Three-dimensional visual emergency management system for mining area

Abstract

The invention relates to the technical field of mining area emergency management, and discloses a mining area three-dimensional visual emergency management system which comprises a data acquisition module and a three-dimensional sand table generation module. The data acquisition module is used for acquiring dynamic data including geological structures (including static exploration and dynamic sensor data), disaster hidden dangers and environmental monitoring; the three-dimensional sand table generation module processes data through a three-dimensional visualization engine, the data preprocessing unit completes unification of a space coordinate system, missing data complementation and abnormal value correction, and the scene rendering unit generates a three-dimensional dynamic sand table through texture mapping, spatial feature fusion, spatial coupling modeling and three-dimensional scene synthesis. The system is provided with functional modules of offline data loading, disaster diffusion simulation, escape route planning and the like, and supports cross-platform rendering and user interaction operation. According to the invention, three-dimensional dynamic display and emergency analysis of mining area multi-source data are realized, and the accuracy and timeliness of disaster early warning and emergency decision making are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of mine emergency management, in particular to a mine three-dimensional visual emergency management system. BACKGROUND

[0002] In the mining operation, due to the complex geological conditions and the changeable operation environment, various safety risks and disaster hidden dangers are always the key factors restricting the safety development of the industry. The traditional mine emergency management means mainly relies on two-dimensional drawings and scattered monitoring data, which is difficult to intuitively and dynamically present the overall geological conditions, disaster hidden danger distribution and environmental change trend of the mine, resulting in the lack of comprehensiveness and timeliness of emergency decision-making.

[0003] From the perspective of geological data management, static geological data and dynamic geological data are usually stored in different systems, and there is a lack of effective fusion mechanism. Static geological data mainly comes from the previous geological exploration platform, which can reflect the basic geological structure of the mine area, but cannot be updated in real time. Dynamic geological data is obtained through the sensor network deployed in the mine area, which can monitor dynamic information such as stratum displacement and stress change in real time. However, there are differences in spatial coordinate system, data format and precision between the two types of data, and traditional methods are difficult to realize accurate docking and comprehensive analysis, resulting in lagging geological risk early warning.

[0004] In terms of disaster hidden danger monitoring, the existing system has a single processing of disaster hidden danger data, which can only realize threshold alarm of single point data and cannot deeply analyze the spatial coupling relationship between different types of disaster hidden dangers. For example, there may be a complex relationship between fault activity and groundwater infiltration, but traditional technology is difficult to reveal this potential relationship through multi-source data fusion, resulting in the inability to predict the formation and development path of disaster chain in advance.

[0005] The processing of environmental monitoring data also faces challenges. The environmental monitoring of the mine area involves multidimensional data such as air quality, dust concentration and harmful gas content. The traditional system has insufficient accuracy in correcting abnormal data and completing missing data, which may lead to distortion of environmental risk assessment. In addition, various monitoring data lack intuitive visual display means, and management personnel are difficult to quickly grasp the real-time changes of the mine area environment, affecting the speed of emergency response.

[0006] In terms of visualization technology, the traditional two-dimensional visualization method cannot truly restore the three-dimensional spatial structure of the mine area, especially in the display effect of underground tunnel layout and three-dimensional distribution of geological structure. For emergency management, the lack of support of three-dimensional dynamic sand table makes it difficult to quickly plan escape routes and simulate disaster diffusion process when disasters occur, resulting in low efficiency of emergency drills and actual rescue work.

[0007] The cross-platform compatibility problem also restricts the popularization and application of the mine emergency management system. The underlying graphic interfaces of different operating systems and hardware devices are different, and the traditional visualization engine is difficult to realize unified rendering across platforms, resulting in inconsistent display effects and operation experience of the system on different terminal devices, and increasing the cost of system deployment and maintenance. SUMMARY

[0008] The purpose of the present application is to provide a mine three-dimensional visualization emergency management system to solve the problems raised in the background art.

[0009] To achieve the above purpose, the present application provides the following technical solution: a mine three-dimensional visualization emergency management system, the system comprising:

[0010] A data acquisition module for acquiring real-time dynamic data of a target mine, the dynamic data including a first spatial data set corresponding to geological structure data, a second spatial data set corresponding to disaster hazard data, and a third spatial data set corresponding to environmental monitoring data, the geological structure data including static geological data obtained from a geological exploration platform and dynamic geological data obtained through a sensor network deployed in the mine;

[0011] A three-dimensional sand table generation module for inputting the dynamic data into a three-dimensional visualization engine for processing;

[0012] A three-dimensional dynamic sand table of the target mine is constructed according to the output results of the three-dimensional visualization engine; the three-dimensional visualization engine includes a data preprocessing unit and a scene rendering unit, wherein the data preprocessing unit is used for spatializing conversion of the dynamic data, and the scene rendering unit is used for dynamic rendering generation based on a preset mine three-dimensional model template and historical disaster data; the scene rendering unit includes a spatial mapping layer, a first rendering layer, a second rendering layer and a display layer connected in turn, the first rendering layer is used for texture mapping processing of a plurality of spatial data sets contained in the dynamic data to generate a spatial feature fusion model; the second rendering layer is used for modeling the spatial coupling relationship between the spatial feature fusion models corresponding to various types of dynamic data to generate spatial correlation topology data; the display layer is used for three-dimensional scene synthesis based on the spatial correlation topology data and the spatial feature fusion model to generate a three-dimensional dynamic sand table.

[0013] Preferably, the spatial coupling relationship between the spatial feature fusion models corresponding to various types of dynamic data is modeled to generate spatial correlation topology data, comprising:

[0014] An edge detection algorithm is used to identify the abrupt boundaries in the spatial feature fusion model, and a boundary distribution sequence corresponding to each type of dynamic data is determined based on the spatial position of each abrupt boundary;

[0015] The sum of Manhattan distances between mutation boundaries with same spatial code in the boundary distribution sequences corresponding to any two types of dynamic data is calculated, and spatial correlation topology data between the any two types of dynamic data is determined based on the sum of Manhattan distances; the sum of Manhattan distances between mutation boundaries with same spatial code in the boundary distribution sequences corresponding to any two types of dynamic data is calculated, including:

[0016] When the number of mutation boundaries in the boundary distribution sequences corresponding to the any two types of dynamic data is inconsistent, the last mutation boundary in the type with less number of mutation boundaries is filled by equidistant interpolation based on the spatial position of the last mutation boundary, and the sum of Manhattan distances between mutation boundaries with same spatial code is calculated based on the filled sequence.

[0017] Preferably, the data preprocessing unit is specifically used for:

[0018] The first spatial data set, the second spatial data set and the third spatial data set are uniformly projected and converted according to a preset spatial coordinate system, to obtain a converted first spatial data set, a converted second spatial data set and a converted third spatial data set; and

[0019] The missing areas in the converted first spatial data set and the converted second spatial data set are completed by using a linear interpolation method, and the abnormal values in the converted third spatial data set are corrected by using a fixed threshold truncation method, to obtain a first target spatial data set, a second target spatial data set and a third target spatial data set.

[0020] Preferably, the first target spatial data set contains processed static geological data and processed dynamic geological data, and the data preprocessing unit is further used for:

[0021] The spatial coincidence degree of the processed static geological data and the processed dynamic geological data in a historical time period is calculated;

[0022] The predicted geological parameters of the processed dynamic geological data in a target region are predicted according to the spatial coincidence degree and the geological parameters of the processed static geological data in the target region;

[0023] A comprehensive geological data set is generated according to the processed dynamic geological data and the predicted geological parameters thereof, and the spatial data corresponding to the comprehensive geological data set is taken as the first target spatial data set.

[0024] Preferably, the first rendering layer includes a texture mapping unit and a feature superposition unit; wherein,

[0025] The texture mapping unit is configured to perform multi-resolution rasterization on each type of spatial data set contained in the dynamic data respectively, so as to extract corresponding texture feature data from each type of spatial data set;

[0026] The feature superposition unit is configured to perform multi-layer superposition on the texture feature data extracted from each type of spatial data set and the original spatial data set, so as to generate a spatial feature fusion model.

[0027] Preferably, the first rendering layer further comprises a feature dimension reduction unit configured to perform principal component analysis dimension reduction processing on the spatial feature fusion model.

[0028] The spatial coupling relationship between the spatial feature fusion models corresponding to each type of dynamic data is modeled, including: performing spatial coupling analysis on the dimension-reduced spatial feature fusion models corresponding to each type of dynamic data, to generate the spatial correlation topology data.

[0029] Preferably, the display layer comprises a visual interactive interface, which supports the user to switch the perspective, superimpose the layers and dynamically zoom the three-dimensional dynamic sand table through operation instructions.

[0030] The dynamic rendering generation based on the preset mine area three-dimensional model template and historical disaster data comprises: performing real-time updating on the geometric grid of the three-dimensional model template through an incremental rendering algorithm, so as to match the changes of the spatial correlation topology data.

[0031] The three-dimensional scene synthesis based on the spatial correlation topology data and the spatial feature fusion model comprises: performing hierarchical drawing on the spatial correlation topology data according to a preset rendering priority, and dynamically fusing with the spatial feature fusion model.

[0032] Preferably, the three-dimensional dynamic sand table comprises the following functional modules:

[0033] An offline data loading module is configured to integrate offline terrain data, disaster hazard distribution map and underground tunnel model data of the mine area, so as to ensure complete display of the three-dimensional scene in a network-free environment; the underground tunnel model comprises three-dimensional modeling of support structure, ventilation path and safety exit.

[0034] A disaster simulation module is configured to dynamically render a disaster diffusion path according to real-time monitoring data, and mark a risk level in the form of color gradient in the three-dimensional scene.

[0035] An emergency passage planning module is configured to automatically generate an optimal escape route in combination with the tunnel model and the disaster diffusion path, and support manual adjustment of path nodes.

[0036] Preferably, the three-dimensional dynamic sand table further comprises:

[0037] A layer management module is used for classified management of all space data in the sand table, supports user-defined layer grouping and display priority; the layer grouping includes basic geographic layer, geological disaster layer, real-time monitoring layer and emergency plan layer;

[0038] An element positioning module is used for quick positioning to specific roadway, equipment or hidden danger point according to user input key word or code, and marking target position in three-dimensional scene with high light flicker effect;

[0039] A data export module is used for exporting user-labeled emergency plan route, hidden danger point distribution and monitoring data into standard GIS format file.

[0040] Preferably, the three-dimensional visualization engine is realized based on a cross-platform rendering framework, and the cross-platform rendering framework is compatible with underlying hardware acceleration functions of different operating systems through abstracted graphic interface.

[0041] Compared with the prior art, the present application has the following beneficial effects:

[0042] In the aspect of data collection and processing, the data collection module integrates three types of dynamic data of geological structure, disaster hidden danger and environmental monitoring, wherein the geological structure data covers static exploration data and dynamic sensor data, and multi-dimensional and full-period monitoring of the geological environment of the mining area is realized. The data preprocessing unit effectively solves the problems of non-uniformity of spatial coordinate systems, data missing and abnormal values of multi-source data through unified projection conversion, linear interpolation completion and fixed threshold truncation, etc., and ensures the data accuracy and reliability input into the three-dimensional visualization engine. Especially for static and dynamic geological data, the spatial coincidence degree is calculated and the comprehensive geological data set is generated, which further improves the accuracy of geological parameter prediction and provides more solid data support for disaster warning.

[0043] The three-dimensional sand table generation module realizes efficient processing and realistic rendering of dynamic data through the innovative architecture of the three-dimensional visualization engine. The spatialization conversion of the data preprocessing unit lays the foundation for subsequent rendering, and the multi-level processing mechanism of the scene rendering unit is highly innovative. The first rendering layer generates a spatial feature fusion model through texture mapping and feature superposition, realizing the deep fusion of multiple spatial data sets, so that geological structures, disaster hazards and environmental monitoring data can intuitively present their respective features and mutual relationships in the three-dimensional scene; the principal component analysis processing of the feature dimension reduction unit reduces the data dimension while ensuring the integrity of the data features, improving the efficiency of subsequent spatial coupling analysis. The second rendering layer accurately identifies the spatial coupling relationship between different data through edge detection algorithm and Manhattan distance calculation, generates spatial correlation topology data, and can reveal the potential relationship between disaster hazards and geological structures, environmental changes, providing a scientific basis for disaster prediction and emergency decision-making. The display layer's visual interactive interface supports view switching, layer superposition and dynamic scaling operations, and cooperates with the incremental rendering algorithm and layered drawing technology to realize real-time updating and dynamic fusion of the three-dimensional dynamic sand table, enabling management personnel to intuitively and flexibly grasp the real-time situation of the mine area.

[0044] The functional module design of the three-dimensional dynamic sand table is closely related to the emergency management needs, and has significant practicality and innovation. The offline data loading module ensures that the system can still fully display the three-dimensional scene in a network-free environment, and the underground tunnel model provides reliable basic data for emergency rescue through fine modeling of support structures, ventilation paths and safety exits. The disaster simulation module dynamically renders the disaster diffusion path based on real-time monitoring data and marks the risk level, which helps to predict the development trend of disasters in advance and provides a basis for developing rescue strategies. The emergency passage planning module automatically generates the optimal escape route in combination with the tunnel model and the disaster diffusion path and supports manual adjustment, improving the efficiency and flexibility of emergency response. The layer management module realizes the classified management and custom display priority of spatial data, making it easy for users to quickly obtain key information according to different needs; the element positioning module's fast positioning and highlighting function helps to quickly lock the location of hidden danger points and equipment, improving the targeting of emergency disposal; the data export module supports exporting emergency plan routes, hidden danger point distribution and other data into standard GIS format files, facilitating data sharing and further analysis.

[0045] The system is implemented based on a cross-platform rendering framework, which abstracts the graphics interface to compatible with the underlying hardware acceleration functions of different operating systems, ensuring the consistency of the display effect and operation experience of the three-dimensional dynamic sand table on different terminal devices, reducing the system deployment and maintenance cost, and improving the system's applicability and scalability. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1A working principle diagram of the mine area three-dimensional visualization emergency management system of the present application;

[0047] Figure 2 A flow chart for generating spatial correlation topological data;

[0048] Figure 3 A flow chart for working of the data preprocessing unit;

[0049] Figure 4 A flow chart for three-dimensional scene synthesis of the display layer. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0051] Please refer to Figures 1-4 The present application relates to a mine area three-dimensional visualization emergency management system, which comprises a data acquisition module and a three-dimensional sand table generation module, and the specific implementation steps are as follows:

[0052] The data acquisition module is used to acquire real-time dynamic data of a target mine area, and the dynamic data comprises a first spatial data set corresponding to geological structure data, a second spatial data set corresponding to disaster hidden danger data, and a third spatial data set corresponding to environmental monitoring data. The geological structure data includes static geological data obtained from a geological exploration platform, such as long-term stable geological information of stratum distribution, rock type, geological fault, etc. in the mine area; and also includes dynamic geological data obtained through a sensor network deployed in the mine area, such as real-time monitoring data of rock stratum displacement, ground stress change, underground water level fluctuation, etc. which change with time. The second spatial data set corresponding to the disaster hidden danger data mainly covers data such as position, scale and stability index of disaster hidden danger points such as landslide, collapse and debris flow that may exist in the mine area. The third spatial data set corresponding to the environmental monitoring data includes real-time monitoring data of environmental parameters such as air quality (such as dust concentration, harmful gas content), water quality, noise, etc. in the mine area.

[0053] The three-dimensional sand table generation module is configured to input the dynamic data into a three-dimensional visualization engine for processing, and construct a three-dimensional dynamic sand table of the target mining area according to an output result of the three-dimensional visualization engine. The three-dimensional visualization engine comprises a data preprocessing unit and a scene rendering unit. The data preprocessing unit is configured to perform spatialization conversion on the dynamic data, specifically, convert dynamic data of different sources and different formats into a unified spatial data format for subsequent processing and rendering. The scene rendering unit performs dynamic rendering generation based on a preset three-dimensional model template of the mining area and historical disaster data. The scene rendering unit comprises a spatial mapping layer, a first rendering layer, a second rendering layer and a display layer connected in sequence. The first rendering layer is configured to perform texture mapping processing on a plurality of spatial data sets contained in the dynamic data to generate a spatial feature fusion model. The second rendering layer is configured to model a spatial coupling relationship between spatial feature fusion models corresponding to various types of dynamic data to generate spatial correlation topology data. The display layer is configured to perform three-dimensional scene synthesis based on the spatial correlation topology data and the spatial feature fusion model to generate a three-dimensional dynamic sand table.

[0054] The application will be further described below in combination with Examples 1 to 5:

[0055] Example 1

[0056] In the process of spatialization conversion of the dynamic data by the data preprocessing unit, two core processes need to be completed: unified projection conversion and data correction and completion. The present embodiment will be described in detail from the aspects of technical principle, operation steps and implementation logic.

[0057] I. Technical implementation of unified projection conversion

[0058] The preset spatial coordinate system is the reference framework of the entire data processing, and its selection needs to consider the geographical location of the mining area, industry standards and subsequent application requirements. For example, if the mining area is located in China and needs to be connected with the national basic geographic information, the national 2000 geodetic coordinate system is usually used; if the mining area is small or has special local modeling requirements, a local coordinate system can also be defined. The purpose of unified projection conversion is to convert spatial data sets (i.e. the first spatial data set, the second spatial data set and the third spatial data set) of different sources and different coordinate systems into a standard format under the same coordinate system, so as to eliminate problems such as data misplacement and inability to superimpose caused by coordinate system differences.

[0059] In a specific operation, for the first spatial data set (geological structure data), the static geological data can come from drilling data of a geological exploration platform, geological profile maps, etc. These data can be based on different local coordinate systems or projection methods when collected. The dynamic geological data is obtained by a sensor network deployed in the mining area, and the coordinates of the sensors are usually represented in the local coordinates of the device installation location. For the second spatial data set (disaster hazard data), it can be obtained by unmanned aerial vehicle aerial survey, ground survey, etc. The aerial survey data often uses the geographic coordinate system (latitude and longitude), while the ground survey data can use the plane rectangular coordinate system. The third spatial data set (environmental monitoring data) comes from various environmental sensors, and its coordinate system can be related to the deployment location and collection method of the sensors, such as a local coordinate system with the monitoring point as the origin.

[0060] The projection conversion process needs the help of the coordinate conversion tool of the geographic information system (GIS). First, identify the original coordinate system parameters of each spatial data set, including the datum, projection type, central meridian, scale factor, etc. For the first spatial data set, if the static geological data is based on the Beijing Coordinate System 1954, and the dynamic geological data is based on a self-defined local coordinate system, they need to be converted into a transition coordinate system first, and then converted into a preset target coordinate system. For the second spatial data set, if the disaster hazard point coordinates obtained by aerial survey are in the WGS84 geographic coordinate system, they need to be converted into the plane rectangular coordinate system through the Gauss-Kruger projection, and then unified with the plane coordinate system of other data sets. If the environmental monitoring data of the third spatial data set is represented by the relative coordinates of the sensors, the absolute coordinates of the sensors need to be determined (such as through GPS positioning or joint survey with the mining area control network), and then the relative coordinates are converted into absolute coordinates and included in the preset spatial coordinate system.

[0061] During the conversion process, attention needs to be paid to the preservation of data precision. For continuous data (such as rock layer displacement field, ground stress distribution), polynomial interpolation method is used for coordinate transformation to ensure the continuity and smoothness of the converted data; for discrete data (such as disaster hazard point coordinates, monitoring point positions), direct application of coordinate conversion formula for accurate calculation. After conversion, the converted first spatial data set, the converted second spatial data set, and the converted third spatial data set are obtained, at which time the three types of data have consistency in spatial position, and subsequent superposition analysis and rendering processing can be performed.

[0062] II. Specific process of data correction and completion

[0063] (I) Linear interpolation to complete missing data

[0064] There may be data missing areas in the converted first and second spatial data sets, due to reasons including sensor failure, data transmission interruption, uneven distribution of geological exploration boreholes, etc. For example, in geological structure data, if the stratum interface data is missing in a certain area due to too large borehole spacing, or the sensor fails to collect rock stratum displacement data for a certain period of time due to equipment failure, a data blank area will be formed.

[0065] The linear interpolation method is based on the assumption of spatial continuity of data, that is, it is assumed that the attribute values between adjacent known data points change linearly. The specific steps are as follows:

[0066] Determine the range of the missing area: Through data visualization or statistical analysis, identify the missing area in the first and second spatial data sets, such as a blank area with no data value in a two-dimensional plane, or a data fault in a three-dimensional space.

[0067] Select adjacent known data points: For each missing data point, search for a certain number of known data points in its neighborhood. The neighborhood range can be determined according to the data distribution density, such as selecting points within a 3x3 grid in a data dense area, and appropriately expanding the search range in a data sparse area.

[0068] Establish a linear interpolation model: Assume that the missing point and the adjacent known points have a linear relationship in the spatial coordinate axes (such as X, Y, Z axes), and fit a linear equation by least squares method. Taking a two-dimensional plane as an example, let the missing point coordinates be (x0, y0), the adjacent known point coordinates be (x i , y i ), and the corresponding attribute value be z i , then the linear interpolation equation can be expressed as z = a x + b y + c, and the coefficients a, b, c are obtained by solving the equation system, and then the attribute value z0 of the missing point is calculated.

[0069] Complete missing data point by point: Repeat the above steps for each data point in the missing area until all missing areas are completed. The completed data needs to be checked for reasonableness, such as comparing the attribute value trend of adjacent areas to ensure that the interpolation result conforms to the geological law or disaster hidden danger distribution characteristics, and to avoid abnormal mutations.

[0070] Through the linear interpolation method, the missing areas in the first and second spatial data sets can be effectively filled, making the data continuous and complete in space, and providing a reliable data basis for subsequent spatial feature fusion and coupling relationship analysis.

[0071] (II) Fixed threshold cut-off correction of abnormal values

[0072] There may be abnormal values in the converted third spatial data set (environmental monitoring data), such as dust concentration monitoring values exceeding the range, harmful gas content data suddenly becoming negative, etc. These abnormal values may be caused by sensor failure, electromagnetic interference, data transmission error, etc. If not corrected, it will lead to distortion of the environmental parameter distribution display in the three-dimensional scene, affecting the accuracy of emergency decision-making.

[0073] The core of the fixed threshold truncation method is to set a reasonable threshold range according to the physical meaning of the environmental monitoring parameters and historical statistical data, and forcibly correct the abnormal values that exceed the range to the threshold boundary value. The specific implementation steps are as follows:

[0074] Determine the threshold range: For each environmental monitoring parameter (such as dust concentration, CO concentration, O2 concentration, etc.), analyze its normal range of variation. For example, the normal range of dust concentration is usually 0-1000 mg / m 3 (depending on the operating environment of the mine), the safety threshold of CO concentration is generally 24 ppm (long-term exposure), and short-term high concentration may reach several hundred ppm, but there will be no negative values or values exceeding the sensor range (such as sensor range 0-1000 ppm, values exceeding 1000 ppm are considered abnormal). The threshold can be set by referring to industry standards, equipment manuals and statistical analysis results of historical monitoring data, such as taking the average value ± 3 times the standard deviation of historical data as the threshold range, or directly using the recommended warning threshold of the equipment manufacturer as the upper limit.

[0075] Identify abnormal values: Traverse each data point in the converted third spatial data set to determine whether its value exceeds the threshold range. For example, if the CO concentration monitoring value at a certain time is -5 ppm or 1500 ppm (assuming the sensor range is 0-1000 ppm), it is determined to be an abnormal value.

[0076] Truncate and correct abnormal values: For values exceeding the upper limit of the threshold, uniformly correct them to the upper limit of the threshold; for values below the lower limit of the threshold (such as negative values), uniformly correct them to the lower limit of the threshold (usually 0). For example, correct the CO concentration of -5 ppm to 0 ppm and 1500 ppm to 1000 ppm. The corrected data needs to be marked with an abnormality mark for subsequent tracing and troubleshooting of sensor failure.

[0077] Through the fixed threshold truncation method, abnormal interference in the third spatial data set can be effectively eliminated, ensuring the reasonableness and reliability of the environmental monitoring data, and making the environmental parameter display in the three-dimensional dynamic sand table conform to the actual working conditions, providing accurate data support for mine environmental safety evaluation and emergency response.

[0078] III. Processing results and data output

[0079] After unified projection conversion, linear interpolation completion and fixed threshold truncation correction, the first target spatial data set, the second target spatial data set and the third target spatial data set are obtained. Among them:

[0080] The first target spatial data set: contains processed static geological data and dynamic geological data. The static geological data is aligned with the dynamic geological data in the same coordinate system after coordinate conversion. The missing geological interface data is completed by linear interpolation. The data continuity and spatial consistency are good, and the geological structure characteristics and dynamic changes of the mining area can be fully displayed.

[0081] The second target spatial data set: after the coordinate unification and the missing completion of the disaster hidden danger data, the position, size and other information of the hidden danger point are accurate and continuous, which is convenient for analyzing the spatial distribution rule of the disaster hidden danger and the correlation with the geological structure.

[0082] The third target spatial data set: the abnormal values in the environmental monitoring data are corrected, and the data distribution conforms to the physical law and the actual environmental conditions, which can truly reflect the environmental quality of the mining area.

[0083] The three types of target spatial data sets serve as the data input of the three-dimensional visualization engine, providing high-quality data sources for subsequent texture mapping processing, spatial coupling relationship modeling and three-dimensional scene synthesis, ensuring that the three-dimensional dynamic sand table can accurately and intuitively display the real-time dynamic information of the mining area, and providing strong technical support for the emergency management of the mining area.

[0084] During the entire data preprocessing process, attention should be paid to the standardization and traceability of the processing flow, such as recording the parameter settings of the projection conversion, the neighborhood range of the linear interpolation, and the basis for setting the threshold, etc. so as to be consulted during the subsequent data verification or system maintenance. At the same time, the data before and after processing can be compared and analyzed through data visualization tools to intuitively show the effect of data preprocessing, ensuring the accuracy and effectiveness of data processing.

[0085] Example 2:

[0086] In the processing process of the first rendering layer, the texture mapping unit and the feature superposition unit work together to convert the dynamic data into a spatial feature fusion model with rich visual features. The following describes Example 2 in detail from the aspects of data processing principle, technical implementation path and system interaction logic.

[0087] I. Technical principle of multi-resolution rasterization processing

[0088] When the texture mapping unit performs multi-resolution rasterization on each type of spatial data set, it needs to follow the principle of hierarchical representation of spatial data. This principle is based on the characteristics of human visual perception: at different observation scales, users have different needs for data details. For example, at the macro scale, users are more concerned about the overall geological structure profile and disaster hazard distribution trend of the mine area; while at the micro scale, they need to view the details of the support structure of the specific roadway and the equipment layout. Multi-resolution rasterization realizes the differentiated expression of the same data at different scales by establishing a pyramid data structure.

[0089] For the first spatial data set of geological structure data, multi-resolution rasterization first discretizes the continuous geological body into regular grids. For example, a three-dimensional geological body is divided into cubic grid cells with a side length of 10 meters, and each grid cell is assigned a corresponding geological attribute value (such as lithology, density, porosity, etc.). At the coarse resolution level, a larger grid size (such as 100 meters) is used, and only the main geological interfaces and structural features are retained; as the resolution increases, the grid size gradually decreases (such as 50 meters, 20 meters, 10 meters), gradually increasing the secondary geological interfaces and local structural details. The data at each level is generated through aggregation algorithms, for example, the grid attribute value at the low resolution level is defined as the statistical characteristics (such as mean, median, maximum, etc.) of all high-resolution grid attribute values contained.

[0090] The multi-resolution processing of the second spatial data set of disaster hazard data focuses on the preservation of spatial distribution characteristics. For planar hazard areas such as landslides and collapses, a simplified polygon boundary is used at the low resolution level to retain their approximate location and area; at the high resolution level, the number of polygon vertices is increased to refine the boundary shape, and internal structural features (such as sliding direction, crack distribution, etc.) are added. For point-like hazard points (such as monitoring equipment locations, dangerous chemical storage points), a single symbol is used at the low resolution level, and a composite symbol containing more attribute information is used at the high resolution level (such as symbol size representing risk level, color representing hazard type).

[0091] The multi-resolution processing of the third spatial data set of environmental monitoring data needs to consider both time and space dimensions. For continuous distribution of environmental parameters such as air quality and water quality, spatial interpolation methods such as Kriging interpolation are used to convert discrete monitoring point data into continuous grid surfaces. At the low resolution level, the time sampling interval is long (such as 1 hour), and the spatial grid size is large (such as 100 meters), mainly reflecting the overall distribution trend of environmental parameters; at the high resolution level, the time sampling interval is shortened (such as 10 minutes), and the spatial grid size is reduced (such as 20 meters), which can capture rapid changes and local anomalies in environmental parameters. The data at each level is generated through time aggregation and spatial downsampling algorithms to ensure data consistency between different resolution levels.

[0092] II. Implementation Path of Texture Feature Extraction

[0093] Texture feature extraction is a key step to convert spatial data sets into visually perceptible textures. This process is based on feature extraction algorithms in computer vision, optimized with the professional characteristics of mining data.

[0094] For geological structure data, the gray level co-occurrence matrix (GLCM) is used to extract the texture of lithology distribution. The gray level co-occurrence matrix describes the spatial correlation of gray values in the image, and by calculating the joint probability distribution of gray values in different directions and distances, the statistical features of texture (such as contrast, correlation, energy, uniformity, etc.) are extracted. In geological data processing, geological attribute values (such as density, acoustic velocity) are mapped to gray values to generate pseudo-gray images, and then the gray level co-occurrence matrix is calculated. For example, layered geological structures are represented as high probability distribution along the diagonal direction in the gray level co-occurrence matrix, while fault structures are represented as local gray level mutations. By analyzing these statistical features, different lithology texture patterns can be identified, such as the granular texture of sandstone and the layered texture of shale.

[0095] The texture feature extraction of disaster hazard data focuses on the identification of spatial distribution patterns. For landslide hazard areas, Gabor filters are used to extract their surface texture features. Gabor filters are a kind of band-pass filter that can detect local texture changes in images at different scales and directions. By adjusting the parameters of the filter (such as wavelength, direction, bandwidth), the boundary features, internal crack texture and vegetation coverage differences of the landslide body can be extracted. For example, the surface texture of active landslides usually shows irregular block distribution, while stable areas have more continuous texture patterns. For collapse hazard points, local binary patterns (LBP) are used to extract their surface roughness features. LBP generates a binary pattern by comparing the gray values of the center pixel and the neighborhood pixels, which is used to describe the microstructure of local texture. The surface of the collapse body usually has high roughness, which is manifested as an increase in the diversity of LBP patterns.

[0096] The texture feature extraction of environmental monitoring data is based on spatio-temporal change pattern analysis. For air quality data, wavelet transform is adopted to extract the multi-scale features of its time series. Wavelet transform can decompose the signal into different frequency components, and analyze the change characteristics of the signal at different time scales. For example, short-term fluctuations in pollutant concentration can be represented by high-frequency components, while long-term trends can be reflected by low-frequency components. By mapping the wavelet coefficients at different scales to color or brightness values, texture images reflecting the spatio-temporal changes of air quality can be generated. For water quality data, principal component analysis (PCA) is used to extract the comprehensive features of multiple monitoring indicators. PCA can project high-dimensional data into low-dimensional space, retaining the main variation information of the data. By mapping the principal component scores obtained by PCA to texture parameters (such as color, transparency), the comprehensive status of water quality and its spatial distribution differences can be visually displayed.

[0097] III. Logic architecture of multi-layer superimposition to generate fusion model

[0098] When the extracted texture feature data is superimposed with the original spatial data set in multiple layers, the layered rendering and transparency mixing technology is adopted to realize the seamless fusion of data.

[0099] The entire superimposition process is based on the layered structure design of the three-dimensional scene. The bottom layer is the base terrain layer, which is composed of the digital elevation model (DEM) and orthophoto map of the mining area, providing the true surface morphology and texture. Above the base terrain layer, the geological structure layer, disaster hidden danger layer and environmental monitoring layer are superimposed in turn. Each layer of data is rendered according to its spatial position and attribute characteristics, and the visibility and fusion degree between layers are controlled through transparency setting.

[0100] The superimposition of the geological structure layer adopts volume rendering technology. The multi-resolution rasterized geological data is combined with the extracted lithological texture features, and the absorption, scattering and emission effects of light passing through the geological body are calculated through the ray casting algorithm to realize the visualization of the internal structure of the geological body. For example, the texture features of different lithologies are modulated by color and transparency, making the geological interface and internal structure clearly visible. For special structures such as faults and folds, the contrast and color saturation of the boundary texture are enhanced to highlight the spatial form and distribution rule.

[0101] The superimposition of the disaster hidden danger layer adopts symbolization and buffer analysis technology. The extracted disaster hidden danger texture features are combined with the spatial position information of the original hidden danger data to generate a symbol layer with visual warning effect. For example, for landslide hidden danger areas, red semi-transparent polygons are used to represent their impact range, and the interior is filled with landslide texture patterns; for collapse hidden danger points, symbols with warning signs are used, and the symbol size and color are dynamically adjusted according to the risk level. At the same time, different levels of risk areas are generated based on buffer analysis, and the visualization transition of risk levels is realized through transparency gradient.

[0102] The superposition of the environmental monitoring layer adopts a combination of isosurface and volume rendering techniques. For continuously distributed environmental parameters (such as dust concentration, harmful gas content), isosurfaces are generated by interpolation, and the extracted spatiotemporal texture features are mapped onto the isosurfaces to represent the size and trend of parameter values through color and transparency changes. For discrete monitoring point data, volume rendering techniques are used to generate the spatial distribution field of monitoring data, and texture features are associated with data values to make the spatial distribution of environmental parameters more intuitive. For example, high concentration areas are represented in red, and low concentration areas are represented in green, and the texture density reflects the gradient of parameter values.

[0103] In the multi-layer superposition process, by adjusting the transparency, brightness, contrast and other parameters of each layer, the organic integration of different types of data is realized. For example, the transparency of the geological structure layer is set to 70%, so that part of the terrain texture of the bottom layer is visible, enhancing the realism of the scene; the transparency of the disaster hazard layer is set to 50%, highlighting the hazard area without blocking the geological and topographic information of the bottom layer; the transparency of the environmental monitoring layer is dynamically adjusted according to the reliability and importance of the data, ensuring that users can clearly perceive the distribution of environmental parameters. Through this multi-layer superposition method, the spatial feature fusion model generated can simultaneously display multiple types of dynamic data and their mutual relationships, providing comprehensive and intuitive decision support for mine emergency management.

[0104] Embodiment 3:

[0105] When modeling the spatial coupling relationship between the spatial feature fusion models corresponding to various types of dynamic data in the second rendering layer, edge detection, distance calculation and sequence filling steps are required. The following describes embodiment 3 in detail from the aspects of technical principles, operation processes and data processing logic, etc.

[0106] The edge detection algorithm is designed based on the physical characteristics of mine dynamic data, and different types of data use different detection strategies. For geological structure data, its abrupt boundary usually corresponds to the discontinuous surface of geological structures such as rock layer interface and fault zone. When using the Canny edge detection algorithm, Gaussian smoothing of the spatial feature fusion model is required to reduce the influence of data noise. The size of the Gaussian smoothing kernel is determined according to the accuracy and noise level of the geological data, and a convolution kernel of 3x3 or 5x5 is generally selected. Next, the amplitude and direction of the image gradient are calculated, and the edge is refined through the non-maximum suppression technique, and only the local gradient maximum points are retained as candidate edge points. Finally, a double-threshold method is used to determine the true edge, the low threshold is used to connect the edge fragments, and the high threshold is used to determine the starting point of the edge, effectively identifying the abrupt boundary in the geological structure and forming the boundary distribution sequence.

[0107] The mutation boundary identification of disaster hazard data focuses on the division between the hazard area and the normal area. For example, the boundary of a landslide body usually shows a sudden change in parameters such as terrain slope and surface displacement rate. For such data, an edge detection algorithm based on region growing is used. First, mark the known hazard points as seed points in the spatial feature fusion model, then gradually expand the region according to the preset growth criteria (such as the attribute difference of adjacent pixels being less than a threshold), until the boundary condition is reached. The boundary condition can be that the attribute difference of adjacent regions exceeds the threshold, or that the edge of the image is reached. In this way, the boundaries of disaster hazard areas such as landslides and collapses can be accurately identified, forming the corresponding boundary distribution sequence.

[0108] The mutation boundary identification of environmental monitoring data mainly targets areas with sharp changes in parameters such as pollutant concentration, temperature and humidity. For example, in a mine, a harmful gas leakage point usually forms an area with a large concentration gradient. For such data, an edge detection algorithm based on gradient operators is used, such as the Sobel operator or the Prewitt operator. These operators detect areas with sharp changes in grayscale values by calculating the gradient in the horizontal and vertical directions of the image. In environmental monitoring data processing, the parameter values are mapped to grayscale values, then the gradient operator is used to calculate the gradient amplitude, and the area with a larger amplitude is the mutation boundary. By thresholding the gradient amplitude image, the mutation boundary of the environmental parameter is extracted, forming the boundary distribution sequence.

[0109] When calculating the sum of the Manhattan distances between mutation boundaries with the same spatial code in the boundary distribution sequences of any two types of dynamic data, the boundary distribution sequences need to be spatially coded first. Spatial coding uses a quadtree or octree structure to divide the three-dimensional space into grid cells at different levels. Each grid cell is assigned a unique code, with the code length representing the spatial resolution and the code value representing the cell's position in space. For the boundary distribution sequences of geological structure data and disaster hazard data, map each mutation boundary point to the corresponding grid cell to obtain its spatial code.

[0110] Before calculating the Manhattan distance, the boundary distribution sequences need to be aligned. Due to differences in data collection density and accuracy, the number of mutation boundaries in the boundary distribution sequences may not be consistent. To ensure that the distance between mutation boundaries with the same spatial code can be calculated, the sequences need to be padded. Based on the spatial position of the last mutation boundary in the sequence with fewer mutation boundaries, insert virtual mutation boundaries around it at a certain interval. The insertion interval is determined according to the spatial distribution characteristics of the data and the calculation accuracy requirements, and is usually half of the average distance between original data points.

[0111] After filling, for each pair of mutation boundaries with the same spatial code in the sequence, the Manhattan distance between them is calculated. In three-dimensional space, the Manhattan distance is defined as the sum of the absolute distances of two points on the X, Y, Z coordinate axes, that is, d = |x1-x2| + |y1-y2| + |z1-z2|. Traverse all pairs of mutation boundaries with the same spatial code, and accumulate their Manhattan distances to obtain the sum of the Manhattan distances between any two types of dynamic data. This distance sum reflects the degree of spatial similarity of the mutation boundaries of the two types of data. The smaller the distance, the closer the mutation boundaries of the two types of data in space, and the stronger the spatial coupling relationship between them.

[0112] In some cases, there may be some mutation boundary points that cannot find corresponding points with the same spatial code. For these points, match them with the nearest mutation boundary points with different spatial codes, and calculate the Manhattan distance between them. To avoid the influence of such inaccurate matching on the results, when accumulating the distance sum, multiply the distance value of such inaccurate matching by a weight coefficient less than 1, which is determined according to the degree of spatial code difference between the matching points. The greater the difference, the smaller the weight coefficient.

[0113] Based on the calculated sum of Manhattan distances, the spatial correlation topology data between any two types of dynamic data is determined. The spatial correlation topology data is represented in a graph structure, where nodes represent different types of dynamic data, edges represent the spatial correlation relationship between them, and the weight of the edge is the inverse of the sum of the Manhattan distances. In this way, the spatial distance relationship is converted into the correlation strength relationship, and the smaller the distance, the greater the correlation strength. This spatial correlation topology data provides important spatial correlation information for subsequent three-dimensional scene synthesis, which can help the system correctly handle the spatial relationship between different types of data during rendering, making the generated three-dimensional dynamic sand table more realistically reflect the actual situation of the mining area.

[0114] Embodiment 4:

[0115] When performing three-dimensional scene synthesis based on the spatial correlation topology data and the spatial feature fusion model in the display layer, the visualization of the data needs to be realized through layered rendering, dynamic fusion, and interactive interface design. The following describes Embodiment 4 in detail from the aspects of scene construction principle, rendering technology implementation, and user interaction mechanism.

[0116] The layered rendering process is based on the importance level division of spatial correlation topology data, and different types of data are given different rendering priorities. Geological structure data, as the basic layer, carries the underlying structural information of the mining area and is set to the lowest rendering priority. Its rendering uses volume rendering technology to discretize the three-dimensional geological body into a regular grid, and calculates the absorption, scattering and transmission effects of light passing through the geological body through the ray casting algorithm. To enhance the display effect of the geological interface, different color mapping schemes are applied to different lithology areas, such as yellow for sandstone, gray for shale, and black for coal seams. At the same time, texture details such as bedding structure and fault scratches are added at the geological interface, and the micro-geological structure is simulated through normal perturbation technology.

[0117] Disaster hazard data, as key information for safety warning, is given the highest rendering priority. For landslide, collapse and other planar hazard areas, semi-transparent polygons are used to superimpose texture patterns. The polygon boundary is emphasized with a red thick line, and the inside is filled with irregular textures to represent the activity level of the hazard. The texture pattern uses a perturbed pattern generated based on Berlin noise to simulate surface deformation and cracks. For point-like hazard points, different sizes and colors of warning symbols are designed according to the risk level. Low-risk points are represented by yellow triangles, medium-risk points are represented by orange pentagons, and high-risk points are represented by red explosion symbols. The symbols use Billboard technology to always face the camera, ensuring that they are clearly visible from any angle.

[0118] The rendering priority of environmental monitoring data is between geological structure and disaster hazard. For continuously distributed environmental parameters (such as dust concentration and harmful gas content), isosurfaces are combined with volume rendering. First, the discrete monitoring point data is converted to a continuous scalar field using the Kriging interpolation algorithm, and then the isosurfaces are extracted using the MarchingCubes algorithm. The color of the isosurface is mapped according to the parameter value, such as green for low concentration, yellow for medium concentration, and red for high concentration. To enhance the spatial hierarchy, the isosurface is added with a transparency gradient, and the transparency of the isosurface in the distance is increased. For discrete monitoring device locations, icons with dynamic particle effects are used to represent them, and the color and density of the particles reflect the changing trend of real-time monitoring data.

[0119] In the dynamic fusion stage, the system listens to the updates of spatial correlation topology data and spatial feature fusion model in real time. When new data arrives, the system first verifies the validity of the data to check its integrity and reasonableness. For updates of geological structure data, such as rock layer displacement monitoring data, the system uses a smooth transition algorithm to achieve seamless switching between old and new data. By calculating the interpolation function between the old and new data fields, the system gradually transitions from old data to new data within a certain time interval (such as 2 seconds), avoiding sudden changes in the scene.

[0120] The update of disaster hazard data uses a warning animation effect. When the stability index of a landslide hazard area changes, the boundary of the area will flash red light, and the disturbance frequency of the internal texture pattern will increase. The system dynamically adjusts the intensity of the warning effect according to the change of the hazard level, and the flashing frequency and brightness of the high-risk area are higher. For newly added hazard points, a growing animation from nothing to something is used, and the icon gradually enlarges from a small point to the full size, accompanied by a wave effect spreading outward, attracting the user's attention.

[0121] The update of environmental monitoring data uses fluid simulation technology. When harmful gas leakage is detected, the system renders the gas diffusion process in real time in the three-dimensional scene according to the location of the leakage source and the diffusion model. The diffusion effect uses a physics-based fluid simulation algorithm, taking into account factors such as gas density, wind speed, and terrain. The gas cloud is realized using a semi-transparent particle system, and the color and transparency of the particles are dynamically adjusted according to the gas concentration. At the same time, streamlines are added to the gas diffusion path to visually show the direction of gas flow.

[0122] The visualization interface is based on WebGL technology and is compatible across platforms, supporting multiple operation modes such as mouse, keyboard, and touch devices. The perspective switching function provides multiple preset perspectives, such as global overhead view, internal tunnel roaming, and hazard point close-up. Users can quickly switch by clicking the perspective buttons on the interface, or freely rotate the scene by dragging the mouse and zoom in or out using the scroll wheel. To ensure smooth operation, the system uses multi-threading technology to separate rendering calculations from user interaction processing, avoiding interface lag.

[0123] The layer superimposition function allows users to display or hide specific types of data as needed. The interface is designed with a layer management panel on the left, listing all available layers, including base terrain, geological structure, disaster hazards, and environmental monitoring. Each layer is accompanied by a checkbox and a transparency slider, allowing users to control the display state of the layer by checking the checkbox and adjust the layer's transparency by dragging the slider. When there are many layers, the panel supports folding and unfolding functions, allowing users to fold layers that are not currently in use to save interface space.

[0124] The dynamic scaling function uses a progressive loading strategy. When the user zooms in on the scene, the system automatically loads higher resolution data. For example, in the global view, the system only loads low-resolution terrain models and sketchy geological structure data; when the user zooms in to the tunnel level, the system will asynchronously load high-resolution geological data, tunnel support structure data, and equipment distribution data for that area. To avoid blank areas during loading, the system displays temporary placeholder graphics, such as low-precision grid models or blurred textures.

[0125] During the interaction, the system provides a real-time feedback mechanism. When the user selects a certain object (such as a hidden danger point or a monitoring device), the object is highlighted, and an information panel is popped up to display its detailed attributes. The information panel adopts a card design, containing basic information, historical data charts, and related multimedia materials. The user can perform further operations through the buttons on the panel, such as viewing historical trend changes or calling up on-site monitoring videos. To improve user experience, all interactive operations are equipped with appropriate animation effects, such as the ripple effect of button clicks and the smooth transition of panel expansion.

[0126] Embodiment 5:

[0127] The integration of functional modules and cross-platform implementation of the three-dimensional dynamic sand table need to be completed through the cooperation of multiple technologies. The following elaborates on Embodiment 5 from the aspects of module design principles, data processing flow, and system architecture, etc.

[0128] The offline data loading module adopts a hierarchical storage strategy, classifying and storing the mine area terrain data, disaster hidden danger distribution map, and underground tunnel model data into different file formats. The terrain data adopts high-precision DEM (Digital Elevation Model) format, storing it as a regular grid elevation value matrix, while supplemented by texture mapping technology, mapping satellite images or aerial photography images onto the terrain surface to enhance the sense of reality. The disaster hidden danger distribution map is stored in vector data format, with each hidden danger point or hidden danger area containing detailed attribute information (such as type, level, formation time, etc.), and fast querying and positioning is achieved through spatial indexing technology. The underground tunnel model data adopts BIM (Building Information Modeling) standard format, containing geometric information, material attributes, construction time, and other parameters of the support structure, while the ventilation path and safety exit information is stored in the form of a topological network, recording the connection relationship between nodes (such as tunnel intersections, ventilation openings) and edges (tunnel segments).

[0129] In a network-free environment, the system preloads the offline data into local storage through a preloading mechanism. At startup, it first loads the basic terrain data and main tunnel model, providing basic three-dimensional scene browsing functions. The user can select the area of interest through the navigation controls on the interface, and the system will dynamically load the detailed data of the area according to the user's selection. To reduce data loading, a spatial indexing structure such as quadtree or octree is used to organize the data in blocks, and only the data blocks within a certain range of the current visible area and its periphery are loaded. At the same time, the system supports data compression technology, reducing the data storage volume without affecting the display precision and improving the loading speed.

[0130] The disaster simulation module is based on fluid dynamics and discrete element theory to construct a disaster diffusion model. For gas disasters such as gas and dust, the Navier-Stokes equation is used to describe fluid motion, taking into account factors such as roadway structure, ventilation conditions, and obstacles affecting airflow. The finite difference method or finite element method is used to discretely solve the equations, dividing the calculation area into regular grids and updating fluid state parameters (such as velocity, pressure, and concentration) at each time step. To improve computational efficiency, GPU acceleration technology is used to distribute calculation tasks to multiple GPU cores for parallel processing.

[0131] When rendering the disaster diffusion path, the system uses a combination of particle systems and volume rendering. For large-scale diffusion areas, volume rendering technology is used to generate continuous concentration field visualization effects, with different concentration levels represented by color mapping. For local high-concentration areas or leakage sources, particle systems are used to simulate discrete gas movement, with particle size, color, and motion trajectory dynamically adjusted based on calculated fluid parameters. Risk level labeling uses a combination of color gradient and transparency changes, with semi-transparent risk area polygons overlaid in the three-dimensional scene, filled with a gradient color from green (low risk) to red (high risk), and transparency gradually decreasing with distance from the leakage source, creating a sense of hierarchy.

[0132] The emergency passage planning module uses the A* algorithm as the core path search algorithm, combining the topological structure of the roadway model and the disaster diffusion path information to generate the optimal escape route. First, the underground roadway network is abstracted as a graph structure, with nodes representing key locations such as roadway intersections and safety exits, and edges representing roadway segments. The edge weights are determined based on factors such as roadway length, difficulty of passage, and risk level. In the event of a disaster, the edge weights are dynamically adjusted based on the real-time disaster diffusion path, with higher-risk areas having higher weights.

[0133] The A* algorithm searches for the optimal path by maintaining an open list and a closed list. Starting from the starting point, the estimated cost (actual cost + heuristic function value) of each adjacent node is calculated, the node with the smallest estimated cost is added to the closed list, and its adjacent nodes are added to the open list. This process is repeated until the end point is found or the open list is empty. In the roadway network, the heuristic function uses Manhattan distance, which takes into account the grid-like structure of the roadway and can more accurately estimate the shortest path.

[0134] The system supports manual adjustment of path nodes, allowing users to add, remove, or move nodes on the automatically generated escape route, and the system will recalculate the path based on the new node positions. To facilitate user operation, an intuitive drag-and-drop interface is provided, allowing users to change the path direction by clicking and dragging the nodes on the path. At the same time, the system will assess the feasibility of the new path in real time, ensuring that the adjusted path does not pass through high-risk areas or impassable obstacles.

[0135] The layer management module organizes all spatial data in the sand table in a tree structure. The root node is the entire three-dimensional scene, and the child nodes are divided into basic geographic layers, geological disaster layers, real-time monitoring layers, and emergency plan layers. Each category is further divided into specific layers, such as the basic geographic layer containing terrain, landform, and water system layers; the geological disaster layer containing landslide, collapse, and debris flow layers. Users can quickly locate the layers they need to operate by expanding or folding the layer groups through the tree control on the interface.

[0136] The custom layer grouping function allows users to create personalized layer collections based on their work needs. Users can add different types of layers to custom groups through drag-and-drop operations, and set names and icons for the groups. Display priority is set through numerical ordering, with each layer assigned a unique integer value. The smaller the value, the higher the display priority. The system renders each layer in order of priority, ensuring that high-priority layers are not obscured by low-priority layers.

[0137] The feature positioning module implements fast positioning based on spatial indexing technology. Spatial indexing is established for each tunnel, device, and hidden danger point, recording their geometric position and attribute information. When the user inputs keywords or codes, the system first performs fuzzy matching in the attribute database to filter out possible target features. Then, based on the spatial position of the target feature, the view angle of the three-dimensional scene is adjusted so that the target feature is located in the center of the viewport.

[0138] The highlight flicker effect is achieved by modifying the rendering attributes of the target feature. For point features, increase their display size and change the color to bright red, while adding periodic transparency changes to create a flickering effect. For surface features, add a glowing outline to the boundary, with the brightness and thickness of the outline changing periodically over time. To ensure that the highlight effect does not affect the display of other features, use depth testing and stencil buffer technology to limit the highlight effect to the geometric range of the target feature.

[0139] The data export module supports the export of multiple standard GIS format files. For emergency plan routes, the system converts them to ESRIShapefile format, containing geometric information (points, lines) and attribute information (route name, length, risk level, etc.) of the route. For hidden danger point distribution data, export as GeoJSON format, retaining detailed information such as the coordinates, type, and level of each hidden danger point. Monitoring data is exported as CSV format, containing time series data and spatial location information.

[0140] During the export process, the system performs format conversion and coordinate system unification. It converts the coordinates in the three-dimensional scene to the coordinate system required by the target GIS format, ensuring that the exported data can be correctly displayed and analyzed in other GIS software. At the same time, the data is compressed and optimized to remove redundant information, improving the efficiency of file transmission and processing. Users can select the export format, file path, and data range through the export dialog box on the interface, and the system will perform data processing and file generation in the background, providing progress feedback.

[0141] The cross-platform rendering framework is based on the WebGL2.0 standard and achieves compatibility with different operating system underlying hardware acceleration functions by abstracting the graphics interface. The framework defines a unified API that encapsulates the differences between underlying graphics hardware, allowing upper-layer application code to be free from specific graphics driver implementation. On Windows systems, the framework utilizes the hardware acceleration functions provided by DirectX11 / 12; on Linux systems, it uses OpenGLES3.0; and on macOS systems, it supports the Metal graphics API.

[0142] To ensure rendering performance, the framework adopts a multi-threaded rendering architecture. The main thread is responsible for handling user interactions and scene management, while the rendering thread performs actual graphics rendering operations. The double-buffering technique is used to perform scene rendering in the background buffer, and then exchange it with the foreground buffer to avoid flickering during rendering. At the same time, texture compression and resource management mechanisms are implemented to optimize the storage and loading of resources such as textures and models, reducing memory usage and loading time.

[0143] On mobile devices, the framework optimizes the interaction experience for touchscreens. It implements gestures such as touch dragging, two-finger scaling, pinch rotating, and other gestures, allowing users to easily navigate and manipulate three-dimensional scenes on small screens. At the same time, adaptive rendering technology is used to dynamically adjust rendering quality and detail levels based on device performance, ensuring smooth operation on low-configuration devices while maintaining visual effects. Through this cross-platform rendering framework, the system can provide consistent user experience and high-performance three-dimensional rendering effects on desktops, mobile devices, and other devices.

[0144] It should be noted that in this article, relational terms such as first and second are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such a process, method, article or device.

[0145] While embodiments of the application have been shown and described, it is to be understood that the application is not limited to the details of the embodiments described, since numerous changes, modifications, substitutions and variations can be made thereto without departing from the spirit and scope of the application as defined by the appended claims and their equivalents.

Claims

1. A three-dimensional visual emergency management system for mining areas, characterized in that: include: A data acquisition module is used to obtain real-time dynamic data of the target mining area, wherein the dynamic data includes a first spatial data set corresponding to geological structure data, a second spatial data set corresponding to disaster hazard data, and a third spatial data set corresponding to environmental monitoring data. The geological structure data includes static geological data obtained from the geological exploration platform and dynamic geological data obtained through the sensor network deployed in the mining area; A three-dimensional sand table generation module is used to input the dynamic data into a three-dimensional visualization engine for processing; Constructing a three-dimensional dynamic sand table of the target mining area according to the output results of the three-dimensional visualization engine; The 3D visualization engine includes a data preprocessing unit and a scene rendering unit, wherein the data preprocessing unit is used to perform spatial conversion on the dynamic data, and the scene rendering unit performs dynamic rendering and generation based on a preset mining area 3D model template and historical disaster data; the scene rendering unit includes a spatial mapping layer, a first rendering layer, a second rendering layer, and a display layer connected in sequence, and the first rendering layer is used to perform texture mapping processing on multiple spatial data sets contained in the dynamic data to generate a spatial feature fusion model; The second rendering layer is used to model the spatial coupling relationship between the spatial feature fusion models corresponding to various types of dynamic data to generate spatial correlation topology data; the display layer is used to synthesize three-dimensional scenes based on the spatial correlation topology data and the spatial feature fusion model to generate a three-dimensional dynamic sandbox.

2. A mining area three-dimensional visualization emergency management system according to claim 1, characterized in that: The spatial coupling relationship between the spatial feature fusion models corresponding to various types of dynamic data is modeled to generate spatial correlation topology data, including: An edge detection algorithm is used to identify mutation boundaries in the spatial feature fusion model, and a boundary distribution sequence corresponding to each type of dynamic data is determined based on the spatial position of each mutation boundary; Calculating the sum of Manhattan distances between mutation boundaries with the same spatial encoding in boundary distribution sequences corresponding to any two types of dynamic data, and determining spatial correlation topological data between the any two types of dynamic data based on the sum of Manhattan distances; calculating the sum of Manhattan distances between mutation boundaries with the same spatial encoding in boundary distribution sequences corresponding to any two types of dynamic data, including: When the number of mutation boundaries in the boundary distribution sequences corresponding to any two types of dynamic data is inconsistent, equidistant interpolation filling is performed based on the spatial position of the last mutation boundary in the sequence with the smaller number of mutation boundaries, and the sum of the Manhattan distances between the mutation boundaries with the same spatial encoding is calculated based on the filled sequence.

3. The three-dimensional visualization emergency management system for mining areas according to claim 1, characterized in that: The data preprocessing unit is specifically used for: performing a unified projection transformation on the first spatial dataset, the second spatial dataset, and the third spatial dataset according to a preset spatial coordinate system to obtain a transformed first spatial dataset, a transformed second spatial dataset, and a transformed third spatial dataset; as well as A linear interpolation method is used to fill in the missing areas in the converted first spatial dataset and the converted second spatial dataset, and a fixed threshold truncation method is used to correct abnormal values ​​in the converted third spatial dataset to obtain a first target spatial dataset, a second target spatial dataset, and a third target spatial dataset.

4. A three-dimensional visualization emergency management system for mining areas according to claim 3, characterized in that: The first target spatial data set includes processed static geological data and processed dynamic geological data, and the data preprocessing unit is further used to: Calculating the spatial overlap between the processed static geological data and the processed dynamic geological data within a historical time period; Predicting the predicted geological parameters of the processed dynamic geological data in the target area based on the spatial overlap and the geological parameters of the processed static geological data in the target area; A comprehensive geological data set is generated based on the processed dynamic geological data and its predicted geological parameters, and spatial data corresponding to the comprehensive geological data set is used as the first target spatial data set.

5. The three-dimensional visualization emergency management system for mining areas according to claim 1, characterized in that: The first rendering layer includes a texture mapping unit and a feature superposition unit; wherein, The texture mapping unit is used to perform multi-resolution rasterization processing on each type of spatial data set included in the dynamic data, so as to extract corresponding texture feature data from each type of spatial data set; The feature superposition unit is used to perform multi-layer superposition of texture feature data extracted from each type of spatial data set with the original spatial data set to generate a spatial feature fusion model.

6. A mining area three-dimensional visualization emergency management system according to claim 5, characterized in that: The first rendering layer further includes a feature dimensionality reduction unit, which is used to perform principal component analysis dimensionality reduction processing on the spatial feature fusion model; The modeling of the spatial coupling relationship between the spatial feature fusion models corresponding to various types of dynamic data includes: performing spatial coupling analysis on the spatial feature fusion models corresponding to various types of dynamic data after dimensionality reduction processing to generate the spatial correlation topology data.

7. The three-dimensional visualization emergency management system for mining areas according to claim 1, characterized in that: The display layer includes a visual interactive interface, which supports users to switch perspectives, overlay layers, and dynamically scale the three-dimensional dynamic sandbox through operation instructions; The dynamic rendering generation based on the preset mining area three-dimensional model template and historical disaster data includes: updating the geometric grid of the three-dimensional model template in real time through an incremental rendering algorithm to match the changes in the spatial correlation topology data; The three-dimensional scene synthesis based on the spatial correlation topology data and the spatial feature fusion model includes: layering the spatial correlation topology data according to a preset rendering priority, and dynamically fusing it with the spatial feature fusion model.

8. The three-dimensional visualization emergency management system for mining areas according to claim 1, characterized in that: The three-dimensional dynamic sandbox includes the following functional modules: An offline data loading module integrates offline topographic data, hazard distribution maps, and underground tunnel model data from the mining area to ensure a complete 3D scene display in an offline environment. The underground tunnel model includes 3D modeling of support structures, ventilation paths, and emergency exits. The disaster simulation module is used to dynamically render the disaster diffusion path based on real-time monitoring data and mark the risk level in the form of color gradient in the three-dimensional scene; The emergency channel planning module is used to automatically generate the optimal escape route by combining the tunnel model and the disaster diffusion path, and supports manual adjustment of path nodes.

9. A three-dimensional visualization emergency management system for mining areas according to claim 8, characterized in that: The three-dimensional dynamic sandbox also includes: The layer management module is used to classify and manage all spatial data in the sandbox, and supports user-defined layer grouping and display priority; the layer grouping includes basic geographic layers, geological disaster layers, real-time monitoring layers, and emergency plan layers; The element positioning module is used to quickly locate specific lanes, equipment or potential hazards based on keywords or codes entered by the user, and mark the target location with a highlighted flashing effect in the 3D scene; The data export module is used to export the emergency plan routes, hidden danger point distribution and monitoring data marked by the user into standard GIS format files.

10. The three-dimensional visualization emergency management system for mining areas according to claim 1, characterized in that: The three-dimensional visualization engine is implemented based on a cross-platform rendering framework, which is compatible with the underlying hardware acceleration functions of different operating systems through an abstract graphics interface.

Images