Coal mining prediction method and coal mining prediction device

Abstract

The invention relates to the technical field of coal mining, in particular to a coal mining prediction method and a coal mining prediction device. The method comprises the steps of obtaining geological basic data, and obtaining a regional geological database based on the geological basic data; based on the regional geological database, carrying out three-dimensional geological modeling by adopting a multi-field coupling modeling method; performing dynamic iteration correction on the three-dimensional geologic model based on the data consistency constraint of the virtual sensor node; production working condition parameters of the working face in the mining process are obtained in real time, the production working condition parameters are converted into constraint conditions of the three-dimensional geologic model, and dynamic modeling in the mining process of the working face is carried out; based on the production working condition parameters and the three-dimensional geologic model, intelligent analysis and multi-field coupling simulation are carried out, and a mining prediction result is obtained; updating the mining decision in real time based on the mining prediction result; the device is applied to the method, and accurate prediction, dynamic optimization and safety guarantee of the mining process are achieved.

Description

Technical Field

[0001] This invention relates to the field of coal mining technology, and more specifically, to a coal mining prediction method and a coal mining prediction device. Background Technology

[0002] With the increasing depth and scale of coal mining, the geological conditions in mines are becoming increasingly complex, constantly facing threats from geological disasters such as roof fractures, expansion of water-conducting fracture zones, abnormal gas outbursts, and rock bursts, seriously affecting safe production. These complex geological conditions place higher demands on mine safety and intelligent management. For building geological support systems, existing technologies mainly rely on single geological exploration methods, such as ground-penetrating radar, 3D seismic surveys, transient electromagnetic drilling, or hydrological boreholes. These methods can reveal the stratigraphic distribution, fault structures, and aquifer conditions in the mining area to a certain extent, but they have the following shortcomings: First, geological modeling is static. Most existing 3D geological models are based on one-time exploration data, lacking a linkage mechanism with subsequent monitoring data after model establishment. As mine excavation and mining progress, roof stress, seepage conditions, and gas migration states constantly change, and traditional models cannot dynamically correct based on real-time monitoring data, leading to significant deviations between predicted results and actual conditions. Second, production disturbances at the working face are not reflected. In existing technologies, the mining parameters of the working face (mining height, advance speed, support method, etc.) are independent of the geological model, and the model fails to reflect the impact of production disturbances on the strata and overlying structure. For example, roof decompression caused by tunneling, fracture zone expansion caused by mining, and aquifer activation cannot be dynamically represented in the model, making it difficult for the prediction results to reflect the true impact of production activities on the geological environment. Third, there is a lack of real-time prediction and feedback optimization. Existing numerical simulation and prediction methods are mostly offline calculations, which cannot achieve real-time disturbance prediction of the coal mine production process. Furthermore, the prediction results fail to establish a linkage with the mine scheduling system, lacking automated risk warning and process optimization mechanisms, making it difficult to promptly translate prediction information into scheduling decisions and reducing the system's application value in actual production.

[0003] Therefore, how to realize an integrated system from multi-source geological data acquisition and three-dimensional dynamic modeling to intelligent prediction and feedback optimization has become a technical problem that urgently needs to be solved in the field of intelligent mining. Summary of the Invention

[0004] The purpose of this invention is to provide a coal mining prediction method and a coal mining prediction device to solve at least one technical problem in the background art.

[0005] In a first aspect, the present invention provides a method for predicting coal mining operations, the method comprising: Obtain basic geological data, and based on the basic geological data, obtain a regional geological database; the basic geological data includes: stratum thickness, lithological distribution, fault structure, folds, collapse columns, aquifer distribution, hydrogeological parameters, coal seam occurrence state and gas occurrence conditions; Based on a regional geological database, a multi-field coupling modeling method is used for three-dimensional geological modeling; the three-dimensional geological modeling includes: geometric modeling, hydrological modeling, gas geological modeling, geostress property modeling, reserve modeling, and hidden disaster-causing modeling. A virtual sensor node network is constructed in the three-dimensional geological model, with each node mapping one-to-one with the actual monitoring equipment in the downhole space. The virtual sensor nodes receive real-time monitoring data from the actual monitoring equipment, and the three-dimensional geological model is dynamically iteratively corrected based on the data consistency constraints of the virtual sensor nodes. The actual monitoring equipment includes stress monitoring equipment, seepage monitoring equipment, gas monitoring equipment, and energy monitoring equipment. The monitoring data includes stress data, seepage data, gas data, and energy data. Real-time acquisition of working face production condition parameters during mining is performed, and these parameters are converted into constraints for the three-dimensional geological model to dynamically model the working face during the mining process. The working face production condition parameters include: actual mining height, advance displacement, advance speed, support structure layout parameters, original rock stress, ventilation parameters, and mining depth. The constraints include: geometric constraints, time loading conditions, mechanical boundary conditions, gas field boundary conditions, and seepage field boundary conditions. Based on the production condition parameters and the three-dimensional geological model, intelligent analysis and multi-field coupled simulation are performed to obtain mining prediction results; the mining prediction results include: overburden migration, fracture zone development, water hazard risk and gas risk; Based on the mining prediction results, mining decisions are updated in real time; the mining decisions include at least: optimizing mining parameters or early warning processing.

[0006] Optionally, the process of obtaining a regional geological database based on the geological baseline data includes: Acquire geophysical data from the surface and borehole data from downhole; The geophysical data, the borehole data, and the real-time monitoring data are normalized to obtain normalized data; the normalization process includes: unified coordinate reference constraints, physical dimension normalization, and attribute consistency verification; Based on the normalized data, a standardized regional geological database is constructed for three-dimensional geological models and dynamic iterative corrections.

[0007] Optionally, the dynamic iterative correction includes: inverting and updating the mechanical field, seepage field, and gas field parameters in the three-dimensional geological model based on real-time monitoring data, and using the deviation between historical monitoring data and current real-time monitoring data as a constraint condition to control the convergence of model parameters.

[0008] Optionally, obtaining the mining prediction result based on the production condition parameters and the three-dimensional geological model includes: Using geological big data intelligent analysis technology, the regional geological structure, coal seam occurrence, and hydrological and gas conditions are intelligently interpreted; Based on the production condition parameters and the real-time updated three-dimensional geological model, a numerical simulation platform is invoked to perform multi-field coupling calculations; the numerical simulation platform includes: FLAC3D simulation platform and COMSOL Multiphysics simulation platform. By combining artificial intelligence inversion optimization algorithms, mining prediction results for overburden migration, fracture zone development, water hazard risk, gas risk, and energy risk are obtained. The overburden migration includes the overburden subsidence and the critical layer fracture height. The fracture zone development includes the development height of the water-conducting fracture zone and its risk coefficient of connection with the aquifer. The water hazard risk includes the probability of local water inrush and the expected inflow. The gas risk includes areas of abnormal gas enrichment and the maximum gas outflow. The energy risk includes the rockburst risk level of energy accumulation areas.

[0009] Optionally, the artificial intelligence inversion optimization algorithm includes: Convolutional neural networks are used to automatically identify faults, folds, and collapse columns in seismic data. Recurrent neural networks or gated recurrent units are used to predict the temporal evolution of seepage and gas fields. Random forests and support vector machines are used to classify and determine the existence and development trend of activated fluid channels.

[0010] Optionally, the dynamic modeling during the working face recovery process includes: By combining artificial intelligence feature extraction algorithms to analyze the production condition parameters, key factors of overburden migration and fracture zone development are identified.

[0011] Optionally, the optimized mining parameters include: adjusting the key factors; The early warning process includes: increasing the number of advanced water exploration boreholes; and automatically triggering an early warning if the predicted risk exceeds a preset threshold in the mining prediction results.

[0012] Secondly, the present invention provides a coal mine mining prediction device applied to the above-mentioned method, the device comprising: The first acquisition module is used to acquire basic geological data and, based on the basic geological data, obtain a regional geological database. The basic geological data includes: stratum thickness, lithological distribution, fault structure, folds, collapse columns, aquifer distribution, hydrogeological parameters, coal seam occurrence state, and gas occurrence conditions. The first modeling module is used to perform three-dimensional geological modeling based on a regional geological database and using a multi-field coupling modeling method. The three-dimensional geological modeling includes: geometric modeling, hydrological modeling, gas geological modeling, geostress property modeling, reserve modeling, and hidden disaster-causing modeling. A dynamic correction module is used to construct a virtual sensor node network in the three-dimensional geological model that maps one-to-one with the actual monitoring equipment in the downhole space. The virtual sensor nodes receive real-time monitoring data from the actual monitoring equipment, and the three-dimensional geological model is dynamically iteratively corrected based on the data consistency constraints of the virtual sensor nodes. The actual monitoring equipment includes stress monitoring equipment, seepage monitoring equipment, gas monitoring equipment, and energy monitoring equipment. The monitoring data includes stress data, seepage data, gas data, and energy data. The second modeling module is used to acquire the production condition parameters of the working face in real time during the mining process, and convert the production condition parameters into the constraints of the three-dimensional geological model to perform dynamic modeling of the working face during the mining process. The production condition parameters include: actual mining height, advance displacement, advance speed, support structure parameters, original rock stress, ventilation condition parameters, and mining depth. The constraints include: geometric constraints, time loading conditions, mechanical boundary conditions, gas field boundary conditions, and seepage field boundary conditions. The analysis and simulation module is used to perform intelligent analysis and multi-field coupled simulation based on the production condition parameters and the three-dimensional geological model to obtain mining prediction results. The mining prediction results include: overburden migration, fracture zone development, water hazard risk, and gas risk. An optimization processing module is used to update mining decisions in real time based on the mining prediction results; the mining decisions include at least: optimizing mining parameters or early warning processing.

[0013] Optionally, the dynamic correction module includes: Stress sensors are used to acquire triaxial stress changes in the roof and coal seam to construct a mechanical field; Seepage sensors are used to acquire groundwater pressure, water level, and permeability parameters to construct a seepage field; Gas sensors are used to acquire gas concentration, outflow rate, and desorption rate to construct a gas field; Vibration monitors and acoustic probes are used to acquire energy release rates and fracture waveforms to construct an energy field.

[0014] Optionally, the device further includes: The display module includes a transparent mine geological cloud platform, which has a three-dimensional visualization interface for intuitively displaying faults, folds, collapse columns and aquifers, and supports remote access and dynamic interaction.

[0015] The coal mining prediction method and apparatus provided by this invention have the following beneficial effects: By using a regional geological database and employing a multi-field coupling modeling method, three-dimensional geological modeling is performed. Combined with monitoring data from preset locations, the three-dimensional geological model is iteratively updated. Furthermore, by combining mining data, intelligent analysis and multi-field coupling simulation are conducted to accurately obtain mining prediction results. This allows for real-time updates to mining decisions based on the mining prediction results, such as optimizing mining parameters or handling early warnings. This achieves accurate prediction, dynamic optimization, and safety assurance of the mining process. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0017] Figure 1 This is a schematic flowchart of the coal mining prediction method provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating a specific example of the coal mining prediction method provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of the coal mining prediction device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the overall architecture of the coal mining prediction method and coal mining prediction device provided in the embodiments of the present invention. Detailed Implementation

[0018] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] The following specific examples, in conjunction with the accompanying drawings, illustrate the concepts. Figures 1-4 The present invention will now be described in further detail.

[0020] Example 1 This invention provides a method for predicting coal mining operations, which includes the following steps: S101. Obtain basic geological data. Based on the basic geological data, obtain the regional geological database. The basic geological data includes: stratum thickness, lithological distribution, fault structure, folds, collapse columns, aquifer distribution, hydrogeological parameters, coal seam occurrence state, and gas occurrence conditions.

[0021] S102, based on a regional geological database, employs a multi-field coupling modeling method for three-dimensional geological modeling; the three-dimensional geological modeling includes: geometric modeling, hydrological modeling, gas geological modeling, geostress property modeling, reserve modeling, and hidden disaster-causing modeling.

[0022] S103, a virtual sensor node network is constructed in the three-dimensional geological model, with each node mapping one-to-one with the actual monitoring equipment in the downhole space. The virtual sensor nodes receive real-time monitoring data from the actual monitoring equipment, and the three-dimensional geological model is dynamically iteratively corrected based on the data consistency constraints of the virtual sensor nodes. The actual monitoring equipment includes stress monitoring equipment, seepage monitoring equipment, gas monitoring equipment, and energy monitoring equipment. The monitoring data includes stress data, seepage data, gas data, and energy data. The stress monitoring equipment is used to monitor stress data, the seepage monitoring equipment is used to monitor seepage data, the gas monitoring equipment is used to monitor gas data, and the energy monitoring equipment (e.g., vibration monitors and acoustic emission probes) is used to monitor energy data.

[0023] It should be noted that data consistency constraints refer to the introduction of unified constraint rules for geological and monitoring data from different sources, times, and monitoring methods during the construction and dynamic correction of 3D geological models. This ensures the consistency of various data in terms of spatial location, physical meaning, and trend of change, preventing physical contradictions or numerical divergence from occurring during iterative model correction. Specifically, the data consistency constraints include at least the following: data acquired by different monitoring devices at the same spatial location or within the same geological unit should meet consistency requirements in terms of trend of change; the values ​​of the same physical quantity under different monitoring times or methods should maintain continuity and rationality; and the model correction results should simultaneously meet the common constraints of multi-source monitoring data, rather than fitting only a single data source. By introducing data consistency constraints, overfitting of the model to only local monitoring data can be avoided, thereby improving the overall reliability and stability of the 3D geological model under multi-source monitoring conditions.

[0024] S104: Real-time acquisition of production condition parameters of the working face during mining, conversion of production condition parameters into constraints of a three-dimensional geological model, and dynamic modeling of the working face during the mining process; the production condition parameters include: actual mining height, advance displacement, advance speed, support structure parameters, original rock stress, ventilation condition parameters, and mining depth; the constraints include: geometric constraints, time loading conditions, mechanical boundary conditions, gas field boundary conditions, and seepage field boundary conditions.

[0025] It should be noted that converting production condition parameters into constraints for a three-dimensional geological model means transforming real-time production state parameters acquired during coal mining into control conditions that can be used to limit or correct the calculation boundaries, loading conditions, or evolution processes of the three-dimensional geological model. Specific conversion methods include, but are not limited to: converting actual mining height, advance position, and advance speed into the geometric boundaries of the goaf and their temporal evolution conditions in the three-dimensional geological model; converting the layout of the support structure into mechanical constraints or equivalent support parameters in the model; and converting production condition parameters such as original rock stress, ventilation conditions, and mining depth into initial stress fields, boundary loading conditions, or time loading conditions in the model.

[0026] S105, based on production condition parameters and a three-dimensional geological model, performs intelligent analysis and multi-field coupled simulation to obtain mining prediction results; the mining prediction results include: overburden migration, fracture zone development, water hazard risk and gas risk.

[0027] S106, Based on the mining prediction results, update the mining decision in real time; the mining decision includes at least: optimizing mining parameters or handling early warnings.

[0028] In this mining method, a three-dimensional geological model is created using a regional geological database and a multi-field coupling modeling method. The three-dimensional geological model is iteratively updated by combining monitoring data from preset locations. Combined with mining data, intelligent analysis and multi-field coupling simulation are performed to accurately obtain mining prediction results. This allows for real-time updates to mining decisions based on the mining prediction results, such as optimizing mining parameters or handling early warnings. This achieves accurate prediction, dynamic optimization, and safety assurance of the mining process.

[0029] In step S101, a regional geological database is obtained based on the geological baseline data, including the following steps: S1011, acquire geophysical data from the surface and borehole data from downhole; S1012, The geophysical data, the borehole data and the real-time monitoring data are normalized to obtain normalized data; the normalization process includes: unified coordinate reference constraint, physical dimension normalization and attribute consistency verification; It should be noted that unified coordinate reference constraint refers to using the same spatial coordinate reference or coordinate transformation rule for multi-source geological data from surface geophysical exploration, downhole drilling, and real-time monitoring, mapping all types of data to a unified three-dimensional spatial coordinate system. Attribute consistency verification refers to checking and correcting the consistency of data representing the same physical or geological attribute in multi-source geological data after coordinate unification. This attribute consistency verification includes at least: verifying the rationality of attributes such as lithology, mechanical parameters, and hydrogeological parameters across different data sources; verifying the consistency of change trends in monitoring data at different time scales; and screening, correcting, or downweighting data with obvious anomalies or conflicts.

[0030] S1013, Based on the normalized data, construct a standardized regional geological database for three-dimensional geological models and dynamic iterative corrections.

[0031] Specifically, through joint surface and underground exploration methods, the coal and rock mass is meticulously explored to obtain regional geological multi-field basic data information, including coal seam occurrence state, faults, folds, collapse columns, aquifer distribution, gas occurrence conditions, hydrogeological parameters, stratum thickness, lithology distribution, etc. Combined with mine production and operation statistics, multi-source fusion and standardization processing are carried out. The standardization processing methods can include noise reduction, interpolation and fusion processing, thereby forming a standardized regional geological database.

[0032] Mine data includes surface and underground geological exploration data, geophysical data, and monitoring data, including data obtained from ground-penetrating radar, 3D seismic data, transient electromagnetic data, and hydrological borehole data. Based on geospatial interpolation methods and attribute modeling technology, the system integrates geospatial data from different data sources, enabling the storage, fusion, and management of multiple data sources.

[0033] In step S102, a multi-field coupling modeling method is used for three-dimensional geological modeling, including: The transparent mine platform is built based on multi-source fusion data to establish a three-dimensional geological model of the coal mine. The model integrates strata, fault structures, aquifers and gas-rich areas, and the model visualization and intelligent management interface are completed through the transparent mine geological cloud platform.

[0034] The 3D geological modeling function for mines involves geometric modeling, hydrological modeling, gas geology modeling, geostress attribute modeling, reserve modeling (coal resources, gas-oil type gas occurrence), and hidden disaster modeling. Through this function, a macroscopic 3D geological spatial structure of the coal mine (strata, coal seams, faults, boreholes) and an attribute 3D structure (hydrology, gas-oil type gas, geostress, reserves, hidden disasters) are constructed, achieving a realistic and detailed description of the mine's geological conditions.

[0035] In step S103, the dynamic iterative correction includes: updating the mechanical field, seepage field, and gas field parameters in the 3D geological model based on real-time monitoring data, and using the deviation between historical monitoring data and current real-time monitoring data as a constraint to control the convergence of model parameters. Specifically, during the dynamic iterative correction of the 3D geological model, the difference between historical monitoring data and current real-time monitoring data is compared and used as a constraint to control the magnitude and direction of model parameter adjustment. When the deviation between the model calculation results and historical and current real-time monitoring data is large, a larger correction of model parameters is allowed. As the model calculation results gradually approach the historical and real-time monitoring data, the adjustment magnitude of model parameters is gradually reduced. When the deviation meets the preset convergence condition, further adjustment of model parameters is stopped or restricted. Through the above methods, the gradual convergence of model parameters is achieved, avoiding model instability or calculation divergence caused by repeated large-scale parameter adjustments, thereby improving the stability and reliability of the dynamic correction of the 3D geological model.

[0036] Specifically, during the 3D geological modeling and correction process of the mine, virtual sensor nodes (preset positions) corresponding to the mine are arranged in the 3D geological model. These sensor nodes correspond to underground stress monitoring, seepage monitoring, gas monitoring, and vibration monitoring equipment. Real-time monitoring data is received, and a dynamic correction algorithm is used to iteratively update the model to reflect the dynamic changes in the mechanical field, seepage field, gas field, and energy field. The multi-field information collected by the sensors includes: stress sensors acquiring triaxial stress changes in the roof and coal body to construct the mechanical field; seepage sensors acquiring groundwater pressure, water level, and permeability parameters to construct the seepage field; gas sensors acquiring gas concentration, emission rate, and desorption rate to construct the gas field; and vibration monitors and acoustic emission probes acquiring energy release rates and rupture waveforms to construct the energy field.

[0037] In step S104, the production condition parameters of the working face correspond to the operating status parameters of the coal mining equipment and the tunneling equipment. These production condition parameters are then transformed into constraints for the three-dimensional geological model. This mainly involves converting the actual mining height, advance displacement, advance speed, support structure parameters, and mining depth into geometric constraints and time-based loading conditions in the numerical model. Mechanical boundary conditions are set in conjunction with the original rock stress, and boundary conditions for the gas field and seepage field are set in conjunction with the ventilation condition parameters. This achieves dynamic modeling of the working face during the mining process. The original rock stress is the original rock stress corresponding to the coal burial depth.

[0038] Dynamic modeling of the working face based on mining parameters includes: analyzing production condition parameters using artificial intelligence feature extraction algorithms to identify key factors affecting overburden migration and fracture zone development. Specifically, inputting coal mine mining parameters, including mining height, advance speed, support method, mining depth, and ventilation conditions; dynamically modeling the working face based on these production condition parameters; and using artificial intelligence feature extraction algorithms to identify key factors influencing overburden migration and fracture zone development.

[0039] Specifically, dynamic modeling of mine working faces comprises two layers: the "static" changes before mining and the "dynamic" changes after mining. The "static" changes before mining are expressed by the aforementioned "3D Geological Modeling of Mines" function, while the "dynamic" changes after mining are realized by the "Dynamic Modeling of Working Faces" function. The dynamic modeling function operates on coal mining faces and tunneling faces, mainly involving production planning and geological disaster simulation. The production planning function is reflected in the adaptive coal cutting of the working face, while the geological disaster simulation function reflects the analysis and prediction of water hazards, rock bursts, and gas (oil-type gas) disasters based on numerical simulation technology.

[0040] In step S105, based on mining parameters and a three-dimensional geological model, mining prediction results are obtained, including: S1051 utilizes intelligent analysis technology based on geological big data to provide intelligent interpretation of regional geological structures, coal seam occurrence, and hydrological and gas conditions. S1052, based on production condition parameters and a real-time updated 3D geological model, calls a numerical simulation platform to perform multi-field coupling calculations; the numerical simulation platform includes: FLAC3D simulation platform and COMSOL Multiphysics simulation platform; S1053, combined with artificial intelligence inversion optimization algorithm, yields mining prediction results for overburden migration, fracture zone development, water hazard risk, gas risk, and energy risk. Overburden migration includes overburden subsidence and key layer fracture height; fracture zone development includes the development height of water-conducting fracture zones and their risk coefficient of connection with aquifers; water hazard risk includes the probability of local water inrush and the expected inflow; gas risk includes areas of abnormal gas enrichment and the maximum gas outflow; and energy risk includes the rockburst risk level of energy accumulation areas.

[0041] Specifically, the artificial intelligence inversion optimization algorithms include: convolutional neural networks, recurrent neural networks / gated recurrent units, random forests, and support vector machines; among them, convolutional neural networks are used to automatically identify faults, folds, and collapse columns in seismic data; recurrent neural networks / gated recurrent units are used to predict the temporal evolution of seepage fields and gas fields; and random forests and support vector machines are used to classify and determine the existence and development trend of activated fluid channels.

[0042] In step S106, optimizing mining parameters includes adjusting key factors; early warning processing includes increasing advance water exploration boreholes; if the predicted risk exceeds a preset threshold in the mining prediction results, an early warning is automatically triggered. Specifically, this step also includes: result output and feedback optimization, outputting the prediction results through a 3D visualization interface and transmitting them to the mine scheduling and decision-making system; when the predicted risk exceeds the preset threshold, an early warning message is automatically generated, and process optimization measures including advance water exploration, grouting reinforcement, and adjustment of advance speed are output to achieve accurate prediction and dynamic optimization of the mining process.

[0043] In this embodiment of the invention, the geological big data intelligent analysis and prediction system utilizes artificial intelligence and big data analytics to process and analyze massive amounts of geological data, providing a system for geological prediction and decision support. By combining multiple data sources such as geological exploration data, seismic data, and geological survey data, it performs intelligent data processing, pattern recognition, and predictive modeling to obtain important information such as geological structure and geological hazard risks. The method also uses historical geological data and monitoring results for model self-learning to improve prediction accuracy. Numerical simulation employs a joint coupled computation of FLAC3D, PFC, UDEC, and COMSOL Multiphysics. FLAC3D is used to simulate the evolution of the overlying strata's mechanical field, while COMSOL is used to simulate fluid-structure interaction and gas migration processes. The results are fed back through a data interface.

[0044] As can be seen from the above, in this embodiment of the invention, geological exploration of coal and rock masses is carried out by combining ground and underground coal mine exploration; multi-field / multi-source geological data of the region are acquired to establish a three-dimensional geological model; intelligent interpretation of multi-field information such as regional geological structure, coal seam occurrence, hydrology and gas is performed by combining artificial intelligence and big data analysis technology; the construction of a mine intelligent management and monitoring platform and a visual operation interface is completed; based on the results of multi-source geological exploration, sensor nodes are arranged in the model to achieve real-time linkage with various monitoring devices such as stress, seepage, gas and vibration, and to achieve dynamic updates of the three-dimensional geological model; a safety monitoring platform is formed through deep integration of multi-source monitoring data and mine production and operation data; the system receives coal mine mining parameters, performs simulation calculations through artificial intelligence and multi-field coupling methods, and outputs prediction results such as overburden migration, fracture zone development, water hazard and gas risk; the prediction results are fed back to the mine scheduling and decision-making system in real time to achieve accurate prediction and dynamic optimization of the mining process. In short, the coal mine mining prediction method provided in this invention, through detailed geological exploration, the establishment of a three-dimensional geological model, and the analysis and construction of a three-dimensional geological model using artificial intelligence and other methods, combined with various monitoring devices in the coal mine, enables dynamic updating and prediction of the three-dimensional geological model, providing a refined and visualized dynamic geological support system for intelligent coal mining. A specific example is given below, including the following steps: S201, multi-source geological data acquisition, involves fine exploration of coal and rock masses through joint surface and underground detection methods to obtain regional multi-field / multi-source geological information. Combined with statistical data from mine production and operation, the information is fused and standardized. The multi-field / multi-source geological information includes coal seam occurrence status, faults, folds, collapse columns, aquifer distribution, and gas occurrence conditions.

[0045] For example, multi-source geological exploration of the mining area is conducted using methods such as ground-penetrating radar, 3D seismic surveys, transient electromagnetic surveys, and hydrogeological drilling to obtain basic data such as stratigraphic thickness, lithological distribution, fault structures, aquifer locations, and hydrogeological parameters. Before being entered into the system, the data undergoes noise reduction, interpolation, and fusion processing to form a standardized regional geological database.

[0046] S202, 3D geological modeling, is based on a regional geological database (multi-source fusion data) and employs a multi-field coupling modeling method to establish a 3D geological model. The model integrates strata, fault structures, aquifers, and gas-rich areas, and utilizes a transparent mine geological cloud platform to create a visualized and intelligent management interface. Sequence modeling and fault constraint algorithms can be incorporated into the modeling process to ensure that the model accurately reflects the coal seam occurrence state, aquifer distribution, and fault structure characteristics.

[0047] S203, Dynamic Model Update: Virtual sensor nodes corresponding to the mine are deployed in the 3D geological model. These sensor nodes correspond to underground stress monitoring equipment, seepage monitoring equipment, gas monitoring equipment, and vibration monitoring equipment. The sensor nodes receive real-time monitoring data from these devices. The system employs dynamic correction algorithms, such as Kalman filtering or Bayesian update algorithms, to iteratively update the 3D geological model, enabling it to reflect real-time dynamic changes in the regional geology, including dynamic changes in the mechanical field, seepage field, gas field, and energy field.

[0048] S204, Mining Parameter Input: Input relevant parameters for coal mine production and operation (mining parameters) into the system. These parameters include mining height, advance speed, support method, mining depth, and ventilation conditions. The system dynamically models the working face based on the mining parameters and analyzes the input parameters using artificial intelligence algorithms (artificial intelligence feature extraction algorithms) to identify the main controlling factors (key factors) that have a critical impact on overburden migration and fracture zone development.

[0049] S205, a simulation and prediction method, utilizes intelligent geological big data analysis technology to intelligently interpret regional geological structures, coal seam occurrence, and hydrological and gas conditions. Based on a real-time updated 3D geological model and input mining parameters, it calls numerical simulation platforms (such as FLAC3D and COMSOL Multiphysics) to perform multi-field coupled calculations, combined with artificial intelligence inversion optimization algorithms, to obtain prediction results for overburden migration, fracture zone development, water hazard risk, and gas risk. The simulation and prediction process, supplemented by artificial intelligence inversion optimization methods, improves the accuracy of the prediction results.

[0050] S206, Result Output and Feedback Optimization: The prediction results are output through a 3D visualization interface and transmitted to the mine scheduling and decision-making system. When the predicted risk exceeds a preset threshold, an early warning message is automatically generated. The system automatically triggers the warning and generates corresponding process optimization suggestions / measures (such as increasing the number of advanced water exploration boreholes, adjusting the advance speed, or implementing grouting reinforcement), achieving accurate prediction and dynamic optimization of the mining process.

[0051] It should be noted that the system in the embodiments of the present invention can be understood as the coal mine mining prediction device described later.

[0052] Example 2 This invention provides a coal mine mining prediction device, which applies the coal mine mining prediction method described in the above embodiments. The device includes: The first acquisition module 100 is used to acquire basic geological data and, based on the basic geological data, obtain a regional geological database. The basic geological data includes: stratum thickness, lithological distribution, fault structure, folds, collapse columns, aquifer distribution, hydrogeological parameters, coal seam occurrence state, and gas occurrence conditions. The first modeling module 200 is used for three-dimensional geological modeling based on a regional geological database and employing a multi-field coupling modeling method. The three-dimensional geological modeling includes: geometric modeling, hydrological modeling, gas geological modeling, geostress property modeling, reserve modeling, and hidden disaster-causing modeling. The dynamic correction module 300 is used to construct a virtual sensor node network in the three-dimensional geological model that maps one-to-one with the actual monitoring equipment in the downhole space. The virtual sensor nodes receive real-time monitoring data from the actual monitoring equipment, and the three-dimensional geological model is dynamically iteratively corrected based on the data consistency constraints of the virtual sensor nodes. The actual monitoring equipment includes stress monitoring equipment, seepage monitoring equipment, gas monitoring equipment, and energy monitoring equipment. The monitoring data includes stress data, seepage data, gas data, and energy data. The second modeling module 400 is used to acquire the production condition parameters of the working face in real time during the mining process, and convert the production condition parameters into constraints of the three-dimensional geological model to perform dynamic modeling of the working face during the mining process. The production condition parameters include: actual mining height, advance displacement, advance speed, support structure parameters, original rock stress, ventilation condition parameters and mining depth. The constraints include: geometric constraints, time loading conditions, mechanical boundary conditions, gas field boundary conditions and seepage field boundary conditions. The analysis and simulation module 500 is used to perform intelligent analysis and multi-field coupled simulation based on mining parameters and a three-dimensional geological model to obtain mining prediction results. The mining prediction results include: overburden migration, fracture zone development, water hazard risk, and gas risk. The optimization processing module 600 is used to update mining decisions in real time based on mining prediction results; the mining decisions include at least: optimizing mining parameters or early warning processing.

[0053] This coal mining prediction device has all the advantages of the aforementioned coal mining prediction methods, which will not be elaborated here.

[0054] Specifically, the first acquisition module includes a geological data acquisition module and a multi-source data fusion module. The geological data acquisition module acquires multi-source geological information of the coal mining area, including geological data collected by ground-penetrating radar, 3D seismic data, transient electromagnetic data, hydrological borehole data, and downhole sensors. It also standardizes the geological data to form regional geological data. The multi-source data fusion module fuses and standardizes monitoring data of mechanical fields, seepage fields, gas fields, and energy fields to form a regional multi-field coupled database, providing data support for subsequent 3D modeling. In other words, the geological data acquisition module and the multi-source data fusion module can acquire and fuse geological data, including data from ground-penetrating radar, 3D seismic data, transient electromagnetic data, and hydrological borehole data, as well as real-time monitoring data from downhole stress sensors, seepage sensors, gas sensors, and vibration monitors.

[0055] The first modeling module includes a 3D geological modeling and correction module for mines. This module is used to build a 3D geological model of the mine based on collected multi-source geological data and to label structural features such as coal seams, faults, folds, collapse columns, and aquifers. This module can combine dynamic correction algorithms (Kalman filtering, Bayesian updating, etc.) to iteratively update model parameters using real-time sensor data, thereby ensuring consistency between the model and actual geological conditions. Specifically, the 3D geological modeling and correction module can use extended Kalman filtering and Bayesian updating algorithms to dynamically correct the model, ensuring that the simulated parameters of the mechanical field, seepage field, gas field, and energy field are consistent with real-time monitoring data. This module can also visualize the coal seam occurrence state, fault structures, and aquifer distribution through sequence modeling and fracture constraint algorithms. The dynamic correction module includes a sensor linkage module, where sensors include: stress sensors, seepage sensors, gas sensors, vibration monitors, and acoustic probes. Stress sensors acquire triaxial stress changes in the roof and coal seam to construct a mechanical field; seepage sensors acquire groundwater pressure, water level, and permeability parameters to construct a seepage field; gas sensors acquire gas concentration, outflow rate, and desorption rate to construct a gas field; and vibration monitors and acoustic probes acquire energy release rates and rupture waveforms to construct an energy field. This module embeds virtual sensor nodes into the 3D geological model, receives real-time data from stress, seepage, gas, and vibration monitoring devices, and iteratively updates the model using a dynamic correction algorithm.

[0056] The second modeling module includes a mining parameter input module and a working face dynamic modeling module. The mining parameter input module receives coal mine production / mining parameters (including mining height, advance speed, support method, mining depth, and ventilation conditions), inputs these parameters into the system, establishes a dynamic model corresponding to the working face, and performs feature extraction and sensitivity analysis using artificial intelligence algorithms. This module can simulate overburden movement, fracture evolution, and gas migration during tunneling and mining, reflecting the impact of production disturbances on the regional geological environment in real time. The working face dynamic modeling module can dynamically simulate the overburden migration pattern and fracture zone development of the working face based on parameters such as mining height, advance speed, support method, and mining depth, and outputs predictions of aquifer connectivity and abnormal gas enrichment.

[0057] The analysis and simulation module includes a geological big data intelligent analysis and prediction module, which utilizes artificial intelligence and big data analysis methods to perform deep learning and intelligent interpretation on the collected multi-source data and dynamic models. This module can call numerical simulation platforms (including FLAC3D, COMSOL Multiphysics, etc.) to perform multi-field coupled calculations, and combine artificial intelligence inversion optimization algorithms such as convolutional neural networks and genetic algorithms to optimize the prediction results and output prediction results such as overburden migration patterns (output overburden subsidence), fracture zone development height, water hazard risk level, and gas risk (e.g., abnormal gas distribution, gas emission volume).

[0058] The optimization processing module includes a feedback optimization module, which analyzes and processes the simulation prediction results and links with the mine scheduling system. When the deviation between the prediction results and the real-time monitoring results exceeds a threshold, an early warning is automatically triggered and optimization suggestions are generated to realize dynamic adjustment of the coal mine mining process.

[0059] In this embodiment of the invention, the device further includes a display module, which includes a transparent mine geological cloud platform. The transparent mine geological cloud platform has a three-dimensional visualization interface for centralized management and visualization of three-dimensional geological models, intuitive display of faults, folds, collapse columns and aquifers, and supports remote access and dynamic interaction. It can integrate geological data, production data and monitoring results into the same platform to realize the transparency and sharing of mine geological information.

[0060] In summary, in this embodiment of the invention, the coal mine mining prediction device includes a geological data acquisition and multi-source data fusion module, a transparent mine geological cloud platform, a mine three-dimensional geological modeling and correction module, a working face dynamic modeling module, and a geological big data intelligent analysis and prediction module. This invention utilizes a combination of surface and underground geological exploration of coal and rock masses. It employs artificial intelligence and big data analytics to intelligently interpret information from multiple fields, including regional geological structure, coal seam occurrence, hydrology, and gas. By acquiring multi-field / multi-source geological data, a three-dimensional geological model is established, and a mine intelligent management and monitoring platform with a visual interface is constructed. Based on the multi-source geological exploration results, sensor nodes are deployed in the model to achieve real-time linkage with monitoring equipment for stress, seepage, gas, and vibration, enabling dynamic updates to the geological model. Through deep integration of multi-source monitoring data and mine production and operation data, a safety monitoring platform is formed. The system receives coal mine mining parameters and dynamically models the working face. Using artificial intelligence and multi-field coupling methods, it simulates and calculates preset mining conditions, outputting predictions for overburden migration, fracture zone development, water hazards, and gas risks. These risk prediction results are fed back to the mine scheduling and decision-making system in real time, achieving accurate prediction and dynamic optimization of the mining process. This invention effectively guides safe coal mining and improves mining efficiency, possessing strong practical value and promotional significance.

[0061] In the description of this invention, it should be noted that the terms "upper", "lower", "front", "horizontal", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0062] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the term "installation" should be interpreted broadly. For example, it can refer to a fixed connection, a detachable connection, or an integral connection; it can refer to a mechanical connection or an electrical connection; it can refer to a direct connection or an indirect connection through an intermediate medium; it can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0063] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting coal mining operations, characterized in that, The method includes: Obtain basic geological data, and based on the basic geological data, obtain a regional geological database; the basic geological data includes: stratum thickness, lithological distribution, fault structure, folds, collapse columns, aquifer distribution, hydrogeological parameters, coal seam occurrence state and gas occurrence conditions; Based on a regional geological database, a multi-field coupling modeling method is used for three-dimensional geological modeling; the three-dimensional geological modeling includes: geometric modeling, hydrological modeling, gas geological modeling, geostress property modeling, reserve modeling, and hidden disaster-causing modeling. A virtual sensor node network is constructed in the three-dimensional geological model, with each node mapping one-to-one with the actual monitoring equipment in the downhole space. The virtual sensor nodes receive real-time monitoring data from the actual monitoring equipment, and the three-dimensional geological model is dynamically iteratively corrected based on the data consistency constraints of the virtual sensor nodes. The actual monitoring equipment includes stress monitoring equipment, seepage monitoring equipment, gas monitoring equipment, and energy monitoring equipment. The monitoring data includes stress data, seepage data, gas data, and energy data. Real-time acquisition of working face production condition parameters during mining is performed, and these parameters are converted into constraints for the three-dimensional geological model to dynamically model the working face during the mining process. The working face production condition parameters include: actual mining height, advance displacement, advance speed, support structure parameters, original rock stress, ventilation condition parameters, and mining depth. The constraints include: geometric constraints, time loading conditions, mechanical boundary conditions, gas field boundary conditions, and seepage field boundary conditions. Based on the production condition parameters and the three-dimensional geological model, intelligent analysis and multi-field coupled simulation are performed to obtain mining prediction results; the mining prediction results include: overburden migration, fracture zone development, water hazard risk and gas risk; Based on the mining prediction results, mining decisions are updated in real time; the mining decisions include at least: optimizing mining parameters or early warning processing.

2. The coal mine mining prediction method according to claim 1, characterized in that, The regional geological database obtained based on the aforementioned geological foundation data includes: Acquire geophysical data from the surface and borehole data from downhole; The geophysical data, the borehole data, and the real-time monitoring data are normalized to obtain normalized data; the normalization process includes: unified coordinate reference constraints, physical dimension normalization, and attribute consistency verification; Based on the normalized data, a standardized regional geological database is constructed for three-dimensional geological models and dynamic iterative corrections.

3. The coal mine mining prediction method according to claim 2, characterized in that, The dynamic iterative correction includes: inverting and updating the mechanical field, seepage field, and gas field parameters in the three-dimensional geological model based on real-time monitoring data, and using the deviation between historical monitoring data and current real-time monitoring data as a constraint condition to control the convergence of model parameters.

4. The coal mine mining prediction method according to claim 1, characterized in that, The mining prediction results obtained based on the production condition parameters and the three-dimensional geological model include: Using geological big data intelligent analysis technology, the regional geological structure, coal seam occurrence, and hydrological and gas conditions are intelligently interpreted; Based on the production condition parameters and the real-time updated three-dimensional geological model, a numerical simulation platform is invoked to perform multi-field coupling calculations; the numerical simulation platform includes: FLAC3D simulation platform and COMSOL Multiphysics simulation platform. By combining artificial intelligence inversion optimization algorithms, mining prediction results for overburden migration, fracture zone development, water hazard risk, gas risk, and energy risk are obtained. The overburden migration includes the overburden subsidence and the critical layer fracture height. The fracture zone development includes the development height of the water-conducting fracture zone and its risk coefficient of connection with the aquifer. The water hazard risk includes the probability of local water inrush and the expected inflow. The gas risk includes areas of abnormal gas enrichment and the maximum gas outflow. The energy risk includes the rockburst risk level of energy accumulation areas.

5. The coal mine mining prediction method according to claim 4, characterized in that, The artificial intelligence inversion optimization algorithm includes: Convolutional neural networks are used to automatically identify faults, folds, and collapse columns in seismic data. Recurrent neural networks or gated recurrent units are used to predict the temporal evolution of seepage and gas fields. Random forests and support vector machines are used to classify and determine the existence and development trend of activated fluid channels.

6. The coal mine mining prediction method according to any one of claims 1-5, characterized in that, The dynamic modeling during the working face mining process includes: By combining artificial intelligence feature extraction algorithms to analyze the production condition parameters, key factors of overburden migration and fracture zone development are identified.

7. The coal mine mining prediction method according to claim 6, characterized in that, The optimized mining parameters include: adjusting the key factors; The early warning process includes: increasing the number of advanced water exploration boreholes; and automatically triggering an early warning if the predicted risk exceeds a preset threshold in the mining prediction results.

8. A coal mine mining prediction device, characterized in that, The apparatus, applicable to the method of any one of claims 1-7, comprises: The first acquisition module is used to acquire basic geological data and, based on the basic geological data, obtain a regional geological database. The basic geological data includes: stratum thickness, lithological distribution, fault structure, folds, collapse columns, aquifer distribution, hydrogeological parameters, coal seam occurrence state, and gas occurrence conditions. The first modeling module is used to perform three-dimensional geological modeling based on a regional geological database and using a multi-field coupling modeling method. The three-dimensional geological modeling includes: geometric modeling, hydrological modeling, gas geological modeling, geostress property modeling, reserve modeling, and hidden disaster-causing modeling. A dynamic correction module is used to construct a virtual sensor node network in the three-dimensional geological model that maps one-to-one with the actual monitoring equipment in the downhole space. The virtual sensor nodes receive real-time monitoring data from the actual monitoring equipment, and the three-dimensional geological model is dynamically iteratively corrected based on the data consistency constraints of the virtual sensor nodes. The actual monitoring equipment includes stress monitoring equipment, seepage monitoring equipment, gas monitoring equipment, and energy monitoring equipment. The monitoring data includes stress data, seepage data, gas data, and energy data. The second modeling module is used to acquire the production condition parameters of the working face in real time during the mining process, and convert the production condition parameters into the constraints of the three-dimensional geological model to perform dynamic modeling of the working face during the mining process. The production condition parameters include: actual mining height, advance displacement, advance speed, support structure parameters, original rock stress, ventilation condition parameters, and mining depth. The constraints include: geometric constraints, time loading conditions, mechanical boundary conditions, gas field boundary conditions, and seepage field boundary conditions. The analysis and simulation module is used to perform intelligent analysis and multi-field coupled simulation based on the production condition parameters and the three-dimensional geological model to obtain mining prediction results. The mining prediction results include: overburden migration, fracture zone development, water hazard risk, and gas risk. An optimization processing module is used to update mining decisions in real time based on the mining prediction results; the mining decisions include at least: optimizing mining parameters or early warning processing.

9. The coal mine mining prediction method according to claim 8, characterized in that, The dynamic correction module includes: Stress sensors are used to acquire triaxial stress changes in the roof and coal seam to construct a mechanical field; Seepage sensors are used to acquire groundwater pressure, water level, and permeability parameters to construct a seepage field; Gas sensors are used to acquire gas concentration, outflow rate, and desorption rate to construct a gas field; Vibration monitors and acoustic probes are used to acquire energy release rates and fracture waveforms to construct an energy field.

10. The coal mine mining prediction method according to claim 8 or 9, characterized in that, The device further includes: The display module includes a transparent mine geological cloud platform, which has a three-dimensional visualization interface for intuitively displaying faults, folds, collapse columns and aquifers, and supports remote access and dynamic interaction.

Images