair plans should be formulated in accordance with the development specifications.

[0130] The governance scheme optimization module is used to optimize the preliminary governance scheme by comprehensively considering the requirements of mine design data and development specifications. If multiple safety hazards in the early warning information exist at the same time, the governance scheme optimization module will comprehensively formulate integrated governance measures to improve the overall governance effect.

[0131] The governance plan generation module generates corresponding mine governance plans based on the optimized governance measures. The governance plan includes specific implementation steps, required materials, time schedule and expected results.

[0132] The working principle of the above technical solution is as follows: the early warning assessment module assesses the severity of safety hazards based on the early warning information, providing a basis for the formulation of a treatment plan.

[0133] Early warning assessment module operation: The module reads the hazard type from the early warning information (such as "rock and soil cracking" or "local collapse"). Combining hazard parameters (such as crack width, depth, propagation rate, or collapse area), it calculates the severity level (such as minor, moderate, or severe) using industry standards or experience models. If the slope crack width is 2 cm and the length is more than 20 meters, the module assesses it as a "severe" hazard and passes it to the next module.

[0134] The remediation plan formulation module develops preliminary remediation measures based on the type of hazard and mine design data. For hazard of soil and rock cracking: It reads mine design data (such as geological structure and slope angle) and suggests reinforcement or support measures, such as anchor bolt reinforcement or shotcrete. For hazard of localized collapse: Based on development specifications, it suggests protection (such as installing protective netting) or repair (such as backfilling and reinforcement) solutions. If a localized collapse hazard is identified, the module generates a preliminary solution: "It is recommended to cover the collapsed area with steel mesh, and simultaneously use concrete for backfilling and support."

[0135] The remediation plan optimization module optimizes the initially formulated remediation plan to improve its effectiveness. It adjusts the details of the plan by comprehensively considering mine design data and development specifications. When multiple hazards coexist: based on the relationship between the hazards (e.g., cracking leading to collapse), a joint remediation plan is developed to reduce redundant remediation and improve resource utilization efficiency. If a slope has both cracks and collapse risks, the optimization module suggests: first, anchor the cracks, then reinforce the collapse area with concrete to ensure the remediation effect.

[0136] The governance plan generation module generates specific mine governance plans based on the optimized measures.

[0137] The remediation plan generation module includes the following details: Implementation steps: such as the installation sequence of support materials. Required materials: such as anchor bolt specifications and concrete quantity. Time schedule: such as daily construction volume and construction period. Expected results: such as data on the improvement in slope stability after support. For example, a specific description of a remediation plan: "In the first stage, 200 anchor bolts with a diameter of 25mm are used to fix the cracked area, which is expected to take 5 days. In the second stage, high-strength concrete is used to cover the collapsed area, which is expected to improve stability by 95%."

[0138] The beneficial effects of the above technical solutions are as follows: The early warning assessment module, based on data and standards, quantitatively assesses potential hazards, providing a scientific basis and avoiding subjective human judgment. The remediation plan formulation module can tailor the most suitable preliminary remediation plan according to the type of hazard and mine conditions, avoiding waste from one-size-fits-all or over-remediation. The remediation plan optimization module formulates comprehensive remediation plans when multiple hazards coexist, effectively reducing remediation conflicts and improving the efficiency of remediation resource utilization and overall effectiveness. The remediation plan generation module outputs a plan that includes specific steps, materials, and timelines, providing clear guidance for construction personnel and ensuring the implementation of remediation measures.

[0139] In another embodiment, the safety hazard assessment includes:

[0140] If the illegal production and construction activities of a mine exceed the illegal production threshold but are less than the preset illegal production limit, local potential risk points are identified.

[0141] If a local potential risk point exceeds the potential risk point threshold but is less than the preset risk point limit, a safety hazard judgment result is generated.

[0142] The working principle of the above technical solution is as follows: Illegal production threshold: a standard value used to measure whether mine production has reached the initial stage of violation. Preset illegal production limit: a higher standard value indicating the severity of the illegal production behavior.

[0143] Localized potential risk points are calculated based on production data or behavioral patterns in specific areas of a mine and serve as indicators to measure the potential safety risks in those areas. For example, if the mining depth in a certain area exceeds the regulations, the equipment aging rate is high, or there are non-compliant behaviors in personnel operation records, it may be identified as a localized potential risk point.

[0144] The judgment process consists of two steps: First, real-time monitoring of the mine's production and construction data to identify whether it exceeds the threshold for illegal production. For example, a mine's daily output may not exceed 10,000 tons. If the actual output is 12,000 tons, it exceeds the threshold but does not reach the preset illegal production limit of 20,000 tons, triggering the next step of analysis. Second, under the premise of illegal production, analyzing whether the local potential risk points in specific areas exceed the set potential risk point threshold. For example, the risk point threshold for a certain area of ​​the mine may be 80 points (based on indicators such as equipment aging and mining depth). However, if the area scores 90 points (less than the 100-point risk point limit), then the area is determined to have potential safety hazards.

[0145] The system generates safety hazard assessment results, combining the analysis of violations and local risk points to produce a systematic safety hazard report. For example, if two areas both have potential risk points, the system will prioritize these two areas and generate a risk level assessment report for decision-makers' reference.

[0146] The beneficial effects of the above technical solution are as follows: By setting clear thresholds and limits, the risk management process of the mine is systematized and standardized, thereby reducing the bias of subjective human judgment. The system can capture illegal production and potential risk points in real time, and issue timely warnings to prevent the further expansion of safety hazards. For example, in the early identification of collapse risks in high-depth mining areas, reinforcement measures can be taken. Through the generated safety hazard reports, managers can prioritize the handling of the highest-risk areas, saving manpower and equipment resources. Intervening at the hazard stage effectively avoids the occurrence of major safety accidents. For example, timely handling of aging equipment areas prevents catastrophic consequences caused by mechanical failure.

[0147] In another embodiment, generating early warning information for the current type of security hazard includes:

[0148] Obtain the severity level associated with each hazard type, and use it as the data group to be analyzed;

[0149] Based on the type of hazard, each data group to be analyzed is sorted, and the data corresponding to each hazard type and severity level is stored in the corresponding data group to be analyzed to obtain the hazard data to be analyzed.

[0150] Obtain safety hazard early warning standard data, and extract early warning rule data from the safety hazard early warning standard data;

[0151] After noise removal processing of the early warning rule data, rule feature vectors are constructed, and standard correlation data are obtained from the safety hazard early warning standard data.

[0152] After clustering the rule feature vectors based on the standardized associated data, a training data set is obtained. The initial model is then trained using the training data set through question-and-answer training to obtain the early warning analysis model.

[0153] The data of potential hazards to be analyzed are input one by one into the preset early warning analysis model to generate early warning information of the current safety hazards;

[0154] Generate tiered early warning information based on the type and severity level of safety hazards;

[0155] The tiered early warning information is input into the early warning analysis model and integrated to generate the final monitoring and early warning information data.

[0156] Clustering of regular feature vectors includes:

[0157]

[0158] J represents minimizing the sum of squared distances from each point to the center of its cluster, n represents the total number of data points, K represents the number of clusters, and r ik The indicator function representing whether data point i belongs to cluster k (1 if it belongs, 0 otherwise), x i Represents data points, μ k This represents the center of cluster k.

[0159] The working principle of the above technical solution is as follows: Obtain the type of hazard (such as soil and rock cracking, local collapse, etc.) and associate it with the severity level (such as low, medium, high). Sort the hazard data for each type according to the hazard type (e.g., by frequency of occurrence or historical data). Store the sorted data in the corresponding hazard type analysis group to form the hazard data to be analyzed.

[0160] The early warning standards and rules processing involves acquiring safety hazard early warning standard data, including a series of rule data for monitoring and assessment. Noise removal (eliminating redundant or abnormal information) is performed on the rule data, key rule features are extracted, and rule feature vectors are constructed. Based on the characteristics of the hazard type and the rule data, standard-related data (such as the impact of different environments and materials) is obtained.

[0161] The process involves clustering regular feature vectors using standardized correlated data to form training datasets. The initial model is then trained using a question-and-answer approach (simulating real-world questions and answers to improve the model's understanding of different potential hazards), resulting in a more accurate early warning analysis model.

[0162] The process involves inputting hazard data and generating early warning information. Hazard data to be analyzed (such as frequency of soil and rock cracking, crack width variations, and locations of localized collapses) is input one by one into the early warning analysis model. The model generates early warning information for the current hazard based on the input data. By considering the hazard type and severity level, a tiered early warning system is generated (e.g., a red alert for high-risk areas).

[0163] The system integrates and generates monitoring and early warning information, then inputs the tiered early warning information back into the early warning analysis model. Through integrated processing, it generates the final monitoring and early warning data. The output information can be used for real-time monitoring, alarms, and the development of response measures. For example, it can be used for cracking and localized collapse of rock and soil in mines.

[0164] Hazard Type Analysis: The severity of soil and rock cracking is categorized into low (initial cracks, no extension), medium (slow crack extension), and high (rapid crack extension or associated collapse) based on crack width, depth, and propagation speed. Regular Feature Vector: Features are constructed based on historical data from the mining area, such as the probability of crack extension during the rainy season, soil strength, and surface water level changes. Clustering Training: Crack data is grouped and trained according to extension speed and frequency, enabling the model to determine the risk level of similar situations. Early Warning Information Generation: Inputting real-time monitoring data, such as crack dynamics and vibration sensor data, generates tiered early warning information (e.g., "Cracks have extended after recent heavy rains; reinforcement is recommended").

[0165] The beneficial effects of the above technical solution are as follows: It classifies, sorts, grades, and clusters complex hazard data, achieving systematic management of hazard information and helping enterprises quickly grasp the status of safety hazards. Through noise processing and the construction of rule-based feature vectors, it reduces false alarms and missed alarms, ensuring high accuracy of early warning information. It generates multi-level early warning information based on real-time input hazard data, supporting dynamic decision-making. Using question-and-answer training, it continuously optimizes the early warning analysis model, adapting it to more types of hazard scenarios and complex risk environments. Tiered early warning information allows enterprises to rationally allocate resources. High-risk areas are prioritized, while medium- and low-risk areas are subject to routine monitoring, saving manpower and resources.

[0166] In another embodiment, generating early warning information about current security risks includes:

[0167] The first potential hazard data to be analyzed is input into the early warning analysis model. The first potential hazard data to be analyzed is then matched for similarity in the early warning analysis model. Based on the matching results, the first early warning result is output.

[0168] Starting with the second potential hazard data to be analyzed, the previous warning result is input into the warning analysis model to obtain the corresponding associated warning results. The first warning result and all associated warning results are then integrated to generate warning information for the current type of safety hazard.

[0169] The working principle of the above technical solution is as follows: Inputting and Preliminary Early Warning of Individual Hazardous Data. The first hazard data (such as a record of increased crack width in mine soil and rock) is input into the early warning analysis model. This data is typically collected by sensors and monitoring equipment (such as displacement gauges or crack gauges). The model calculates the similarity between the input data and a hazard feature library through feature analysis. The feature library contains features of common hazards, such as "crack width change rate ≥ 1 mm / d, which may indicate local instability." For example, if the crack width increases from 1 mm to 5 mm, the matching result shows: "Crack expansion may induce local collapse, matching degree 90%." Outputting the First Early Warning Result: The model generates a preliminary early warning based on the matching result. For example: "Significant expansion of cracks in soil and rock, which may indicate instability risk; enhanced monitoring is recommended."

[0170] Iterative analysis of correlated hazard data begins with the second hazard data point: when new hazard data is input (e.g., increased local settlement of soil and rock mass), the previous warning result (risk of crack propagation) is input into the model as additional information. Correlation warning result generation: the model comprehensively analyzes the new data with the previous warning result. For example, "crack propagation + significant increase in settlement" may indicate a more serious hazard (e.g., local instability or collapse). Example output: "Crack propagation in soil and rock mass accompanied by increased settlement, with a high overall risk level, may induce local collapse."

[0171] Integrate early warning information, combining the initial early warning result (risk of crack propagation) with subsequent related early warning results (risk of instability caused by settlement) to generate the final early warning information. Generate early warning reports for specific hazard types: Comprehensively analyze all data to generate unified risk assessment reports and recommendations for specific hazard types (such as instability of mine rock and soil). For example: "Crack propagation (5mm) and settlement (15cm) in mine rock and soil are intensifying simultaneously, with a high probability of instability. It is recommended to stop operations and initiate emergency reinforcement measures."

[0172] First hazard data (crack expansion):

[0173] Input: The crack width expands from 1 mm to 5 mm at a rate of 1 mm / d.

[0174] Similarity matching:

[0175] The match rate with "rapidly expanding cracks may lead to local collapse" in the known hazard feature database is 90%.

[0176] Warning output: Preliminary results: "Cracks are spreading rapidly, risk level: medium, real-time monitoring is recommended."

[0177] Second hidden danger data (increased settlement):

[0178] Input: The settlement of the soil and rock monitoring point increased from 0 to 15cm, with a change rate of 3cm / d.

[0179] Correlation analysis: Based on the previous result (crack propagation), the model infers that settlement and crack propagation may be manifestations of the same hidden danger process. The output correlation warning is: "Intensified settlement and crack propagation may trigger local instability; risk level: high."

[0180] Integrated Result: Comprehensive Early Warning Information: "The expansion of cracks in the mine's rock and soil (5mm / d) is accompanied by increased settlement (15cm), and the risk level of instability is high. It is recommended to stop operations and implement reinforcement measures."

[0181] The beneficial effects of the above technical solution are as follows: By progressively analyzing hazard data, the model can dynamically track the development trend of hazards. Correlation analysis of multiple hazard indicators (such as crack propagation, settlement changes, and tilt angle changes) can uncover more complex safety issues. Based on similarity matching and correlation analysis, the early warning information generated by the model is more accurate. Comprehensive early warning information provides clear recommendations (such as reinforcement, cessation of work, and evacuation of personnel), helping managers make quick decisions and prevent disasters.

[0182] In another embodiment, graded early warning information is generated based on the type and severity level of the safety hazard, including:

[0183] When the hazard type is rock and soil cracking and the severity level is low, a primary warning message is triggered;

[0184] When the hazard type is rock and soil cracking and the severity level is high, a medium-level early warning information is triggered;

[0185] When the type of hazard is local collapse and the severity level is low, an intermediate warning message is triggered;

[0186] When the hazard type is local collapse and the severity level is high, an advanced early warning information is triggered.

[0187] The working principle of the above technical solution is as follows: The early warning system first identifies the type of hidden danger based on the input data, such as "rock and soil cracking" or "local collapse". Rock and soil cracking: Cracks appear on the surface or inside the rock and soil mass, which is usually an early manifestation of instability. Local collapse: Local collapse of the rock and soil mass, which is usually a subsequent result of cracking, settlement, etc.

[0188] Severity Level: The system assesses the severity of potential hazards based on real-time data, such as crack width, settlement, and collapse volume. Severity is typically categorized as low (no immediate danger) or high (immediate danger exists).

[0189] The system sets different warning levels based on the type and severity of the hazard: Rock and soil cracking (low level): Triggers a basic warning, indicating a potential hazard. Rock and soil cracking (high level): Triggers a medium warning, indicating a risk of accelerated deterioration. Localized collapse (low level): Triggers a medium warning, indicating a possible danger. Localized collapse (high level): Triggers a high-level warning, indicating immediate action is required.

[0190] Data processing and evaluation workflow: Monitoring data input: Sensors or devices collect real-time monitoring data, such as crack width, settlement, or collapse volume. The model identifies the type of hazard based on a feature library, for example, "cracks extending to 3mm are identified as soil and rock cracking." The severity is assessed based on preset thresholds: Soil and rock cracking: Crack width < 5mm: Low level. Crack width ≥ 5mm: High level. Local collapse: Collapse volume < 2 cubic meters: Low level. Collapse volume ≥ 2 cubic meters: High level. Early warning trigger: The system triggers corresponding early warning information according to rules.

[0191] For example: Cracking of soil and rock mass, with relatively low severity:

[0192] Input data: Crack width is 2mm.

[0193] Hazard identification: determined to be "cracks in the soil and rock mass".

[0194] Severity assessment: Width < 5mm, level is low.

[0195] Warning triggered: Basic warning, message: "Risk of crack propagation, requires regular monitoring."

[0196] The soil and rock mass is cracked, and the degree of cracking is relatively high.

[0197] Input data: Crack width is 8mm.

[0198] Hazard identification: determined to be "cracks in the soil and rock mass".

[0199] Severity assessment: Width ≥ 5mm, level is high.

[0200] Warning triggered: Intermediate warning, message: "Cracks are expanding significantly, posing a risk of local instability."

[0201] Localized collapse, relatively minor:

[0202] Input data: The collapse volume is 1 cubic meter.

[0203] Hazard identification: determined to be "partial collapse".

[0204] Severity assessment: Volume < 2 cubic meters, level is low.

[0205] Warning triggered: Intermediate warning, message: "Local collapse phenomenon, protective measures need to be strengthened."

[0206] Localized collapse, with a high degree of severity:

[0207] Input data: The collapse volume is 3 cubic meters.

[0208] Hazard identification: determined to be "partial collapse".

[0209] Severity assessment: Volume ≥ 2 cubic meters, level is high.

[0210] Warning triggered: Advanced warning, message: "Severe local collapse, it is recommended to stop work and carry out emergency reinforcement."

[0211] The beneficial effects of the above technical solution are as follows: Combining hazard type with severity to trigger different levels of early warning information helps clarify management levels. Severity classification reduces overreaction to minor hazards and avoids neglecting major hazards. Different early warning levels correspond to different handling measures. Through tiered early warning information, decision-makers can quickly match appropriate response strategies. By classifying hazard types and severity, the system can more accurately identify the development stage of hazards.

[0212] In another embodiment, the initially formulated governance plan is optimized, including:

[0213] Several potential safety hazards were identified, including risks such as cracking of the mine's rock and soil and local collapses;

[0214] At the moment of each warning message, identify the types of existing safety hazards and analyze the relationships between the hazards;

[0215] Based on the mutual influence between potential hazards, the preliminary remediation plan was optimized and the details of the remediation measures were adjusted.

[0216] For situations where multiple safety hazards exist simultaneously, comprehensive management measures should be formulated to improve the overall management effectiveness, including:

[0217] When there is a risk of soil and rock cracking and local collapse, the optimized treatment plan includes prioritizing anchor bolt support measures based on the cracking risk and implementing concrete reinforcement schemes based on the collapse risk to ensure the treatment effect.

[0218] Based on the relationships between different potential hazards, and taking into account the severity of the hazards, risk assessment results, and technical requirements for remediation, a joint remediation plan should be developed to avoid redundant remediation and improve resource utilization efficiency.

[0219] The remediation plan is adjusted in real time based on the actual monitoring data and remediation results of each potential hazard to ensure the coordination and effectiveness of various measures.

[0220] In comprehensive management measures, the corresponding support structure, reinforcement technology and protective materials are adjusted according to the spatiotemporal characteristics of different hazards to ensure the feasibility and economy of the management plan.

[0221] The relationships between different potential hazards are quantitatively analyzed using the following formula:

[0222]

[0223] R represents the comprehensive hazard risk assessment value, ω i R represents the weight value of hazard i (e.g., the weight of collapse caused by cracking is higher than that of simple settlement). i This represents the individual risk assessment value of hazard i (such as specific indicators like crack propagation rate and collapse impact range), and n represents the total number of hazard types. This formula is used to perform weighted calculations on different hazards to determine the priority of each hazard in the comprehensive management plan.

[0224] The optimized governance plan is based on the following steps:

[0225] Acquire monitoring data of each current hidden danger, and combine the interaction between hidden dangers (such as the risk of collapse caused by crack propagation) to calculate the comprehensive treatment effect of the current hidden danger using a preset dynamic model;

[0226] Based on the calculated comprehensive treatment effect, the specific parameters of the treatment measures of the anchor bolt support and reinforcement scheme are adjusted to form the final optimized treatment scheme;

[0227] If the actual effect of the optimization plan deviates from the expected effect, the plan will be further optimized based on dynamic monitoring data and real-time feedback to form a closed-loop feedback mechanism to improve the governance effect.

[0228] The working principle of the above technical solution is as follows: real-time data of the mine's rock and soil are collected using monitoring equipment (such as surface deformation sensors, inclinometers, etc.) to identify potential safety hazards. For example, if the width of cracks on the surface of the rock and soil reaches a critical value, or if the rock strata tilt beyond the stable range, the potential for cracking or collapse can be identified.

[0229] Based on monitoring data and the interactions between potential hazards, pre-set dynamic models (such as crack propagation models and rock mass stability models) are used to assess the impact of remediation measures on overall risk. Preliminary remediation measures are designed for different hazards; for example: cracking risk: anchor bolt support is prioritized to enhance structural integrity. Local collapse risk: concrete reinforcement is used to stabilize high-risk areas.

[0230] After implementing remediation measures, the effectiveness is monitored, and the remediation plan is adjusted based on real-time data. If the effect deviates from expectations, such as if the cracking rate is not significantly reduced, the influencing factors are analyzed, and parameters such as the number of anchor bolts or the amount of concrete are adjusted to form a closed-loop feedback mechanism. For multiple potential hazards, joint remediation plans are developed to avoid redundant remediation. For example, when both cracking and collapse risks exist simultaneously, unified support and reinforcement measures are implemented to ensure safety while improving resource utilization efficiency.

[0231] Based on the dynamic changes of potential hazards in time and space, optimize support and reinforcement schemes. For example, if cracks are mainly concentrated in a certain mining section, increase the anchoring density in that area. As the risk of collapse increases over time, a phased treatment strategy can be adopted to gradually expand the treatment scope.

[0232] The beneficial effects of the above technical solutions are as follows: They systematically analyze and identify potential hazards, and determine remediation priorities through weighted calculations, ensuring timely and effective remediation of high-risk areas and reducing the likelihood of mine accidents. Through joint remediation schemes, they avoid redundant construction and resource waste. For example, the optimized anchor bolt support and concrete reinforcement schemes address both cracking and collapse hazards. A closed-loop feedback mechanism ensures that the remediation scheme is adjusted in real time based on monitoring data, adapting to the complex changes in the mining environment and ensuring that remediation measures remain highly effective.

[0233] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A monitoring system for the treatment of cracking and local collapse of rock and soil in mines based on satellite remote sensing, characterized in that, include: Satellite remote sensing units are used to conduct full-range three-dimensional monitoring of mining areas and acquire real-world three-dimensional data of the mining areas; The data analysis unit is used to analyze real-world 3D data and obtain analysis results, including illegal production and construction activities in the mine, local potential risk points, and types of safety hazards. The monitoring and early warning unit is used to generate early warning information for the current type of safety hazard based on the analysis results. The types of safety hazards include rock and soil cracking and local collapse. The monitoring and early warning unit includes: an early warning information generation module, which generates early warning information for the current type of safety hazard based on the safety hazard judgment results. The types of safety hazards include rock and soil cracking and local collapse. The early warning information generation module is further used to: obtain the severity level associated with each type of hazard as a data group to be analyzed; Based on the type of hazard, each data group to be analyzed is sorted, and the data corresponding to each hazard type and severity level is stored in the corresponding data group to be analyzed to obtain the hazard data to be analyzed. Obtain safety hazard early warning standard data, and extract early warning rule data from the safety hazard early warning standard data; After noise removal processing of the early warning rule data, rule feature vectors are constructed, and standard correlation data are obtained from the safety hazard early warning standard data. After clustering the rule feature vectors based on the standardized associated data, a training data set is obtained. The initial model is then trained using the training data set through question-and-answer training to obtain the early warning analysis model. The data of potential hazards to be analyzed are input one by one into the preset early warning analysis model to generate early warning information of the current safety hazards; Generate tiered early warning information based on the type and severity level of safety hazards; The tiered early warning information is input into the early warning analysis model and integrated to generate the final monitoring and early warning information data; The generation of early warning information for current safety hazards includes: The first potential hazard data to be analyzed is input into the early warning analysis model. The first potential hazard data to be analyzed is then matched for similarity in the early warning analysis model. Based on the matching results, the first early warning result is output. Starting with the second potential hazard data to be analyzed, the previous warning result is input into the warning analysis model to obtain the corresponding associated warning results. The first warning result and all associated warning results are then integrated to generate warning information for the current type of safety hazard. The governance and management unit is used to formulate corresponding mine governance plans based on current early warning information, combined with mine design data and development specifications.

2. The monitoring system for mine rock and soil cracking and local collapse control based on satellite remote sensing as described in claim 1, characterized in that, Satellite remote sensing units include: The orbit planning module is used to plan the optimal observation orbit for satellites based on the geographical features and monitoring needs of the mining area, so as to ensure comprehensive coverage of the mining area; The data acquisition triggering module is used to synchronously trigger the data acquisition process when the satellite passes through the mining area based on the orbit planning results, and to record the acquisition time and location information in real time. The data acquisition module is used to acquire lidar data and multispectral data after the data acquisition process is triggered; The stereo image synthesis module is used to fuse lidar data and multispectral data based on 3D modeling technology to generate real-world 3D image data of the mining area.

3. The monitoring system for mine rock and soil cracking and local collapse control based on satellite remote sensing as described in claim 1, characterized in that, The data analysis unit includes: The data receiving module is used to receive real-scene 3D data and extract the initial analysis object from the real-scene 3D data; The initial analysis determination module is used to determine the analysis results for the initial analysis object. The analysis results include illegal production and construction activities in the mine, local potential risk points, and corresponding types of safety hazards. The second analysis and determination module is used to identify and classify multiple new potential analysis objects based on the analysis results, and determine the analysis results for each new potential analysis object until the number of consecutive local potential risk points identified in the new analysis results reaches a preset number threshold, or the current analysis object is the last analysis object. The analysis results report generation module is used to generate a comprehensive analysis results report based on each analysis result and the corresponding safety hazard type, and output the report content, including a list of illegal production and construction activities in the mine, details of local potential risk points, and a classification description of the safety hazard type.

4. The monitoring system for mine rock and soil cracking and local collapse control based on satellite remote sensing as described in claim 1, characterized in that, The monitoring and early warning unit also includes: The first early warning module is used to identify safety hazards based on the analysis results and obtain the results of the safety hazard assessment.

5. The monitoring system for mine rock and soil cracking and local collapse control based on satellite remote sensing as described in claim 1, characterized in that, The governance and management unit includes: The early warning assessment module is used to assess the severity of the type of safety hazard based on the current early warning information; The governance plan formulation module is used to determine the necessary reinforcement or support measures when the early warning information indicates the potential for cracking in the soil and rock mass, in conjunction with the mine design data. When the early warning information indicates a potential for localized collapse, corresponding protection or repair plans should be formulated in accordance with the development specifications. The governance scheme optimization module is used to optimize the preliminary governance scheme by comprehensively considering the requirements of mine design data and development specifications. If multiple safety hazards in the early warning information exist at the same time, the governance scheme optimization module will comprehensively formulate integrated governance measures to improve the overall governance effect. The governance plan generation module generates corresponding mine governance plans based on the optimized governance measures. The governance plan includes specific implementation steps, required materials, time schedule and expected results.

6. The monitoring system for mine rock and soil cracking and local collapse control based on satellite remote sensing according to claim 4, characterized in that, Safety hazard assessment includes: If the illegal production and construction activities of a mine exceed the illegal production threshold but are less than the preset illegal production limit, local potential risk points are identified. If a local potential risk point exceeds the potential risk point threshold but is less than the preset risk point limit, a safety hazard judgment result is generated.

7. The monitoring system for mine rock and soil cracking and local collapse control based on satellite remote sensing as described in claim 1, characterized in that, Graded early warning information is generated based on the type and severity level of safety hazards, including: When the hazard type is rock and soil cracking and the severity level is low, a primary warning message is triggered; When the hazard type is rock and soil cracking and the severity level is high, a medium-level early warning information is triggered; When the type of hazard is local collapse and the severity level is low, an intermediate warning message is triggered; When the hazard type is local collapse and the severity level is high, an advanced early warning information is triggered.

8. The monitoring system for mine rock and soil cracking and local collapse control based on satellite remote sensing as described in claim 5, characterized in that, The preliminary governance plan will be optimized, including: At the moment of each warning message, identify the types of existing safety hazards and analyze the relationships between the hazards; Based on the mutual influence between potential hazards, the preliminary remediation plan was optimized and the details of the remediation measures were adjusted. For situations where multiple safety hazards exist simultaneously, comprehensive management measures should be formulated to improve the overall management effectiveness, including: When there is a risk of soil and rock cracking and local collapse, the optimized treatment plan includes prioritizing anchor bolt support measures based on the cracking risk and implementing concrete reinforcement schemes based on the collapse risk to ensure the treatment effect. Based on the relationships between different potential hazards, and taking into account the severity of the hazards, risk assessment results, and technical requirements for remediation, a joint remediation plan should be developed to avoid redundant remediation and improve resource utilization efficiency. The remediation plan is adjusted in real time based on the actual monitoring data and remediation results of each potential hazard to ensure the coordination and effectiveness of various measures. In comprehensive management measures, the corresponding support structure, reinforcement technology and protective materials are adjusted according to the spatiotemporal characteristics of different hazards to ensure the feasibility and economy of the management plan. The optimized governance plan is based on the following steps: Acquire monitoring data for each current hazard, and combine the interactions between hazards to calculate the comprehensive mitigation effect of the current hazard using a pre-set dynamic model; Based on the calculated comprehensive treatment effect, the specific parameters of the treatment measures of the anchor bolt support and reinforcement scheme are adjusted to form the final optimized treatment scheme; If the actual effect of the optimization plan deviates from the expected effect, the plan will be further optimized based on dynamic monitoring data and real-time feedback to form a closed-loop feedback mechanism to improve the governance effect.

Images