Open pit coal mine closed pit disaster multi-dimensional evaluation and prevention system

Abstract

The invention relates to the technical field of disaster risk management, in particular to an open pit coal mine closed pit disaster multi-dimensional evaluation and prevention system which comprises a data acquisition module, a data analysis and processing module, a disaster prediction and early warning module and a prevention measure recommendation module. By collecting multi-source heterogeneous data such as geological structures, hydrological conditions and environmental influences, key disaster-inducing factors are identified, and a disaster development area is delineated; constructing a multi-field coupling numerical model coupling an underground water seepage field, a rock-soil stress field and a pollutant transport field, and predicting a disaster evolution trend; and carrying out space matching and economic optimization in combination with the dynamic risk distribution map to generate an accurate prevention scheme. The system has the capabilities of full-process automation, multi-disaster coupling simulation and intelligent measure recommendation, and is suitable for mine closed pit treatment and disaster early warning scenes.

Description

Technical Field

[0001] This invention relates to the field of disaster risk management technology, and in particular to a multi-dimensional assessment and prevention system for disasters caused by the closure of open-pit coal mines. Background Technology

[0002] With the continuous increase in resource extraction intensity, a large number of open-pit coal mines are gradually entering the closure stage. After the closure of open-pit coal mines, the geological structure is fractured, the hydrological cycle is altered, and the ecosystem is degraded due to long-term disturbance, which can easily induce various secondary disasters such as slope instability, surface subsidence, groundwater inrush, and migration of harmful substances. These disasters not only endanger the lives and property of residents around the mining area, but also cause long-term damage to the ecological environment.

[0003] Existing open-pit coal mine disaster prevention and control technologies generally suffer from the following problems: First, data acquisition methods are limited, making it difficult to achieve unified collection and fusion of multi-dimensional geological, hydrological, and environmental data, resulting in a lack of basis for disaster cause analysis; second, disaster identification relies on experience-based judgment and lacks factor extraction and zoning methods based on historical data and spatial modeling; third, early warning and prediction are mostly based on single-field physical process simulations, failing to reflect the coupled evolutionary relationship between multiple disasters; and fourth, the formulation of preventive measures lacks refined extrapolation and economic optimization mechanisms, often making implementation difficult. Summary of the Invention

[0004] This invention provides a multi-dimensional assessment and prevention system for disasters in closed open-pit coal mines, which is a systematic solution that integrates data collection, factor identification, disaster simulation and measure recommendation, so as to achieve scientific, quantitative and intelligent disaster prevention and control in closed areas.

[0005] A multi-dimensional assessment and prevention system for closure disasters in open-pit coal mines includes a data acquisition module, a data analysis and processing module, a disaster prediction and early warning module, and a prevention measure recommendation module, wherein;

[0006] The data acquisition module collects multi-source heterogeneous data of the target open-pit coal mine closed area and uploads the data to the data analysis and processing module; the multi-source heterogeneous data includes geological structure data, hydrological condition data and environmental impact data.

[0007] The data analysis and processing module receives the multi-source heterogeneous data and uses the built-in algorithm model to fuse and analyze the multi-source heterogeneous data, identify potential key disaster-causing factors, and delineate disaster development zones.

[0008] The disaster prediction and early warning module receives information on key disaster-causing factors and disaster development zones, predicts the disaster evolution trend through numerical simulation, and calculates and generates a dynamic risk distribution map of the entire mining area.

[0009] The preventive measures recommendation module, based on the received dynamic risk distribution map, calls the built-in preventive measures knowledge base for matching, and automatically generates and outputs a spatially accurate prevention plan for each zone.

[0010] Optionally, the data acquisition module includes a geological structure data acquisition unit, a hydrological condition data acquisition unit, an environmental impact data acquisition unit, and a data integration unit.

[0011] Optionally, the data acquisition module includes:

[0012] The geological structure data acquisition unit obtains rock core samples and records the distribution information of rock and soil layers through geological drilling, detects underground geological structures and records the attitude information of structural planes through geophysical methods, scans the surface morphology of open-pit mines through a three-dimensional laser scanner and generates three-dimensional geomorphological information of the mines, and integrates the rock and soil layer distribution information, the attitude information of structural planes and the three-dimensional geomorphological information of the mines into complete geological structure data.

[0013] The hydrological condition data acquisition unit continuously monitors and records groundwater level data through water level gauges installed in monitoring wells, monitors and records pore water pressure data through piezometers buried in slopes and pit bottoms, and collects and records water chemical composition data through a network of multi-parameter water quality sensors deployed at groundwater outlets and surface water bodies. The groundwater level data, pore water pressure data, and water chemical composition data are integrated into complete hydrological condition data.

[0014] The environmental impact data acquisition unit acquires surface deformation interferograms of the mining area through synthetic aperture radar satellites and calculates them into surface deformation data; it acquires vegetation indexes through multispectral UAV aerial photography and inverts them into vegetation coverage data; it obtains heavy metal content and pH value through ground soil sampling and testing and records them as soil pollution data; and it integrates the surface deformation data, the vegetation coverage data and the soil pollution data into complete environmental impact data.

[0015] The data integration unit receives the geological structure data, hydrological condition data, and environmental impact data, and performs format standardization, time synchronization, and spatial registration processing. Finally, it packages the data into multi-source heterogeneous data that the system can recognize and uploads it to the data analysis and processing module.

[0016] Optionally, the data analysis and processing module includes a data preprocessing unit, a disaster-causing factor analysis unit, and a disaster zone delineation unit.

[0017] Optionally, the data analysis and processing module includes:

[0018] The data preprocessing unit receives multi-source heterogeneous data from the data acquisition module, performs data cleaning on the multi-source heterogeneous data to remove outliers and noise, performs format standardization to unify the data structure and measurement units, performs spatiotemporal registration to make all data have the same timestamp and spatial coordinate system, and generates preprocessed multi-source heterogeneous data.

[0019] The disaster-causing factor analysis unit receives the preprocessed multi-source heterogeneous data, calls the principal component analysis algorithm in the built-in algorithm model to perform dimensionality reduction processing on the preprocessed multi-source heterogeneous data to extract the main feature variables, calls the grey relational analysis algorithm to calculate the correlation degree between each feature variable and historical disaster events, and selects the key disaster-causing factor with the largest contribution from all feature variables based on the variance contribution rate and correlation degree ranking results.

[0020] The disaster zone delineation unit receives the key disaster-causing factors, performs spatial interpolation of the measured data or calculated values ​​of each key disaster-causing factor in a three-dimensional geological model to generate disaster factor isosurfaces, uses a weighted superposition analysis method to superimpose and calculate multiple disaster factor isosurfaces, delineates potential disaster development areas exceeding the safety threshold in three-dimensional space according to the set threshold, and outputs the disaster development areas and their corresponding key disaster-causing factor information to the disaster prediction and early warning module.

[0021] Optionally, the disaster prediction and early warning module includes a model building unit, a simulation prediction unit, and a risk mapping unit.

[0022] Optionally, the disaster prediction and early warning module includes:

[0023] The model building unit receives key disaster-causing factors and disaster development zone information from the data analysis and processing module. Based on the physical parameters of the key disaster-causing factors and the spatial geometric characteristics of the disaster development zone, it uses the finite element method to construct a multi-field coupled numerical model that couples the groundwater seepage field, the soil and rock stress field, and the pollutant transport field, and assigns values ​​to the key disaster-causing factors as model input parameters.

[0024] The simulation prediction unit runs the multi-field coupled numerical model to simulate the changes in slope stability, groundwater flow path and pollutant migration and diffusion process in the disaster development area under different working conditions. It predicts the disaster evolution trend in a specific future time period through time-series iterative calculation, including three types of quantitative indicators: landslide failure probability, surface subsidence, and pollutant concentration distribution.

[0025] The risk mapping unit uses the risk matrix analysis method to comprehensively quantify the probability of disaster occurrence and the degree of potential harm based on the three types of quantitative indicators of the disaster evolution trend, generates a risk quantification value with spatiotemporal characteristics, and performs visualization rendering on the three-dimensional geographic information system platform. Finally, it outputs a dynamic risk distribution map of the entire mining area and transmits the dynamic risk distribution map to the prevention measure recommendation module.

[0026] Optionally, the preventive measures recommendation module includes a risk analysis unit, a measure matching unit, and a solution generation unit.

[0027] Optionally, the preventative measures recommendation module includes:

[0028] The risk analysis unit receives a dynamic risk distribution map from the disaster prediction and early warning module, performs spatial gridding processing on the dynamic risk distribution map, extracts the risk level, risk type and spatial coordinate information of each grid cell, and generates gridded risk data.

[0029] The measure matching unit calls the built-in preventive measure knowledge base to perform rule matching between the risk level and risk type in the gridded risk data and the preventive measure entries pre-stored in the preventive measure knowledge base. The rule matching includes risk level threshold matching, disaster type matching and geological condition matching, and outputs a set of candidate preventive measures corresponding to each grid unit.

[0030] The scheme generation unit performs an economic and technical feasibility optimization analysis on the candidate prevention measures set, calculates the cost-benefit ratio of different measure combinations, selects the optimal measure combination, and generates a zoned precision prevention scheme that includes project type, implementation location, technical parameters, and cost estimate.

[0031] The beneficial effects of this invention are:

[0032] This invention, by constructing a data acquisition system that integrates multi-source heterogeneous data, achieves high-precision acquisition of all elements related to the geological structure, hydrological conditions, and environmental impacts of closed open-pit coal mine areas. Compared to traditional single-method or qualitative observation approaches, the data acquisition module of this invention integrates geological drilling, geophysical imaging, 3D laser scanning, monitoring well deployment, remote sensing inversion, and sensor networks, ensuring the multidimensionality, timeliness, and spatial continuity of the data. This provides a solid data foundation for subsequent disaster factor identification and simulation, greatly improving the accuracy and quantifiability of early disaster identification.

[0033] This invention introduces an algorithm system combining principal component analysis and grey relational analysis to quantitatively correlate multidimensional data of key disaster-causing factors with historical disaster events, solving the problems of unclear correlations and development mechanisms among factors causing closed-pit disasters. Simultaneously, by constructing isosurfaces of disaster factors in a three-dimensional geological model and employing a weighted overlay analysis method, the system can accurately delineate potential disaster development zones spatially, providing spatial constraints for numerical simulation and early warning. This approach is more objective, scale-adaptive, and automated than traditional methods based on empirical judgment or two-dimensional partitioning.

[0034] This invention constructs a multi-field coupled numerical model of groundwater seepage field, soil and rock stress field, and pollutant transport field. Combined with disaster evolution trend simulation and dynamic risk distribution map visualization mechanisms, it achieves cross-scale, multi-factor, and temporal evolution prediction of various disasters such as landslides, subsidence, and pollution, forming an integrated disaster prediction and early warning module. Simultaneously, the preventative measures recommendation module, based on a process of risk analysis, knowledge base rule matching, and cost-benefit ratio optimization, automatically outputs precise regional prevention and control plans including location, type, parameters, and cost. This significantly improves the intelligence and engineering operability of disaster prevention and mitigation response, realizing a shift from "experience-based prevention" to "data-driven + intelligent recommendation" in pit closure remediation. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a schematic diagram of the system flow according to an embodiment of the present invention;

[0037] Figure 2 This is a flowchart illustrating the prevention measure recommendation module in an embodiment of the present invention. Detailed Implementation

[0038] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0039] It should be noted that the use of terms such as "an embodiment," "an embodiment," "an exemplary embodiment," and "some embodiments" in the specification indicates that the described embodiment may include a specific feature, structure, or characteristic, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, when a specific feature, structure, or characteristic is described in connection with an embodiment, implementing such a feature, structure, or characteristic in conjunction with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the art.

[0040] Generally, terms can be understood at least partly from their use in context. For example, depending at least partly on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in a singular sense, or a combination of features, structures, or characteristics in a plural sense. Additionally, the term "based on" can be understood not necessarily to convey an exclusive set of factors, but rather, alternatively, depending at least partly on the context, to allow for the presence of other factors that are not necessarily explicitly described.

[0041] like Figures 1-2 As shown, a multi-dimensional assessment and prevention system for closure disasters in open-pit coal mines includes a data acquisition module, a data analysis and processing module, a disaster prediction and early warning module, and a prevention measure recommendation module, wherein;

[0042] The data acquisition module collects multi-source heterogeneous data from the closed area of ​​the target open-pit coal mine and uploads the data to the data analysis and processing module. The multi-source heterogeneous data includes geological structure data, hydrological condition data, and environmental impact data, specifically:

[0043] The data acquisition module includes a geological structure data acquisition unit, a hydrological condition data acquisition unit, an environmental impact data acquisition unit, and a data integration unit. It is used to collect multi-source heterogeneous data of the target open-pit coal mine closed area and provide standardized input for the subsequent data analysis and processing module.

[0044] The geological structure data acquisition unit obtains complete core samples by drilling core boreholes with a diameter of no less than Φ89mm and a depth of no less than 60m in the key structural zone of the closed pit area and using hydraulic core drilling equipment. The core samples are then subjected to lithological description, joint measurement, weathering degree determination and interlayer identification to form rock and soil layer distribution information with depth labels.

[0045] This unit employs high-frequency seismic wave excitation and reception, with receiver points spaced 5m apart along the survey line to acquire subsurface wave velocity profiles. Shortwave radar scanning, combined with ground-penetrating radar, is used to obtain shallow structural anomaly echo images, and a resistivity imaging system is employed to acquire electrical distribution images. The combined analysis of these three results extracts structural features such as fault fracture zones, fold axes, and dipping structural planes, calculating the strike, dip, and dip angle of the structural planes to construct their attitude information.

[0046] Simultaneously, the unit deploys a laser scanning system, with base stations set at the top of the pit slope and the center of the platform, to collect three-dimensional laser point clouds at no fewer than six stations, with a single-point ranging accuracy better than 5mm. After point cloud fusion and noise removal, a digital elevation model of the closed pit area is generated, and parameters such as platform width, slope height, slope angle, and polygonal line position are extracted to output the three-dimensional geomorphological information of the pit.

[0047] Within this unit, the distribution information of soil and rock layers, the occurrence information of structural planes, and the three-dimensional geomorphological information of the mining pit are converted into a unified format (CSV format), data tags are attached, spatial reference points are registered, and then packaged to generate complete geological structure data.

[0048] The hydrological data acquisition unit deploys five monitoring wells in the closed pit area, each with a depth of no less than 50m. Submersible pressure level gauges are used, and water level data is collected every 10 minutes. All level gauges are connected to a communication gateway via a data acquisition unit, continuously outputting timestamped groundwater level data.

[0049] Vibrating wire piezometers were installed in the slope deformation-sensitive areas and at the bottom of the mining pit. The burial depths were set to three layers: 10m, 20m, and 30m. The pore water pressure changes were collected and transmitted to the central server in real time via a wireless acquisition terminal to output the pore water pressure data.

[0050] Multi-parameter water quality sensors were installed at historical water inflow points, ditch catchment points, and water accumulation areas in sampling pits to collect pH, conductivity, dissolved oxygen, total phosphorus, sulfate, and manganese ion concentrations, automatically once per hour. Each sampling point was uniformly coded using an RTU module to achieve on-site, fixed-point, and quantitative data collection, outputting structured water chemical composition data.

[0051] This unit performs unit conversion, outlier removal, and unified time label processing on all data, and archives the data according to well location number and measuring point coordinates, ultimately generating complete hydrological condition data.

[0052] The environmental impact data acquisition unit calls the Sentinel-1 synthetic aperture radar satellite to acquire images of the closed pit area, and extracts surface deformation information at a 3-month scale based on the temporal interferometric differential method (D-InSAR) to form surface deformation data in units of pit coordinate grid.

[0053] The drone is equipped with a multispectral imaging system to acquire four-channel images of red, green, blue and near-infrared light. The normalized vegetation index (NDVI) is calculated using the pixel method, and the vegetation coverage percentage is output at a resolution of 0.1. After the grid number is labeled, the vegetation coverage data is generated.

[0054] A sampling point was set at every 100m×100m grid. Topsoil samples (0-20cm deep) were collected using the ring cutter method. The total lead, total cadmium, and hexavalent chromium content were tested in the laboratory, and the pH value was measured. The detection methods were atomic absorption spectrophotometry and potentiometry, respectively. The test results were standardized to form soil pollution data.

[0055] All environmental impact data within this unit undergoes unified coordinate system transformation, resolution resampling, and image data vectorization, ultimately integrating into a well-structured and fully coded environmental impact data set.

[0056] The data integration unit receives the aforementioned geological structure data, hydrological condition data, and environmental impact data, and performs unified processing on all data, as follows:

[0057] All data were uniformly converted to GeoTIFF (raster) and GeoJSON (vector) formats, and multi-level file directories were established according to data type, time label, and sampling point number to achieve format standardization;

[0058] A clock-synchronization-based timestamp correction method is adopted to unify the time of all data to Beijing time (UTC+8), with each hour as the basic time step, to generate an aligned data time series and achieve time synchronization processing;

[0059] Based on the GNSS reference point coordinate system of the closed pit area, all spatial data are uniformly projected onto the WGS84 spatial reference frame, and spatial resampling processing based on nearest neighbor interpolation is performed to achieve spatial registration processing.

[0060] All processed data is uploaded to the data analysis and processing module via encrypted transmission, and used for subsequent functions such as disaster-causing factor analysis, disaster area delineation, disaster prediction and early warning, and prevention measure recommendation.

[0061] The data analysis and processing module receives multi-source heterogeneous data and uses built-in algorithm models to fuse and analyze the data, identify potential key disaster-causing factors, and delineate disaster development zones. Specifically:

[0062] The data analysis and processing module includes a data preprocessing unit, a disaster-causing factor analysis unit, and a disaster zone delineation unit.

[0063] The data preprocessing unit receives geological structure data, hydrological condition data, and environmental impact data from the data acquisition module and processes them according to the following steps:

[0064] Data cleaning: Boundary value removal, sliding window smoothing, and spatial outlier identification are performed on each type of data to remove noise and abrupt changes that do not conform to physical laws;

[0065] Standardized format: The unified data structure is a 5-tuple format (x,y,z,t,v), where x,y,z represent spatial coordinate values, t represents timestamps, and v represents variable values;

[0066] Spatiotemporal registration: unify timestamps to UTC+8 and 1-hour granularity; unify spatial coordinate system to WGS84, and perform 3D interpolation to complete all spatial point data to ensure spatiotemporal consistency.

[0067] The processed data forms standardized, structured preprocessed multi-source heterogeneous data, which is stored in the analysis buffer.

[0068] The disaster-causing factor analysis unit receives the preprocessed data and, based on principal component analysis and grey relational analysis algorithms, identifies the key disaster-causing factors of closed-pit disasters. Its processing flow and mathematical model are as follows:

[0069] Principal Component Analysis (PCA): To reduce redundancy among multidimensional features and improve subsequent analysis efficiency, the feature variables of each spatial sample are used to construct an original data matrix. Let there be m spatial samples in the closed region, and each spatial sample contain n feature variables. The original data matrix is ​​defined as follows:

[0070] X = [x ij ] m×n , where x ij This represents the value of the j-th feature variable of the i-th sample.

[0071] To eliminate differences in the dimensions of the variables, a standardization operation was performed on the original data matrix. The standardized data matrix is ​​as follows: Among them, z ij For the standardized variable value, μ j Let σ be the mean of the j-th feature variable. j Let represent the standard deviation of the j-th feature variable.

[0072] After standardization, calculate the covariance matrix C of the standardized data:

[0073] Subsequently, eigenvalue decomposition was performed on the covariance matrix C to obtain the eigenvalues ​​λ1,λ2,...,λ n and the corresponding eigenvectors e1, e2, ..., e n Each principal component f k It can be calculated using the following formula:

[0074] f k =Ze k , where f k Let e ​​represent the k-th principal component. k is the k-th eigenvector corresponding to the covariance matrix.

[0075] The amount of information carried by each principal component is measured by the "variance contribution rate," which is calculated using the following formula:

[0076]

[0077] The cumulative contribution rate is the proportion of total information in the top k principal components:

[0078]

[0079] When the cumulative contribution rate is greater than or equal to 85%, the first k principal components can be considered to basically represent the information content of the original variables. At this point, the original variables participating in the composition of the principal components are the main characteristic variables, which will be used as input variables for grey relational analysis.

[0080] Grey Relational Analysis (GRA): After obtaining the main characteristic variables, to further evaluate the correlation between these variables and historical disaster events, grey relational analysis is used to screen out the truly catastrophic variables. Let the historical disaster events constitute a reference sequence X0, defined as:

[0081] X0={x0(1),x0(2),...,x0(m)};

[0082] Where x0(k) represents the risk level or disaster occurrence value of the k-th sample in historical disaster events. Each key feature variable constitutes a comparison sequence X. j ;

[0083] X j ={x j (1),x j (2),...,x j (m)};

[0084] Where, x j (k) is the value of the j-th variable in the k-th sample.

[0085] Calculate the correlation coefficient ξ of each variable at each time point. j (k), the formula is as follows:

[0086]

[0087] Where, Δ j (k)=|x0(k)-x j (k)| represents the absolute difference between the reference sequence and the comparison sequence. This represents the minimum difference between all variables at all times. ρ represents the maximum difference between all variables at all times, and ρ is the resolution coefficient.

[0088] Calculate the grey relational degree r of variable j.j :

[0089]

[0090] That is, the average correlation coefficient over all time periods, representing the overall correlation between the variable and historical disasters.

[0091] By combining the contribution rate and grey relational degree in the principal component analysis, the comprehensive weight w of each variable is calculated. j :

[0092] w j =α·contribution rate j +(1-α)·r j ;

[0093] Among them, w j Let represent the overall weight of the j-th variable, and α be the weight adjustment coefficient, ranging from [0,1]. In this embodiment, it is set to 0.5, representing the contribution rate. j r represents the variance contribution rate of the variable in principal component analysis. j This indicates the degree of grey relational relationship.

[0094] All variables are assigned a comprehensive weight w j Sort the variables and select the top 3 as the key disaster-causing factors in the final output, along with their variable names, units, weight values ​​and spatial distribution information, for use by disaster area delineation units.

[0095] The disaster zone delineation unit is used to identify potential high-incidence areas of disasters in three-dimensional geological space based on the aforementioned disaster-causing factor analysis results, and submits the identification results to the disaster prediction and early warning module as the basis for model-driven and risk map drawing.

[0096] Specifically, the analysis process for delineating disaster zones includes the following three steps: spatial interpolation modeling, weighted overlay analysis, and threshold determination delineation, as follows:

[0097] Spatial interpolation modeling: First, for each key disaster-causing factor output by the disaster-causing factor analysis unit, its original data is typically the values ​​of several discrete measurement points at (x, y, z) in three-dimensional space. To perform continuous spatial analysis of the entire closed pit area, these discrete point values ​​need to be interpolated in three dimensions to generate a continuous spatial value field. Inverse distance weighted interpolation (IDW) is used to model each key disaster-causing factor, generating its value field function in three-dimensional space.

[0098] Let the spatial distribution function of the i-th critical disaster-causing factor be: F i (x,y,z);

[0099] Among them, F i(x,y,z) represents the predicted value of the i-th critical disaster-causing factor at spatial location (x,y,z), where i = 1, 2, ..., n, and n is the number of critical disaster-causing factors selected.

[0100] Through this process, the system can construct a spatially continuous distribution map of each key disaster-causing factor within the entire closed pit area, providing input for subsequent comprehensive risk assessment.

[0101] Weighted overlay analysis: After completing the spatial value field construction of each key disaster-causing factor, these value field information need to be integrated in order to assess whether there is a potential disaster hazard at a certain spatial location.

[0102] The fusion method employs a weighted superposition approach, which linearly combines multiple factors according to their disaster-causing intensity weights. Let w be the comprehensive weight obtained by each key disaster-causing factor in the grey relational analysis. i The formula for calculating the comprehensive disaster risk intensity value R(x,y,z) at a certain point (x,y,z) is as follows:

[0103]

[0104] Where R(x,y,z) represents the comprehensive disaster risk intensity value of spatial location (x,y,z) in the closed pit area, and F i (x,y,z) represents the spatial value of the i-th critical disaster-causing factor, w i The weight of the i-th critical disaster-causing factor is ∑w i =1.

[0105] The essence of this step is to construct a comprehensive disaster intensity assessment field, enabling the system to quantify its disaster risk level at any spatial location.

[0106] Threshold determination: In order to identify high-risk areas from the disaster risk intensity field, the system sets a safety threshold T. This threshold can be determined by historical disaster event data or expert experience, and represents the critical value of disaster intensity.

[0107] When the comprehensive risk intensity R(x,y,z) of a point in space exceeds a certain threshold, it can be considered to possess the potential risk of disaster development. The judgment criteria are as follows:

[0108] R(x,y,z)≥T;

[0109] All three-dimensional spatial units that meet this condition constitute a continuous potential high-incidence area of ​​disasters, i.e., a disaster development zone.

[0110] After identifying all grid cells that meet the conditions, the system extracts their boundary coordinate sets and includes the corresponding key disaster-causing factor values ​​and their spatial distribution information, which are then encapsulated into a structured data package for the disaster area.

[0111] Through the above three-step process of interpolation, superposition, and judgment, the disaster zone delineation unit realizes the quantitative identification of potential disaster risk space in closed pit areas, providing scientific support for achieving regional-level refined disaster prevention management.

[0112] The disaster prediction and early warning module receives information on key disaster-causing factors and disaster development zones, predicts the evolution trend of disasters through numerical simulation, and calculates and generates a dynamic risk distribution map of the entire mining area. Specifically:

[0113] The disaster prediction and early warning module is used to simulate the evolution trend of disasters in closed open-pit coal mines through multiple physical processes based on key disaster-causing factors and disaster development zone information. It then generates spatialized and dynamic risk distribution results through time-series extrapolation and risk quantification analysis. This module includes a model building unit, a simulation and prediction unit, and a risk mapping unit. Its processing flow is as follows:

[0114] The model building unit receives key disaster-causing factor data and three-dimensional spatial boundary coordinate information of the disaster development zone from the data analysis and processing module, and constructs a multi-field coupled simulation model to reveal the disaster driving mechanism and future evolution trend.

[0115] First, the boundary coordinate set of the disaster development zone is imported into a modeling tool (such as COMSOL Multiphysics or FLAC3D) to construct a three-dimensional geometric model. Based on the measurement point data and spatial interpolation results, an unstructured mesh is divided into the model. The meshing element types include tetrahedral, hexahedral, and prism composite elements to ensure model accuracy and computational convergence.

[0116] Then, based on the physical meaning of each type of key disaster-causing factor, it is mapped to different physical domains in the three-field coupling model:

[0117] Factors belonging to groundwater dynamics (such as groundwater level and pore water pressure) are assigned to the groundwater seepage field to solve for the pore flow velocity under Darcy's law.

[0118] Factors belonging to the soil and rock stability category (such as structural plane dip angle and shear strength parameters) are assigned to the soil and rock stress field, and the Mohr-Coulomb or Drucker-Prager constitutive model is executed.

[0119] Factors belonging to the category of environmental pollution (such as heavy metal concentration and water chemical composition) are assigned to the pollutant transport field to control the boundary conditions of the convection-diffusion-adsorption equation.

[0120] The entire model is solved using the finite element method. The three fields are constrained by boundary flow, pore pressure stress coupling, and solute influence intensity, constructing a unified multi-field coupled numerical model of groundwater seepage field, soil stress field, and contaminant transport field. In this model, each field iterates synchronously using a uniform time step, and the state variables of each field receive feedback input from other fields, forming a closed-loop coupling.

[0121] The simulation prediction unit runs the aforementioned multi-field coupled numerical model, inputs different simulation conditions, and conducts a series of simulation experiments to predict the evolution trend of the disaster development zone in a specific future period. Simulation conditions include, but are not limited to:

[0122] Changes in surface rainfall intensity (simulating heavy rain conditions);

[0123] Slope retaining structure failure (simulated engineering disturbance);

[0124] Groundwater level rises or surges (simulating a sudden water inrush);

[0125] The initial concentration of pollutants increases (simulating a pollution event response).

[0126] For each operating condition setting, the model iterates sequentially until the target prediction time point (e.g., 7 days, 30 days, 90 days), outputting the following three types of disaster evolution trend indicators:

[0127] Landslide failure probability: The slope failure probability distribution is calculated based on the safety factor method (FS method) and reliability analysis methods (such as FORM);

[0128] Surface subsidence: Outputs a surface subsidence field based on the vertical displacement of nodes;

[0129] Pollutant concentration distribution: Based on the convection-diffusion-adsorption equation, the concentration cloud map of the main pollutants in the model domain is output.

[0130] All three types of indicators are output in raster form, with clear spatial location, units of measurement, and numerical changes over time, forming a complete dataset of disaster evolution trends.

[0131] After obtaining three types of quantitative indicators of disaster evolution trends, the risk mapping unit uses the risk matrix analysis method to comprehensively calculate the probability of disaster occurrence and the degree of potential harm at different spatial locations, forming a quantitative risk assessment value.

[0132] The specific method is as follows:

[0133] Define five levels of disaster occurrence probability (extremely low, relatively low, moderate, relatively high, extremely high) and five levels of hazard severity (minor, moderate, severe, major, catastrophic);

[0134] A 5×5 two-dimensional risk matrix is ​​constructed, and the three types of indicators are standardized and mapped into the matrix cells.

[0135] The comprehensive risk quantification value R for each location q (x,y,z,t) is calculated as follows:

[0136] R q (x,y,z,t)=P d (x,y,z,t)·H s (x,y,z,t);

[0137] Among them, R q (x,y,z,t) represents the combined risk value at time t and spatial location (x,y,z), P d (x,y,Z,t) represents the probability of landslide failure or anomaly in settlement, with values ​​ranging from 0 to 1. s (x,y,z,t) is a normalized hazard score, based on the deposition amount or pollution concentration mapping value, ranging from 1 to 5.

[0138] Finally, the risk quantification value R of all grid cells is calculated. q Displayed in a 3D geographic information system platform (such as ArcGIS Pro or Cesium platform), the risk is visualized using color gradients or risk heatmaps, outputting a dynamic risk distribution map with complete spatiotemporal dimensions.

[0139] This dynamic risk distribution map will be spatially registered according to the actual terrain of the closed pit area, has a multi-level query interface, supports classification and display by disaster type, time and level, and is transmitted to the prevention measure recommendation module through the interface for the generation of refined response strategies.

[0140] The preventative measures recommendation module, based on the received dynamic risk distribution map, calls upon the built-in preventative measures knowledge base for matching, automatically generating and outputting spatially-based, zone-specific, and precise preventative plans. Specifically:

[0141] The disaster prevention measure recommendation module, after obtaining disaster prediction and early warning results, automatically generates spatially accurate, clearly categorized, and economically reasonable prevention and control measure implementation plans by combining spatial risk information with a knowledge base matching mechanism. This module consists of a risk analysis unit, a measure matching unit, and a plan generation unit, and its specific implementation method is as follows:

[0142] The risk analysis unit receives the dynamic risk distribution map output from the disaster prediction and early warning module. First, it performs three-dimensional spatial discretization on the map and uses an equilateral regular grid or an adaptive quadtree block strategy to construct a two-dimensional / three-dimensional spatial grid system with a spatial resolution of no more than 20m×20m.

[0143] For each grid cell, extract the following three types of information:

[0144] Risk level: Based on the risk quantification value R of each unit in the dynamic risk distribution map. q (x,y,z,t) is mapped to integer levels (1 to 5) according to the classification rules, corresponding to five risk levels: low, relatively low, medium, relatively high, and extremely high, respectively.

[0145] Risk type: Based on the proportion of the dominant influencing factors in the risk indicators (e.g., if the probability of landslide failure is the highest, it is defined as landslide risk), it is automatically classified into landslide type, subsidence type, pollution type, or compound type.

[0146] Spatial coordinate information: includes spatial labels such as the spatial center point coordinates (x, y, z) of each grid cell, the partition number, the surface elevation, and the boundary outline.

[0147] The above three types of data are packaged in a standard structure to generate gridded risk data.

[0148] The measure matching unit has a built-in structured knowledge base of preventive measures, which is hierarchically constructed according to disaster type, risk level, geological structure conditions, and constraints. Each preventive measure entry contains the following fields:

[0149] Disaster types (landslide, subsidence, pollution);

[0150] Scope of application of risk levels (e.g., level 3 and above);

[0151] Geological conditions applicable labels (such as weak interlayers, high groundwater level, easily diffusible sand layers);

[0152] Names of preventative measures (e.g., anchor reinforcement, drainage blind ditch, geomembrane barrier, counterweight platform construction, etc.);

[0153] Technical parameters (specifications, depth, anchoring force, construction methods, etc.);

[0154] Unit cost and applicable scale range.

[0155] During the measure matching process, the system sequentially performs three types of rule matching:

[0156] Risk level threshold matching: Select all measures that are applicable to the current grid risk level or higher;

[0157] Disaster type matching: Filter measures that match the risk type of the current grid cell;

[0158] Geological condition matching: Based on the spatial geological structure tags provided by the data analysis and processing module, the applicable geological prerequisites (such as fractured zones, saturated silt, inclined structural surfaces, etc.) marked in the measures knowledge base are matched.

[0159] The final output is a set of candidate preventive measures for each grid cell, and each set contains several preventive measures that can be selected.

[0160] The scheme generation unit combines and evaluates the candidate prevention measures set for each grid unit and optimizes the scheme. The goal is to output a final prevention scheme with a clear project type, complete technical parameters, and reasonable cost estimate, while meeting the risk reduction effect.

[0161] The specific process is as follows:

[0162] Cost-benefit ratio calculation: For each candidate measure, calculate its cost-benefit ratio (CEB) per unit area. i :

[0163]

[0164] Among them, E i This indicates the degree of risk reduction (risk level decrease, in levels) brought about by the measure in the simulation model, C. i The implementation cost per unit area or unit volume (in yuan / square meter or yuan / cubic meter), CEB i The cost-benefit ratio of the candidate measures.

[0165] Optimal combination selection: Arrange and combine multiple candidate measures (such as using two measures together), calculate the total cost and total benefit increment after combination, and select the combination measure with the highest cost-benefit ratio under the condition of meeting the minimum risk suppression threshold (such as risk level reduction ≥ 2 levels) as the final output.

[0166] Partition-based precise prevention solution generation: Each final solution output by the system must include the following information fields:

[0167] Project types (such as anchor bolt support, anti-slide pile installation, filter layer laying, etc.);

[0168] Implementation location (accurate to spatial grid number or center point coordinates);

[0169] Technical parameters (such as construction depth, spacing, material type, construction sequence, etc.);

[0170] Cost estimation (unit cost and total project cost calculated based on market prices and design parameters);

[0171] Construction suggestions and precautions (such as whether the work area needs to be closed, construction period, construction sequence, etc.).

[0172] The final results corresponding to all grid units will be integrated into a structured, precise prevention plan for each zone, which can be exported as a table file, engineering drawing, or GIS layer. It can be distributed, deployed, or displayed graphically through the platform system.

[0173] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0174] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An open-pit coal mine closed pit disaster multi-dimensional assessment prevention system, characterized in that, The system comprises a data collection module, a data analysis and processing module, a disaster prediction and early warning module, and a preventive measure recommendation module. The data collection module collects multi-source heterogeneous data of the target open-pit coal mine closed area and uploads the data to the data analysis and processing module. The data analysis and processing module receives the multi-source heterogeneous data, fuses and analyzes the data using an embedded algorithm model, identifies potential key disaster-causing factors, and delineates a disaster development area. The disaster prediction and early warning module receives the key disaster-causing factors and disaster development area information, predicts the disaster evolution trend through numerical simulation, and calculates the dynamic risk distribution map of the entire mine area. The preventive measure recommendation module calls the built-in preventive measure knowledge base for matching based on the received dynamic risk distribution map, automatically generates and outputs a partitioned precise prevention scheme with spatial location.

2. The multi-dimensional disaster assessment and prevention system for closed pit of an open-pit coal mine according to claim 1, characterized in that, The data collection module comprises a geological structure data collection unit, a hydrological condition data collection unit, an environmental impact data collection unit, and a data integration unit.

3. The multi-dimensional disaster assessment and prevention system for closed pit of an open-pit coal mine according to claim 2, characterized in that, The data collection module comprises: The geological structure data collection unit obtains core samples through geological drilling and records rock-soil layer distribution information, detects underground geological structures through geophysical methods and records structure surface occurrence information, scans the surface morphology of the open-pit through a three-dimensional laser scanner and generates three-dimensional topographic information of the pit, and integrates the rock-soil layer distribution information, structure surface occurrence information, and pit three-dimensional topographic information into complete geological structure data. The hydrological condition data collection unit continuously monitors and records underground water level data through water level meters installed in monitoring wells, monitors and records pore water pressure data through piezometers buried in slopes and pit bottoms, collects and records water chemical component data through water quality multi-parameter sensor networks arranged at underground water outlets and surface water bodies, and integrates the underground water level data, pore water pressure data, and water chemical component data into complete hydrological condition data. The environmental impact data collection unit obtains mine area surface deformation interferograms through synthetic aperture radar satellites and calculates surface deformation data, obtains vegetation index through multi-spectral unmanned aerial vehicle aerial photography and inverts vegetation coverage data, obtains heavy metal content and pH value through ground soil sampling and testing and records soil pollution data, and integrates the surface deformation data, vegetation coverage data, and soil pollution data into complete environmental impact data. The data integration unit receives the geological structure data, hydrological condition data, and environmental impact data, and performs format standardization, time synchronization, and spatial registration processing, finally packages the multi-source heterogeneous data recognizable by the system and uploads it to the data analysis and processing module.

4. The multi-dimensional disaster assessment and prevention system for closed pit of an open-pit coal mine according to claim 3, characterized in that, The data analysis and processing module comprises a data preprocessing unit, a disaster-causing factor analysis unit, and a disaster area delineation unit.

5. The multi-dimensional disaster assessment and prevention system for closed pit of an open-pit coal mine according to claim 4, characterized in that, The data analysis and processing module comprises: The data preprocessing unit receives multi-source heterogeneous data from the data acquisition module, performs data cleaning on the multi-source heterogeneous data to eliminate abnormal values and noise, performs format standardization to unify data structure and measurement units, performs spatio-temporal registration to make all data have the same time stamp and spatial coordinate system, and generates preprocessed multi-source heterogeneous data; The disaster factor analysis unit receives the preprocessed multi-source heterogeneous data, calls a principal component analysis algorithm in a built-in algorithm model to perform dimensionality reduction processing on the preprocessed multi-source heterogeneous data to extract main characteristic variables, calls a grey correlation analysis algorithm to calculate the correlation degrees of each characteristic variable and historical disaster events, and selects the key disaster factors with the largest contribution degrees from all characteristic variables according to variance contribution rates and correlation degree sorting results; The disaster area delineation unit receives the key disaster factors, performs spatial interpolation on the measured data or calculated values of each key disaster factor in the three-dimensional geological model to generate a disaster factor contour surface, performs superimposed calculation on multiple disaster factor contour surfaces by using a weighted superimposed analysis method, delineates a potential disaster development area that exceeds a safety threshold in the three-dimensional space according to a set threshold, and outputs the disaster development area and corresponding key disaster factor information to the disaster prediction and early warning module.

6. The multi-dimensional disaster assessment and prevention system for closed pit of an open-pit coal mine according to claim 5, characterized in that, The disaster prediction and early warning module includes a model construction unit, a simulation prediction unit, and a risk mapping unit.

7. The multi-dimensional disaster assessment and prevention system for closed pit of an open-pit coal mine according to claim 6, characterized in that, The disaster prediction and early warning module includes: The model construction unit receives key disaster factors and disaster development area information from the data analysis and processing module, constructs a multi-field coupled numerical model coupling a groundwater seepage field, a rock-soil stress field, and a pollutant transport field by using a finite element method based on physical parameters of the key disaster factors and spatial geometric characteristics of the disaster development area, and assigns values to the key disaster factors as model input parameters; The simulation prediction unit runs the multi-field coupled numerical model, simulates changes in slope stability, groundwater flow paths, and pollutant migration and diffusion processes in the disaster development area under different working conditions, predicts disaster evolution trends in a specific future time period, including three types of quantitative indexes, i.e., landslide damage probability, surface subsidence, and pollutant concentration distribution, through time series iteration calculation; The risk mapping unit comprehensively quantitatively calculates disaster occurrence probability and potential harm degree by using a risk matrix analysis method according to the three types of quantitative indexes of the disaster evolution trend, generates risk quantitative values with spatio-temporal characteristics, and visually renders the risk quantitative values on a three-dimensional geographic information system platform, and finally outputs a dynamic risk distribution map of the entire mining area space and transmits the dynamic risk distribution map to the prevention measure recommendation module.

8. The multi-dimensional disaster assessment and prevention system for closed pit of an open-pit coal mine according to claim 7, characterized in that, The prevention measure recommendation module includes a risk analysis unit, a measure matching unit, and a scheme generation unit.

9. The multi-dimensional disaster assessment and prevention system for closed pit of an open-pit coal mine according to claim 8, characterized in that, The prevention measure recommendation module includes: The risk analysis unit receives a dynamic risk distribution map from the disaster prediction and early warning module, performs spatial gridding processing on the dynamic risk distribution map, extracts risk level, risk type, and spatial coordinate information of each grid cell, and generates gridded risk data; The measure matching unit calls a built-in prevention measure knowledge base to perform rule matching between the risk level and risk type in the grid risk data and the pre-stored prevention measure entries in the prevention measure knowledge base, the rule matching includes risk level threshold matching, disaster type matching and geological condition matching, and outputs a candidate prevention measure set corresponding to each grid unit; The scheme generation unit performs economic and technical feasibility optimization analysis on the candidate prevention measure set, calculates the cost-benefit ratio of different measure combinations, selects an optimal measure combination, and generates a partitioned accurate prevention scheme including engineering type, implementation location, technical parameters and cost estimation.

Images