A coal mining subsidence prediction method and system based on cloud computing

Abstract

The application relates to a coal mining subsidence prediction method and system based on cloud computing and relates to the technical field of coal mining subsidence prediction, which comprises the following steps: a user collects mining subsidence geological information through a data acquisition module and uploads the information to a Web server; a geological modeling analysis module performs grid division, data interpolation, elimination and summarization on the geological information, forms an underground mining differential unit information file and a ground calculation point coordinate file; a mining subsidence prediction calculation module divides a calculation process into tasks and performs parallel calculation, adopts a probability integral method to calculate the mining subsidence deformation of each mining differential unit on a ground point and uploads the deformation to a Web database; a graphic processing module performs graphic processing on a prediction result, generates a deformation amount drawing and a deformation extreme value statistical table and feeds back the drawing and the table to the Web server; and a man-machine interaction module is used to download a calculation result. The prediction calculation and the graphic processing are performed in the cloud, the software cost and the operation difficulty of the user are reduced, and the automation degree and the calculation efficiency are improved.

Description

Technical Field

[0001] This application relates to the field of coal mining subsidence technology, and in particular to a cloud computing-based method and system for predicting coal mining subsidence. Background Technology

[0002] Coal mine subsidence prediction is a crucial component of coal mine design and environmental impact assessment. After large-scale mining of underground coal resources, the overlying rock strata lose their original mechanical support and shift, leading to subsidence phenomena such as surface sinking, tilting, horizontal movement, and curvature deformation. Mining subsidence prediction technology aims to calculate and predict the deformation of the mining area and surrounding surfaces in advance using mathematical models and relevant geological survey data. It visually demonstrates the patterns and extent of surface deformation within the predicted area, thus providing reliable data support for mine environmental protection, surface building protection, and geological disaster prevention.

[0003] In existing coal mine subsidence prediction technologies, most calculations and drawing generation are performed using desktop software or applications running locally on a single computer. Existing desktop subsidence prediction software typically requires engineers to install complex professional software environments on their local computers and manually input large amounts of borehole data, contour lines, and coal seam thickness information for geological modeling and calculations. After obtaining the basic calculation results, many existing prediction software programs cannot independently automate the mapping process to transform the abstract calculation matrix into intuitive deformable contour maps. Instead, they require secondary development using third-party professional drawing software (such as the foreign software Surfer) or manual data import for mapping. During this process, users need to perform tedious manual configuration, format conversion, and map debugging for massive amounts of scattered data.

[0004] From the existing methods of single-machine subsidence prediction and reliance on third-party desktop software for mapping, the following major technical problems can be deduced: The current model requires users to purchase and install specialized mapping software, which not only significantly increases the cost of software licensing and use but also leads to a heavy dependence on foreign software copyrights within the industry. Simultaneously, the entire calculation and mapping process is highly dependent on the processing power of the local computer. Since mining subsidence prediction involves complex spatial calculations of massive surface grid points and underground differential units, single-machine computation is extremely time-consuming, severely consuming engineers' computer resources and time. Furthermore, the data processing and drawing generation process is complex and lacks automation, resulting in difficult human-computer interaction and extremely low overall prediction efficiency. Summary of the Invention

[0005] To address the aforementioned issues, this application provides a cloud computing-based method and system for predicting coal mine subsidence.

[0006] Firstly, this application provides a cloud computing-based method for predicting coal mine subsidence, employing the following technical solution: A cloud computing-based method for predicting coal mine subsidence includes the following steps: S1. The data acquisition module collects geological information on mining subsidence and submits it to the Web server through the human-computer interaction module. S2. The geological modeling and analysis module performs grid division and data interpolation on the uploaded mining subsidence geological information to form an underground mining differential unit information file and a ground calculation point coordinate file, which are then uploaded to the Web database. S3. The mining subsidence prediction calculation module acquires the task and uses parallel computing to calculate the mining subsidence deformation of ground points based on the underground mining differential unit information file and the ground calculation point coordinate file using the probability integral method, obtaining the predicted result. S4. The graphics post-processing module graphically processes the predicted result, generating deformation maps and deformation extreme value statistics tables, which are then fed back to the Web server. S5. The deformation maps and deformation extreme value statistics tables are downloaded to the local machine through the human-computer interaction module.

[0007] By adopting the above technical solution, the heavy data interpolation, prediction calculation and graphics processing are all transferred to the cloud server for automatic completion using cloud computing architecture. Only basic data needs to be submitted via the network and the final results need to be downloaded. There is no need to install complex calculation software and third-party graphics software locally. This reduces the computing load requirements of the client, reduces the hardware performance dependence on the local professional drawing environment, improves the efficiency of prediction work, and realizes the fully automated collaborative processing.

[0008] Preferably, the geological information of mining subsidence includes the ground elevation of the subsidence area, the thickness of the underground coal seam, the elevation of the coal seam floor, and the coordinates of the underground mining area; S1 specifically includes: extracting the geological information of mining subsidence through the data acquisition module and writing it into a text file; filling in the starting coordinates of the ground calculation area, the length and width of the calculation area, the underground differential size, and the rock movement parameters through a form in a web browser and uploading them to the web server; and uploading the text file containing the geological information of mining subsidence to the web server through a web browser and storing it.

[0009] By adopting the above technical solutions, the acquisition methods and standardized upload channels for basic environmental data are specified in detail, which can ensure that the prerequisite parameters required for geological models are accurately and completely transmitted to the cloud, laying a reliable data foundation for human-computer interaction and subsequent calculations.

[0010] Preferably, step S2 specifically includes: reading data from the Web server, dividing the calculation area into grids according to a preset interval and performing high-order scribing interpolation to form the coordinates and elevation of each ground calculation point; dividing the underground mining area into differential units according to a specified differential size, and performing high-order scribing interpolation on the coal seam floor elevation and coal seam thickness of each differential unit to form the underground mining differential unit information file.

[0011] By adopting the above technical solution and using the interpolation method in the field of Gauss-Kreig geostatistics, the spatial morphology of the surface and underground coal seams can be constructed in a more refined manner, making the three-dimensional data of grid points and differential units closer to the real geological characteristics of the mining area, thereby improving the physical accuracy of the subsidence deformation calculation.

[0012] Preferably, after forming the downhole mining differential unit information file, the method further includes: calculating the dip and dip angle data of each differential point based on the coordinates of the differential point and adjacent differential points; deleting the geological data of differential units outside the specified mining area through a whitening operation; collecting the geological data of each downhole differential point according to the set parameter order, and uploading it to the Web database.

[0013] By adopting the above technical solutions, data interference from ineffective mining areas was effectively filtered out, key dip angle and dip parameters affecting the rock mass movement mechanism were supplemented, the storage structure of the cloud data set was optimized, and the redundant calculations for subsequent subsidence prediction were reduced.

[0014] Preferably, S3 specifically includes: after the main thread of the mining subsidence prediction calculation module detects the calculation task, it divides all ground grid points into multiple calculation tasks according to the number of CPU cores available in the system, and starts multiple calculation threads; each calculation thread uses the probability integral method to predict the mining subsidence deformation value of the ground points in the assigned task; after each calculation thread completes the calculation, it summarizes the calculation results and uploads them to the Web database.

[0015] By adopting the above technical solutions, the advantages of multi-core computing power under the cloud server architecture are fully utilized. Through dynamic task decomposition and multi-threaded parallel computing mechanisms, the estimated computing time for massive grid points is significantly shortened, effectively avoiding resource congestion and crashes caused by single-machine computing.

[0016] Preferably, step S4 specifically includes: reading the expected results from the Web database and writing the subsidence, horizontal movement, tilt deformation, and curvature deformation into different files; statistically analyzing the maximum values ​​of various deformations across the entire area using the main thread and uploading the statistical results to the Web database; starting multiple graphics processing threads, generating deformation contour maps from the expected results using the automatic drawing submodule, and automatically performing deformation statistics on deformation-sensitive areas using the automatic partitioning statistics submodule.

[0017] By adopting the above technical solution, the classification and analysis of massive calculation results were realized, and the cloud-based mapping component was driven in parallel using multi-threading technology. This not only automatically generates intuitive contour maps of various deformations, but also performs targeted deformation statistics for sensitive protected areas such as villages, replacing the traditional manual local mapping process.

[0018] Preferably, after starting multiple graphics processing threads, the method further includes: after all graphics processing threads complete the graphics processing, compressing and packaging the deformation extreme value statistics table and the deformation contour line drawing into a compressed file, and uploading the compressed file to the Web server.

[0019] By adopting the above technical solution, massive amounts of drawing files and analysis reports are centrally packaged and compressed, significantly reducing the bandwidth consumption pressure of network transmission. This allows users to quickly download all final results with one click, greatly improving the convenience of human-computer interaction.

[0020] Secondly, this application provides a cloud computing-based coal mine subsidence prediction system, comprising: a data acquisition module for collecting mining subsidence geological information and submitting the mining subsidence geological information to a web server through a human-computer interaction module; a geological modeling and analysis module for performing grid division and data interpolation on the uploaded mining subsidence geological information to form an underground mining differential unit information file and a ground calculation point coordinate file, and uploading them to a web database; a mining subsidence prediction calculation module for acquiring tasks and using a parallel computing method to calculate the mining subsidence deformation of ground points based on the underground mining differential unit information file and the ground calculation point coordinate file using a probability integral method to obtain the prediction result; a graphics post-processing module for graphically processing the prediction result, generating deformation amount drawings and deformation extreme value statistics tables, and feeding them back to the web server; the human-computer interaction module is also used to download the deformation amount drawings and the deformation extreme value statistics table to the local machine.

[0021] By adopting the above technical solutions, a closed-loop system architecture for software and hardware collaborative operation was constructed. Each functional module has a clear division of labor, realizing a cloud-based automated collaborative ecosystem for data acquisition, modeling, calculation, and mapping, replacing the single-machine subsidence prediction system.

[0022] Preferably, the system further includes a hardware server, which includes the web server, graphics server, and cloud computing server.

[0023] By adopting the above technical solutions, the stable operation of the entire cloud computing system is supported at the physical level. Each dedicated server undertakes network interaction, graphics rendering, and high-concurrency computing tasks, ensuring the system's high availability and rapid response capability when facing large-scale mining data.

[0024] Preferably, the mining subsidence prediction calculation module includes a task partitioning submodule and a mining subsidence prediction calculation submodule; the task partitioning submodule is used to divide the ground points to be calculated into several tasks and call the available CPU cores of the device for allocation; the mining subsidence prediction calculation submodule is used to use multi-core parallel computing technology to predict the deformation of the allocated ground points and feed back the calculation progress to the main thread.

[0025] By adopting the above technical solutions, the internal execution logic of the core cloud computing module has been further refined, making the scheduling and allocation of computing resources more intelligent and reasonable, and achieving seamless connection between real-time progress monitoring and parallel sinking value solving.

[0026] In summary, this application includes at least one of the following beneficial technical effects: 1. By adopting a technical solution that automatically performs heavy tasks such as geological modeling, mining subsidence calculation, and graphic post-processing in the cloud computing background, the computational load requirements of the client are reduced, the hardware performance dependence on the local professional drawing environment is reduced, users do not need to purchase and install foreign professional drawing software systems, eliminate the dependence on third-party software copyrights, and the operation threshold is significantly lowered.

[0027] 2. By adopting a multi-core parallel computing architecture and dynamic task partitioning mechanism, the probability integral calculation of massive ground grid points is distributed to multiple computing threads in the cloud for synchronous processing, which greatly improves the data calculation speed, alleviates the long-term occupation of local computing resources by traditional prediction methods, improves data processing efficiency, and allows operators to carry out other engineering design work in parallel after issuing instructions without occupying local computing resources. Attached Figure Description

[0028] Figure 1 This is a flowchart illustrating a cloud-based method for predicting coal mine subsidence.

[0029] Figure 2 This is a schematic diagram of a cloud-based coal mining subsidence prediction system.

[0030] Figure 3This is a schematic diagram of how a cloud computing server works.

[0031] Figure 4 This is a schematic diagram of how a graphics server works.

[0032] Figure 5 This is a schematic diagram of the workflow of the graphics post-processing module.

[0033] Figure 6 It is a schematic diagram of the drawing and illustration graphic settings file automatically generated by the system.

[0034] Figure 7 This is a schematic diagram illustrating the calculation of mining subsidence deformation values ​​in a cloud computing-based method for predicting coal mine subsidence.

[0035] Figure 8 This is a schematic diagram of the 3D preview interface of the ground subsidence after the graphics processing is completed.

[0036] Figure 9 It is a contour map of the surface subsidence value.

[0037] Figure 10 This is a schematic diagram of one of the parameter setting interfaces of the human-computer interaction module.

[0038] Figure 11 This is a schematic diagram of another parameter setting interface for the human-computer interaction module. Detailed Implementation

[0039] The present application will be further described in detail below with reference to all the accompanying drawings.

[0040] like Figure 1 As shown in the figure, this application discloses a cloud computing-based method for predicting coal mine subsidence.

[0041] S1, the user collects information such as the ground elevation of mining subsidence, underground coal seam thickness, coal seam floor elevation, and underground mining area through the data acquisition module, and submits the basic parameters and mining subsidence geological information to the web server through the human-computer interaction module. This step aims to complete the standardized extraction and cloud upload of geological calculation data, and it is specifically divided into the following sub-steps: S101: This data acquisition module, developed using the Autolisp language within CAD software (such as AutoCAD or GstarCAD), allows users to quickly extract key spatial data, including existing ground elevations in subsidence areas, coal seam thicknesses obtained from underground boreholes, coal seam floor contour elevations, and boundary extreme coordinates of the underground mining area, guided by the software's prompts and via mouse clicks. The extracted discrete and continuous parameters are automatically converted and written to a local standardized text document (txt file). This data acquisition module is used to extract geological spatial data, aiming to reduce errors from manual data entry and improve the accuracy of data extraction.

[0042] S102: Users manually fill in the starting coordinates of the ground calculation area, the expected length and width of the calculation range, the differential size of the divided downhole mining unit, and the key rock movement parameters used to guide the calculation (such as the expected subsidence coefficient, the main influence angle tangent, the horizontal movement coefficient, etc.) through the front-end interactive form of the web browser. After filling in the information, the user uploads it to the web server as the pre-requisite parameters for the task for caching.

[0043] S103: The user uploads the txt text file generated in S101, which contains ground elevation, coordinates of deformation-sensitive areas such as villages and railways, coal seam floor elevation, coal seam thickness, and mining area coordinates, to the cloud web server via a web browser.

[0044] S104: For uploaded files such as ground coal seam thickness, floor elevation, and mining area coordinates, users can use a web browser to access the online editor interface to verify and modify the data before the expected task starts, so as to correct any defects in the source data in a timely manner.

[0045] S105: After the user verifies that all geological information and rock movement parameters are correct on the Web, they click the "Submit and Start Prediction" button. The Web server then completes the structured storage of all data and lists the prediction task as "Pending Calculation" in the background system task queue, thereby triggering the subsequent geological modeling and analysis process.

[0046] S2, after the geological modeling and analysis module detects a new calculation task, it performs gridding, data interpolation, data filtering, and summarization on the uploaded geological information, ultimately generating a file containing the information of the underground mining differential units and a file containing the coordinates of the ground calculation points, which is then uploaded to the web database. The purpose of this step is to transform discrete geological exploration data into a three-dimensional spatial entity model that can be used for calculus calculations. Specifically, this involves: S201: The geological modeling and analysis module deployed in the cloud first reads the task data just stored in the Web server database, and writes the extracted raw parameters into temporary txt files on the backend according to data categories (such as surface data, downhole thickness data, bottom plate data, etc.).

[0047] S202: The background program calls the underlying interface of graphics processing engines such as Surfer to divide the user-defined ground calculation area into planar grids at specified intervals. Subsequently, based on the existing discrete terrain data file, it uses Gaussian Skrig interpolation (a geostatistical interpolation algorithm based on spatial correlation) to precisely interpolate the elevation of each grid point on the ground, thereby forming a ground calculation point coordinate file covering the entire area and containing elevation attributes, and then uploads it to the web server database.

[0048] S203: The program calls the underlying engine to divide the possible underground mining area (this range is determined by expanding the boundary extreme values ​​of the mining area uploaded by the user by a certain safety margin) into tens of thousands or even hundreds of thousands of underground mining differential units according to the specified differential size. For the center point of each differential unit, the Gaussian Skriger interpolation method is also used to calculate the elevation of the coal seam floor and the coal seam thickness to accurately reconstruct the three-dimensional undulation of the underground coal seam.

[0049] S204: In order to accurately input the rock stratum dip angle parameters in subsequent calculations, the system automatically calculates the dip and dip angle data of each differential unit using trigonometric functions based on the three-dimensional coordinate difference between the current differential point and its adjacent differential points, and stores it in a separate file in the background.

[0050] S205: The system performs a whitening operation to forcibly delete invalid data such as coal seam floor elevation, coal seam thickness, dip direction, and dip angle of differential units outside the specified mining polygon area, thereby significantly reducing the amount of subsequent invalid calculations.

[0051] S206: The system will retain the geological data of each effective differential point in the well, strictly follow the order of parameters such as X coordinate, Y coordinate, elevation, coal seam thickness, dip, dip angle, subsidence coefficient, horizontal movement coefficient, tangent of the main influence angle, and mining influence propagation angle, merge and assemble the data set to generate a standard underground mining differential unit information file, and finally upload it to the Web server database for the calculation engine to call.

[0052] S3, the mining subsidence prediction calculation module, acquires tasks and employs a multi-core parallel computing approach. The calculation process is divided into tasks, with each task using a probability integral method to calculate the mining subsidence deformation of the ground point within the mining differential unit, yielding the predicted result. Specific breakdowns are as follows: S301: The main thread of the mining subsidence prediction calculation module is in a real-time polling state. When it detects that the ground grid point coordinate data file related to the calculation task and the geological data set file of each differential point in the well have been uploaded to the Web database and are ready, it immediately reads the massive amount of data into the memory of the cloud computing server.

[0053] S302: The main thread of the expected computing module is based on the number of available physical or logical CPU cores of the current cloud computing server detected by the background program. (Typically, 2-3 cores are reserved to ensure the stable operation of the underlying operating system, and the remaining kernels are all used for computation.) The tens of thousands of ground grid point tasks are evenly divided according to the principle of load balancing. Sub-tasks are collected and launched simultaneously. An asynchronous computing thread.

[0054] S303: Each awakened computing thread independently calculates the deformation value of the assigned ground grid points using the classic probability integral method, based on its assigned task.

[0055] The calculation formula is as follows: Where j is the index of the current calculation point in the calculation task; Represented as the coordinates, m, of the j-th ground calculation point; The elevation of the j-th ground calculation point is represented in meters. This is represented by the surface subsidence value at the j-th ground calculation point; It is represented as the surface inclination value in the X direction of the j-th ground calculation point, in mm / m; The value is represented as the surface inclination in the Y direction of the j-th ground calculation point, in mm / m; It is represented by the surface curvature value in the X direction of the j-th ground calculation point. ; It is represented by the surface curvature value in the Y direction of the j-th ground calculation point. ; It is represented as the horizontal deformation value of the ground surface in the X direction at the j-th ground calculation point, in mm / m; It is represented as the horizontal deformation value of the ground surface in the Y direction at the j-th ground calculation point, in mm / m; This represents the total number of differential units within all downhole mining areas; This is the index of the current unit of differentiation among all units of differentiation; The coordinates, m, of the i-th downhole differential mining unit are represented. The elevation is represented by the i-th downhole differential mining unit; Characterized as the coal seam thickness of the i-th underground differential mining unit, in meters; The coal seam dip angle, denoted as °, is represented by the i-th underground differential mining unit. Characterized as the subsidence coefficient; The maximum surface subsidence caused by the i-th downhole differential mining unit is represented in m; The radius of influence of the i-th downhole differential mining unit on the j-th surface calculation point is represented in m. Characterized as the main influencing angle, in °; Characterized as a horizontal shift coefficient; Characterized as the angle of propagation of the mining impact, in °.

[0056] In practice, the program calculates the impact of each underground differential mining unit on its corresponding surface calculation point and performs integral superposition. The calculations include extracting seven core subsidence indicators: subsidence, horizontal displacement deformation along the X and Y directions, tilt deformation along the X and Y directions, and curvature deformation along the X and Y directions. This parallel mechanism avoids single-machine memory overflow issues and shortens the computation cycle for massive grids.

[0057] S304: After all distributed computing threads have finished running and released resources, the system main thread uniformly collects and organizes the deformation calculation results of each point in the memory of all child threads, summarizes and generates a single comprehensive deformation prediction result file for the entire region, and securely uploads it to the web server database. The business process is as follows: Figure 3 As shown.

[0058] S4, the post-processing module, categorizes and graphically processes the predicted results, automatically generating contour maps of various deformation amounts and statistical tables of deformation extreme values, completing closed-loop feedback. The business process is as follows: Figure 4 As shown, the specific breakdown is as follows: S401: The graphics post-processing module reads the expected settlement deformation results from the Web server database. After data decomposition, it separates the seven types of data, namely settlement, horizontal movement (X / Y direction), tilt deformation (X / Y direction), and curvature deformation (X / Y direction), and writes them into seven independent text txt files.

[0059] S402: The main thread of graphics post-processing first traverses all deformation files in the entire region, selects the maximum value (extreme value) of these seven types of deformation in the entire region for statistical archiving, and uploads the statistical results of the deformation extreme value to the Web database first.

[0060] S403: The graphics post-processing module then launches seven parallel graphics processing sub-threads, each loading a graphics post-processing component based on a secondary development and encapsulation of the Surfer engine. This component is internally divided into an automatic drawing sub-module and an automatic partitioning and statistics sub-module, such as... Figure 5 As shown.

[0061] The workflow of the automatic drawing submodule is as follows: After the system reads in the single-type deformation data, it generates a GRD mesh file, and then generates a contour map. The system adaptively determines the contour line spacing based on the maximum deformation extreme value read from S402 (for example, when the maximum subsidence deformation is less than 1000mm, contour lines are drawn with a spacing of 200mm, and labels of 10mm, 200mm, 400mm, etc. are automatically added; when the maximum value exceeds 5000mm, contour lines are drawn with a spacing of 500mm and labeled accordingly). Then, the font size is determined according to the "Coal Mine Mapping Standard" and the user-defined scale, generating a complete LVL configuration parameter file (e.g., ...). Figure 6 (As shown) and loads into the rendering engine, finally automatically exporting a DWG format deformed contour drawing file.

[0062] The workflow of the automatic partitioning statistics submodule is as follows: The system reads the coordinate data of deformation-sensitive protection zones such as villages and railways submitted by users in the S1 stage, calls the "whitening" function in the rendering engine to cut off all deformation GRD data outside the boundary of the sensitive zone, retains only the deformation data within the partition, and reads the local deformation extreme values ​​within each protection zone and uploads them to the Web database.

[0063] S404: The system backbone process compiles the final deformation statistics results for the entire area and each sensitive zone and writes them into an Excel spreadsheet file. Then, this report file is combined with the previously generated 7 deformation contour DWG drawing files (such as...). Figure 9 (As shown) They are compressed and packaged together at a high ratio to form a small ZIP compressed file, which is then sent to the web server.

[0064] S5, users interact through the human-computer interaction module (such as...) Figure 10 to Figure 11 (As shown) In the web browser interface, the task status is displayed as "Computing" (e.g., Figure 7 The status bar (as shown) has been changed to "Completed," and a 3D rendering preview of the sinking effect can be seen in the front-end status bar (e.g.). Figure 8 (As shown). Users only need to click the "Download Expected Results" button to directly download the compressed package containing the calculation results Excel spreadsheet and DWG drawings to their local computer, and then conduct engineering drawing review and scheme demonstration in their local CAD software.

[0065] This application also discloses a cloud computing-based coal mine subsidence prediction system.

[0066] like Figure 2 As shown, the system's hardware architecture physically deploys a hardware server cluster, which includes a web server for network interaction and database access, a graphics server for handling complex primitive rendering and graphics format conversion, and a cloud computing server for providing computing power support. At the software architecture level, this system is highly modular, specifically including the following core modules: The data acquisition module, deployed on the user's local computer terminal, is seamlessly embedded into existing mainstream domestic or international CAD drawing software using secondary development languages ​​such as Autolisp. Its main function is to semi-automatically convert the geological spatial information of the subsidence area on the drawings into a standard TXT text document that the system can recognize by intercepting mouse click events and capturing graphic element attributes, thus eliminating errors from manual data transcription.

[0067] The human-computer interaction module, developed using front-end technologies such as JavaScript and built upon a web server, provides engineering technicians with a lightweight B / S architecture operating platform. This module integrates a parameter configuration panel, a basic data file upload interface, an online source data errata editor, a real-time monitoring panel for expected status, and a result / material download channel.

[0068] The geological modeling and analysis module is deployed on a cloud computing server backend. This module receives raw survey data and uses Gaussian scrimmage interpolation and spatial envelope whitening algorithms to reconstruct an array of differential units of the underground coal seam and an array of surface grid nodes that can be directly called by calculus algorithms.

[0069] The graphics post-processing module runs on a graphics server with a robust graphics processing instruction set. This module incorporates dual engines for automatic plotting and automatic zoning statistics. Through multi-threaded concurrent calls to underlying graphics library components (such as the SurferCOM interface), it maps and adaptively renders digital matrices into compliant engineering contour drawings (DWG), outputting local extreme value Excel reports for sensitive areas such as villages, and then performs packaging and compression.

[0070] The mining subsidence prediction calculation module, also deployed on a cloud computing server, includes a task partitioning submodule and a mining subsidence prediction calculation submodule. The task partitioning submodule probes the operating system for currently available CPU physical core resources and, after reserving a safety threshold, splits several ground calculation points into several task packages. The mining subsidence prediction calculation submodule schedules a multi-core parallel mechanism, using an improved probability integral method to drive each sub-thread to complete the spatial matrix transformation solution, and reports the percentage calculation progress to the main thread in real time.

[0071] Taking a mining area in Zhaolou Coal Mine, Shandong Province as an example, after the calculation is completed, users can see a 3D preview of the subsidence in the calculation status bar. Clicking "Download Expected Results" will download the compressed file containing the expected results. The subsidence contour map downloaded by the system is in an easily editable dwg format.

[0072] The working principle of this embodiment is as follows: The data acquisition module on the CAD end extracts coal mine geological exploration data and outputs it as standard text, which is then transmitted to the cloud via the network through the Web human-computer interaction module. The geological modeling and analysis module constructs an underground three-dimensional mesh model using Gaussian scribal interpolation and removes redundant interference using whitening technology. The mining subsidence prediction calculation module utilizes dynamic task allocation and multi-threaded parallel technology to break through the traditional single-machine computing power bottleneck, completing the probability integral matrix solution for massive grid points in a short time. The graphics post-processing module drives the underlying rendering components in parallel, not only adaptively adjusting the contour line spacing to automatically generate DWG drawings, but also statistically analyzing and packaging the subsidence extreme values ​​of sensitive areas such as villages. The entire process leverages the centralized advantages of cloud computing resources to achieve cloud-hosted management of heavy computation and heavy rendering tasks, freeing users from high local software licensing costs and computing power limitations, and realizing efficient and automated operation of subsidence prediction services.

[0073] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A cloud computing-based method for predicting coal mine subsidence, characterized in that, Includes the following steps: S1, collect geological information on mining subsidence through the data acquisition module, and submit the geological information on mining subsidence to the Web server through the human-computer interaction module; S2, the geological modeling and analysis module performs grid division and data interpolation on the uploaded mining subsidence geological information to form an underground mining differential unit information file and a ground calculation point coordinate file, and uploads them to the Web database; S3, the mining subsidence prediction calculation module acquires the task and adopts a parallel computing method. It uses the probability integral method to calculate the mining subsidence deformation of the ground point based on the information file of the downhole mining differential unit and the coordinate file of the ground calculation point, and obtains the prediction result. S4, the graphics post-processing module performs graphical processing on the expected results, generates deformation drawings and deformation extreme value statistics tables, and feeds them back to the Web server; S5, the deformation drawing and the deformation extreme value statistics table are downloaded to the local machine through the human-computer interaction module.

2. The method for predicting coal mine subsidence based on cloud computing according to claim 1, characterized in that, The geological information regarding the mining subsidence includes the ground elevation of the subsidence area, the thickness of the underground coal seam, the elevation of the coal seam floor, and the coordinates of the underground mining area; S1 specifically includes: The geological information of the mining subsidence is extracted by the data acquisition module and written into a text file; The starting coordinates of the ground calculation area, the length and width of the calculation range, the downhole differential dimensions, and the rock movement parameters are filled in through a form in a web browser and then uploaded to the web server. The text file containing the geological information of the mining subsidence is uploaded to the web server via a web browser and stored.

3. The method for predicting coal mine subsidence based on cloud computing according to claim 1, characterized in that, S2 specifically includes: The data in the Web server is read, the calculation area is divided into grids according to a preset interval and high-order scribing interpolation is performed to form the coordinates and elevation of each ground calculation point; The underground mining area is divided into differential units according to a specified differential size, and the coal seam floor elevation and coal seam thickness of each differential unit are calculated using high-Skelige interpolation to form the underground mining differential unit information file.

4. The method for predicting coal mine subsidence based on cloud computing according to claim 3, characterized in that, After forming the downhole mining differential unit information file, the following is also included: Calculate the dip and tilt angle data for each differential point based on the coordinates of the differential point and its adjacent differential points; Deleting differential unit geological data outside the specified mining area using the whitening operation; The geological data of each downhole differential point are collected according to the set parameter order and uploaded to the Web database.

5. The method for predicting coal mine subsidence based on cloud computing according to claim 1, characterized in that, S3 specifically includes: After the main thread of the mining subsidence prediction calculation module detects the calculation task, it divides all ground grid points into multiple calculation tasks according to the number of CPU cores available in the system, and starts multiple calculation threads. Each of the computing threads uses the probability integral method to predict the mining subsidence deformation value of the ground points in the assigned task; After each computing thread completes its calculation, the results are aggregated and uploaded to the Web database.

6. The method for predicting coal mine subsidence based on cloud computing according to claim 1, characterized in that, S4 specifically includes: The predicted results are read from the Web database, and the subsidence, horizontal movement, tilt deformation, and curvature deformation are written into different files respectively. The maximum values ​​of various deformation quantities in the entire region are statistically analyzed using the main thread, and the statistical results are uploaded to the Web database. Multiple graphics processing threads are started. The automatic drawing submodule generates deformation contour maps from the expected results, and the automatic partitioning and statistics submodule performs automatic deformation statistics on deformation-sensitive areas.

7. The method for predicting coal mine subsidence based on cloud computing according to claim 6, characterized in that, After starting multiple graphics processing threads, the process also includes: After all the graphics processing threads have completed the graphics processing, they compress and package the deformation extreme value statistics table and the deformation contour line drawing into a compressed file, and then upload the compressed file to the web server.

8. A cloud computing-based coal mine subsidence prediction system, characterized in that, The system is used to implement a cloud computing-based method for predicting coal mine subsidence as described in any one of claims 1 to 7, the system comprising: The data acquisition module is used to collect geological information on mining subsidence and submit the geological information on mining subsidence to the Web server through the human-computer interaction module; The geological modeling and analysis module is used to perform grid division and data interpolation on the uploaded mining subsidence geological information, forming underground mining differential unit information files and ground calculation point coordinate files, and uploading them to the Web database; The mining subsidence prediction calculation module is used to acquire tasks and use a parallel computing method to calculate the mining subsidence deformation of ground points based on the information file of the downhole mining differential unit and the coordinate file of the ground calculation point using the probability integral method, so as to obtain the prediction results. The graphics post-processing module is used to graphically process the expected results, generate deformation drawings and deformation extreme value statistics tables, and feed them back to the Web server; The human-computer interaction module is also used to download the deformation drawing and the deformation extreme value statistics table to the local machine.

9. A cloud computing-based coal mine subsidence prediction system according to claim 8, characterized in that, The system also includes a hardware server, which includes the web server, graphics server, and cloud computing server.

10. A cloud computing-based coal mine subsidence prediction system according to claim 8, characterized in that, The mining subsidence prediction calculation module includes a task division submodule and a mining subsidence prediction calculation submodule; The task partitioning submodule is used to divide the ground points that need to be calculated into several tasks and call the available CPU cores of the device for allocation; The mining subsidence prediction calculation submodule is used to predict the deformation of the allocated ground points using multi-core parallel computing technology, and to feed back the calculation progress to the main thread.

Images