Coal mine well disaster identification method based on deep learning algorithm and related equipment

Abstract

The invention discloses a coal mine well disaster identification method based on a deep learning algorithm and related equipment. The method comprises the following steps: acquiring laser scanning point cloud data, geological drilling data and geophysical prospecting data of a coal mine well; dividing the target area into a plurality of three-dimensional grid units, and performing interpolation on the three-dimensional grid units to obtain a first three-dimensional geologic model; adding rock stratum and coal seam parameters into the first three-dimensional geologic model to obtain a second three-dimensional geologic model; associating the second three-dimensional geologic model with the big data model to obtain monitoring data of the current season; performing feature extraction on the current season and the monitoring data to obtain a multi-dimensional time sequence; inputting the multi-dimensional time sequence into a deep learning model to obtain a water level predicted value; the water level predicted value is input into the big data model for correction; performing spatial combination on the corrected water level predicted value and the second three-dimensional geological model to obtain a dynamic three-dimensional geological model; and based on the dynamic three-dimensional geologic model, determining the probability of water damage of the coal mine well.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of disaster identification, and in particular to a coal mine disaster identification method and related equipment based on a deep learning algorithm. Background Art

[0002] The most common disaster in coal mines at present is flooding. This is because the groundwater system is complex and it is difficult to avoid the infiltration of surface water during the construction of coal mines. If accompanied by extreme weather events such as heavy rain and floods, there may be hidden dangers of flooding in coal mines, which may cause damage to the equipment and infrastructure in the coal mines, and even trigger secondary disasters and cause casualties. Therefore, it is very necessary to monitor the hidden dangers of flooding in coal mines.

[0003] At present, the main method is to arrange observation wells around coal mines and regularly measure the groundwater level in the observation wells through water level gauges installed in the observation wells. The changing trend of the groundwater level can then be obtained, and the possibility of water disasters in coal mines can be predicted based on experience.

[0004] For the above monitoring methods, the water level changes in the observation well may lag behind the actual water situation changes in the coal mine, because the groundwater level usually changes with the seasons. For example, in the rainy season, the groundwater level may generally rise; in the dry season, the water level may drop. Therefore, when a sudden water situation occurs, the water level changes caused by it may be masked by seasonal changes. Therefore, seasonal groundwater level changes may mask the short-term changes caused by sudden water situations in the mine. This results in the current coal mine water hazard monitoring method being inaccurate.

[0005] There is currently no better solution to the above problems. Summary of the invention

[0006] The embodiments of the present application provide a coal mine disaster identification method and related equipment based on a deep learning algorithm, which are used to improve the accuracy of water hazard monitoring in coal mines.

[0007] To achieve the above objectives, the embodiments of the present application adopt the following technical solutions:

[0008] In a first aspect, a method for identifying coal mine disasters based on a deep learning algorithm is provided, the method comprising:

[0009] Obtain laser scanning point cloud data, geological drilling data and geophysical data of coal mines;

[0010] Dividing the target area including the coal mine into a plurality of three-dimensional grid units, and interpolating the laser scanning point cloud data, the geological drilling data and the geophysical data into corresponding three-dimensional grid units to obtain a first three-dimensional geological model;

[0011] Add rock layer parameters and coal layer parameters to the first three-dimensional geological model to obtain a second three-dimensional geological model;

[0012] Associating the second three-dimensional geological model with a pre-constructed big data model, and obtaining the current season, and obtaining monitoring data of the current season according to the big data model, wherein the monitoring data includes weather parameters, water level data, and environmental data;

[0013] Extract features of the current season, the weather parameters, the water level data, and the environmental data to obtain a multidimensional time series;

[0014] Input the multidimensional time series into a pre-built deep learning model to obtain a water level prediction value for each time unit in a future preset time period;

[0015] Inputting the water level prediction value of each time unit in a future preset time period into the big data model to perform data correction on the water level prediction value of each time unit in a future preset time period;

[0016] The revised water level prediction value is spatially combined with the second three-dimensional geological model to obtain a dynamic three-dimensional geological model including water level changes in a future preset time period;

[0017] Based on the dynamic three-dimensional geological model, the probability of water damage occurring in the coal mine is determined.

[0018] In a possible implementation of the first aspect, the target area including the coal mine is divided into a plurality of three-dimensional grid units, and the laser scanning point cloud data, the geological drilling data, and the geophysical data are interpolated to the corresponding three-dimensional grid units to obtain a first three-dimensional geological model, including:

[0019] Acquire the area of ​​the target area, and determine the density of the laser scanning point cloud data according to the laser scanning point cloud data;

[0020] Determine the grid resolution according to the area of ​​the target area and the density of the laser scanning point cloud data;

[0021] Dividing the target area into a plurality of three-dimensional grid units according to the grid resolution;

[0022] In the case where any one of the three-dimensional grid units contains the laser scanning point cloud data, the geological drilling data and / or the geophysical data, the laser scanning point cloud data, the geological drilling data and / or the geophysical data are interpolated to the three-dimensional grid unit.

[0023] In another possible implementation manner of the first aspect, determining the grid resolution according to the area of ​​the target region and the density of the laser scanning point cloud data includes:

[0024] Taking a region of a preset area as a basic unit, and dividing the target region into a plurality of sub-regions according to the basic unit;

[0025] Calculate the average density and standard deviation of the laser scanning point cloud data in each of the sub-areas;

[0026] A preset adaptive calculation formula is used to calculate the grid resolution of each sub-area according to the average density and the standard deviation, wherein when the grid resolution is less than a first preset value, the grid resolution is set to the first preset value, and when the grid resolution is greater than a second preset value, the grid resolution is set to the second preset value, and the second preset value is greater than the first preset value.

[0027] In another possible implementation of the first aspect, associating the second three-dimensional geological model with a pre-built big data model, acquiring the current season, and acquiring monitoring data of the current season according to the big data model includes:

[0028] Constructing an octree spatial index structure of the current season, wherein the octree spatial index structure includes a plurality of cubic subspaces and a plurality of nodes storing boundary coordinates and monitoring data;

[0029] Calculating the center point coordinates of each of the three-dimensional grid cells in the second three-dimensional geological model;

[0030] According to the boundary coordinates, a target cube subspace including the center point coordinates is queried in the octree space index structure, and a node corresponding to the center point coordinates is determined in the target cube subspace, and the monitoring data corresponding to the center point coordinates is the monitoring data stored in the node.

[0031] In another possible implementation of the first aspect, the deep learning model is a long short-term memory network, the long short-term memory network includes an input layer, a first LSTM layer, a second LSTM layer, a third LSTM layer, a fully connected layer and an output layer, and the multidimensional time series is input into a pre-built deep learning model to obtain a water level prediction value for each time unit in a future preset time period, including:

[0032] The multidimensional time series is input through the input layer, and passes through the first LSTM layer, the second LSTM layer, the third LSTM layer and the fully connected layer respectively, and the water level prediction value of each time unit in the future preset time period is obtained through the output layer.

[0033] In another possible implementation manner of the first aspect, inputting the water level prediction value of each time unit in the future preset time period into the big data model to perform data correction on the water level prediction value of each time unit in the future preset time period includes:

[0034] Determine the seasonal index, water level upper limit, water level lower limit and water level trend of the current season according to the big data model;

[0035] Smoothing the water level prediction value according to the seasonal index, the water level upper limit value and the water level lower limit value;

[0036] According to the water level trend, the smoothed water level prediction value is trend corrected to obtain the water level prediction value after data correction.

[0037] In another possible implementation of the first aspect, spatially combining the corrected water level prediction value with the second three-dimensional geological model to obtain a dynamic three-dimensional geological model including water level changes in a future preset time period includes:

[0038] defining the boundary and initial water level of the water body in the second three-dimensional geological model;

[0039] Calculate the water level elevation at each time point according to the revised water level prediction value;

[0040] Using a preset bilinear difference interpolation algorithm, the water level surface elevation at each time point is diffused to a preset three-dimensional water level model to obtain a water level surface grid model;

[0041] fusing the water level surface grid model with the second three-dimensional geological model to obtain a fused model;

[0042] In the fusion model, a three-dimensional model of each time point is constructed, and based on the three-dimensional model of each time point, a dynamic three-dimensional geological model including water level changes in a future preset time period is obtained.

[0043] In another possible implementation of the first aspect, determining the probability of water damage occurring in the coal mine based on the dynamic three-dimensional geological model includes:

[0044] Calculating key parameters corresponding to the three-dimensional model at each time point, the key parameters including the volume proportion of the dangerous area, the water level rise rate and the carrying capacity of the drainage system;

[0045] The probability of water damage occurring is calculated using a preset water damage calculation formula based on the key parameters.

[0046] In a second aspect, the present application provides a coal mine disaster identification device based on a deep learning algorithm, comprising:

[0047] a memory configured to store instructions; and

[0048] The processor is configured to call the instructions from the memory and implement the above-mentioned coal mine disaster identification method based on deep learning algorithm when executing the instructions.

[0049] In a third aspect, the present application provides an electronic device, including:

[0050] The above-mentioned coal mine disaster identification device based on deep learning algorithm.

[0051] Through the above technical scheme, the rock layer parameters and coal seam parameters are added to the first three-dimensional geological model to obtain the second three-dimensional geological model, which can more accurately reflect the actual geological conditions of the coal mine; the second three-dimensional geological model is associated with the big data model, and the big data model is used to obtain the monitoring data of the current season, so as to take into account the seasonal changes of the coal mine; the water level prediction using the pre-built deep learning model can improve the prediction accuracy, and the water level prediction value is corrected by using the big data model, which can further improve the prediction accuracy; the corrected water level prediction value is spatially combined with the second three-dimensional geological model to obtain a dynamic three-dimensional geological model, which can more intuitively display the future water level changes, and the probability of water disasters can be evaluated based on the dynamic three-dimensional geological model, thereby effectively improving the accuracy of water disaster monitoring in coal mines. In summary, by taking into account the seasonal changes of coal mines, the problem of seasonal changes covering up sudden water conditions is effectively solved. At the same time, through the use of deep learning models and big data models, the accuracy of water disaster monitoring in coal mines can be effectively improved.

[0052] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A schematic diagram of a process for identifying coal mine disasters based on a deep learning algorithm provided in an embodiment of the present application;

[0054] Figure 2 A schematic diagram of a target area and a three-dimensional grid unit provided in an embodiment of the present application;

[0055] Figure 3 A schematic diagram of the structure of a deep learning model provided in an embodiment of the present application.

[0056] Description of Reference Numerals

[0057] 1 target area 10 3D grid units DETAILED DESCRIPTION

[0058] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0059] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back...), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0060] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0061] Figure 1 The following schematically shows a flow chart of a method for identifying coal mine disasters based on a deep learning algorithm according to an embodiment of the present application. Figure 1 As shown, an embodiment of the present application provides a method for identifying coal mine disasters based on a deep learning algorithm, which may include the following steps.

[0062] S101, obtaining laser scanning point cloud data, geological drilling data and geophysical data of the coal mine;

[0063] S102, dividing the target area including the coal mine into a plurality of three-dimensional grid units, and interpolating the laser scanning point cloud data, geological drilling data and geophysical data into corresponding three-dimensional grid units to obtain a first three-dimensional geological model;

[0064] S103, adding rock layer parameters and coal seam parameters to the first three-dimensional geological model to obtain a second three-dimensional geological model;

[0065] S104, associating the second three-dimensional geological model with the pre-constructed big data model, and obtaining the current season, and obtaining monitoring data of the current season according to the big data model, the monitoring data including weather parameters, water level data and environmental data;

[0066] S105, extracting features of the current season, weather parameters, water level data and environmental data to obtain a multidimensional time series;

[0067] S106, inputting the multidimensional time series into a pre-built deep learning model to obtain a water level prediction value for each time unit in a future preset time period;

[0068] S107, inputting the water level prediction value of each time unit in the future preset time period into the big data model to perform data correction on the water level prediction value of each time unit in the future preset time period;

[0069] S108, spatially combining the revised water level prediction value with the second three-dimensional geological model to obtain a dynamic three-dimensional geological model including water level changes in a future preset time period;

[0070] S109. Determine the probability of water damage in coal mines based on the dynamic three-dimensional geological model.

[0071] First, the laser scanning point cloud data, geological drilling data and geophysical data of the coal mine are obtained. Specifically, the laser scanning point cloud data includes underground tunnel structure data, shaft structure data and chamber structure data.

[0072] In the specific implementation, the laser scanning point cloud data consists of discrete three-dimensional coordinate points, which are used to describe the geometric shape of the coal mine. The laser scanning point cloud data can be obtained by a laser rangefinder, which can measure the three-dimensional coordinates of the surface of an object by emitting a laser beam and receiving a reflected signal to accurately capture the underground tunnel structure data, shaft structure data and chamber structure data. Among them, the underground tunnel structure data includes the width, height, length and spatial distribution of the tunnel; the shaft structure data includes the position, diameter and shape of the shaft; the chamber structure data includes the size and spatial layout of the chamber.

[0073] Geological drilling data can be obtained by drilling and sampling around coal mines, including information such as the distribution, thickness, inclination, etc. of coal seams and various rock formations, as well as the physical properties of rock formations, such as density and porosity.

[0074] Geophysical data can obtain underground structure information through geophysical exploration methods, such as seismic wave detection, resistivity measurement, etc. In addition, in this embodiment, historical construction data of coal mines can also be obtained, and mining history, support methods, drainage systems, etc. can be obtained through the historical construction data of coal mines.

[0075] The above multi-source data collection can help identify potential water risk areas. For example, through laser scanning point cloud data, it can be found that the width of a certain tunnel gradually narrows from 3 meters at the entrance to 2.5 meters, the length is 500 meters, and the height remains at 2.8 meters; while geological drilling data may show that there is a 2-meter-thick water-bearing sandstone layer 50 meters above the tunnel.

[0076] The target area 1 including the coal mine is divided into a plurality of three-dimensional grid units 10, each of which represents a unit volume in space. Figure 2 A schematic diagram of a target area and a three-dimensional grid unit provided in an embodiment of the present application is shown, such as Figure 2 As shown, when the target area 1 is a cube, the three-dimensional grid unit 10 is a cube grid unit.

[0077] In this embodiment, the laser scanning point cloud data, geological drilling data and geophysical data can be interpolated into corresponding three-dimensional grid cells using spatial interpolation technology. For the laser scanning point cloud data, the laser scanning point cloud data can be distributed into the grid using the nearest neighbor interpolation method.

[0078] Since geological drilling data is usually discrete point data, it can be extended in space using Kriging interpolation. Geophysical data can be three-dimensionally interpolated using geostatistical methods. For example, assuming that the target area 1 including the coal mine is 100mx100mx50m, it can be divided into 1mx1mx1m cubic grid units (three-dimensional grid units 10), totaling 500,000 units. Then, each three-dimensional grid unit is assigned a corresponding attribute value, such as a lithology code, a coal code, etc., and the laser scanning point cloud data, geological drilling data, and geophysical data are interpolated into the corresponding three-dimensional grid unit 10, respectively, to finally obtain the first three-dimensional geological model, wherein the first three-dimensional geological model is a voxel model containing multiple geological attributes.

[0079] Adding rock layer parameters and coal seam parameters to the first three-dimensional geological model can enhance the geophysical property characterization capability of the first three-dimensional geological model. Specifically, the rock layer parameters include permeability coefficient and strength parameters, and the coal seam parameters include thickness parameters, inclination parameters and burial depth parameters.

[0080] The permeability coefficient in the rock formation parameters describes the permeability of the rock to the fluid, usually in units of Darcy (D) or centimeters per second (cm / s). For example, the permeability coefficient of sandstone may be 10 -3 to 10 -5 cm / s, while shale may be as low as 10 -8cm / s. Strength parameters include rock compressive strength, tensile strength, cohesion and internal friction angle. Coal seam parameters include thickness parameters, inclination parameters and burial depth parameters. The uniaxial compressive strength of rock can be determined by rock mechanics tests, while the permeability coefficient can be determined by hydraulic fracturing tests.

[0081] For each three-dimensional grid unit, corresponding rock layer parameters can be added according to its corresponding lithology; similarly, corresponding coal seam parameters can be added according to its corresponding coal properties.

[0082] Afterwards, the second three-dimensional geological model is associated with the pre-constructed big data model. The pre-constructed big data model includes multi-source data such as historical geological data, historical disaster records, and environmental monitoring data. The association process first needs to establish a spatial correspondence between the second three-dimensional geological model and the pre-constructed big data model to ensure that each three-dimensional grid unit in the second three-dimensional geological model can accurately correspond to the geographical location in the big data model. In specific implementation, it can be achieved through spatial indexing technology. For example, a spatial index structure can be used to accelerate the spatial query and matching process. Through association, the second three-dimensional geological model can better capture regional geological characteristics and long-term environmental change trends, thereby improving the accuracy and reliability of disaster prediction.

[0083] Get the current season and obtain the corresponding monitoring data according to the big data model. In this embodiment, the current season can be realized by the system clock or user input. According to the determined season, the corresponding monitoring data is extracted from the big data model. Weather parameters include temperature, precipitation, humidity, wind speed, etc. For example, the daily average temperature, daily cumulative precipitation, daily changes in relative humidity, etc. can be obtained. Water level data includes surface water level and groundwater level, wherein the surface water level includes the water level changes of rivers and lakes near the coal mine shaft, and the groundwater level can be obtained through observation wells. Environmental data includes soil moisture content, surface runoff, vegetation coverage and surface settlement, and environmental data can be obtained through a geographic information database.

[0084] The current season, weather parameters, water level data and environmental data are feature extracted to obtain a multidimensional time series. For seasonal information, one-hot encoding can be used to convert it into numerical features. For example, spring can be encoded as [1,0,0,0], summer as [0,1,0,0], and so on. For weather parameters, statistical features such as moving average and standard deviation can be calculated to capture short-term and long-term weather change trends. For example, the average temperature and precipitation standard deviation of the past 7 days can be calculated. Trend, periodicity and autocorrelation features can be extracted from water level data, such as using wavelet transform to decompose water level changes at different scales. The soil moisture content in the environmental data can be calculated to calculate the saturation index, the surface runoff can be extracted to extract the base flow separation index, and the vegetation coverage can be calculated to calculate the growing season index. Among them, the current season can be represented by digital encoding to represent the periodic time series, and the weather parameters, water level data and environmental data are all time series data. Therefore, the current season, weather parameters, water level data and environmental data can be feature extracted to obtain a multidimensional time series.

[0085] It should be noted that in the process of feature extraction of the current season, weather parameters, water level data, and environmental data, it is necessary to ensure that all data are aligned in time so that all sequences have the same time interval.

[0086] Next, the multidimensional time series is input into the pre-built deep learning model to obtain the water level forecast value for each time unit in the future preset time period.

[0087] Among them, the pre-built deep learning model can effectively capture the long-term dependencies and nonlinear patterns in time series data. In the model training process of this embodiment, the Adam optimizer is used, the learning rate is set to 0.001, the batch size is 64, and the number of training rounds is 1000 rounds. The loss function is the root mean square error. In the prediction stage, the deep learning model uses a sliding window method, and inputs a time series of a fixed length (such as the past 30 days) each time to predict the water level in the future (such as the next 7 days). Then the sliding window is used to take part of the prediction results as new inputs, and the next time period is predicted until the entire preset time period is covered. The final output is the water level prediction value for each time unit (such as every day or every hour) in the future preset time period.

[0088] The water level prediction value of each time unit in the future preset time period is input into the big data model for data correction, and the water level trend of the water level prediction value can be corrected according to the water level trend of the current season in the big data model.

[0089] The revised water level prediction value is spatially combined with the second three-dimensional geological model to obtain a dynamic three-dimensional geological model including water level changes in a preset time period in the future. Specifically, first, the revised water level prediction data is converted into a format suitable for three-dimensional spatial representation, such as using an interpolation algorithm to convert discrete water level data into a continuous water level surface. Then, the water level surface data is spatially superimposed and fused with the second three-dimensional geological model.

[0090] In one embodiment, spatial analysis software such as ArcGIS or QGIS can be used in combination with a 3D modeling tool such as Gocad or Petre l. Specific steps may include: (1) importing water level prediction data into ArcGIS to create a water level grid; (2) registering the water level grid with the second 3D geological model in the same coordinate system; (3) fusing the water level grid with the second 3D geological model in Gocad to obtain a dynamic 3D geological model.

[0091] The dynamic three-dimensional geological model not only contains static geological structure information, but also incorporates dynamic water level information that changes over time, which can intuitively show the impact of water level changes on coal mines.

[0092] Finally, the probability of water damage in coal mines is determined based on the dynamic three-dimensional geological model. First, key risk factors need to be extracted from the dynamic three-dimensional geological model, such as the volume proportion of the dangerous area, the speed of water level rise, the strength of the rock formation, the location of the fault, and the bearing capacity of the drainage system. Based on the risk factors, the probability of water damage in coal mines is calculated using the preset calculation formula.

[0093] In this embodiment, rock layer parameters and coal seam parameters are added to the first three-dimensional geological model to obtain a second three-dimensional geological model, which can more accurately reflect the actual geological conditions of the coal mine; the second three-dimensional geological model is associated with the big data model, and the monitoring data of the current season is obtained by using the big data model, so as to take into account the seasonal changes of the coal mine; the water level prediction using the pre-built deep learning model can improve the prediction accuracy, and the water level prediction value is corrected by using the big data model, which can further improve the prediction accuracy; the corrected water level prediction value is spatially combined with the second three-dimensional geological model to obtain a dynamic three-dimensional geological model, which can more intuitively display future water level changes, and the probability of water disasters can be evaluated based on the dynamic three-dimensional geological model, thereby effectively improving the accuracy of water disaster monitoring in coal mines. In summary, by taking into account the seasonal changes of coal mines, the problem of seasonal changes covering up sudden water conditions is effectively solved. At the same time, through the use of deep learning models and big data models, the accuracy of water disaster monitoring in coal mines can be effectively improved.

[0094] In one implementation of this embodiment, a target area including a coal mine is divided into a plurality of three-dimensional grid units, and laser scanning point cloud data, geological drilling data, and geophysical data are interpolated to corresponding three-dimensional grid units to obtain a first three-dimensional geological model, including the following steps:

[0095] S201, obtaining the area of ​​the target area, and determining the density of the laser scanning point cloud data according to the laser scanning point cloud data;

[0096] S202, determining the grid resolution according to the area of ​​the target area and the density of the laser scanning point cloud data;

[0097] S203, dividing the target area into a plurality of three-dimensional grid units according to the grid resolution;

[0098] S204: When any three-dimensional grid unit contains laser scanning point cloud data, geological drilling data and / or geophysical data, interpolate the laser scanning point cloud data, geological drilling data and / or geophysical data to the three-dimensional grid unit.

[0099] First, obtain the area of ​​the target area, which can be done through surveying or GIS systems. The unit of area can be square meters or square kilometers. Next, determine its density based on the laser scanning point cloud data. The density of laser scanning point cloud data is usually expressed as the number of points per square meter, which can be obtained by calculating the total number of points divided by the covered area. For example, if there are 80 million points in an area of ​​800,000 square meters, then the average density is 100 points / square meter.

[0100] Determine the grid resolution based on the area of ​​the target area and the density of the laser scanning point cloud data. Generally speaking, the grid resolution should be less than or equal to the average point spacing of the point cloud data to ensure that important terrain details are not lost. The following formula can be used to estimate the appropriate grid resolution:

[0101]

[0102] For example, if the point cloud density is 100 points / square meter, the grid resolution is about 0.1 meters. However, in a specific implementation, each grid resolution may be different. If there is a 5-meter-wide fault in the target area, the grid resolution should be much smaller than 5 meters in order to accurately represent the fault.

[0103] Afterwards, the target area is divided into multiple three-dimensional grid cells according to the determined grid resolution. First, the spatial range of the entire model needs to be determined, including the minimum and maximum coordinate values ​​in the three directions of X, Y, and Z. Then, according to the previously determined grid resolution, equidistant divisions are performed in each direction. For example, if the range in the X direction is 0-1000 meters and the grid resolution is 1 meter, then there will be 1000 grids in the X direction. The Y and Z directions are processed similarly. In this way, the entire target area is divided into a large number of cubic or rectangular grid cells. Each grid cell has a unique spatial coordinate and index, which facilitates the interpolation and attribute assignment of subsequent data. In the specific implementation, in order to improve the computational efficiency, higher resolution grids can be used in key areas, and lower resolution grids can be used in areas with little change, that is, adaptive grid division is performed.

[0104] In the case where any three-dimensional grid unit contains laser scanning point cloud data, geological drilling data and / or geophysical data, the laser scanning point cloud data, geological drilling data and / or geophysical data are interpolated to the three-dimensional grid unit.

[0105] In this embodiment, the laser scanning point cloud data, geological drilling data and geophysical data can be interpolated into corresponding three-dimensional grid cells using spatial interpolation technology. For the laser scanning point cloud data, the laser scanning point cloud data can be distributed into the grid using the nearest neighbor interpolation method.

[0106] Since geological drilling data is usually discrete point data, it can be extended in space using Kriging interpolation. Geophysical data can be interpolated in three dimensions using geostatistical methods. For example, assuming that the target area including the coal mine is 100mx100mx50m, it can be divided into 1mx1mx1m cube grid cells (three-dimensional grid cells), totaling 500,000 cells. Then assign corresponding attribute values ​​to each cell, such as lithology code, coal code, etc., and interpolate the laser scanning point cloud data, geological drilling data and geophysical data into the corresponding three-dimensional grid cells respectively, and finally obtain the first three-dimensional geological model, wherein the first three-dimensional geological model is a voxel model containing multiple geological attributes.

[0107] The first three-dimensional geological model constructed in this embodiment integrates laser scanning point cloud data, geological drilling data and geophysical data, and can accurately reflect the geological characteristics and spatial distribution of the target area. The accuracy of the first three-dimensional geological model is ensured by interpolating laser scanning point cloud data, geological drilling data and / or geophysical data into three-dimensional grid units.

[0108] In one implementation of this embodiment, determining the grid resolution according to the area of ​​the target area and the density of the laser scanning point cloud data includes the following steps:

[0109] S301, taking a region of a preset area as a basic unit, and dividing a target region into a plurality of sub-regions according to the basic unit;

[0110] S302, calculating the average density and standard deviation of the laser scanning point cloud data in each sub-area;

[0111] S303. Using a preset adaptive calculation formula, the grid resolution of each sub-area is calculated according to the average density and the standard deviation, wherein when the grid resolution is less than a first preset value, the grid resolution is set to the first preset value; and when the grid resolution is greater than a second preset value, the grid resolution is set to the second preset value, and the second preset value is greater than the first preset value.

[0112] First, a region of a preset area is used as a basic unit, and the target region is divided into multiple sub-regions according to the basic unit. The basic unit can be determined based on the total area of ​​the target region and the complexity of the terrain. For example, for a target region with a total area of ​​10 square kilometers, the basic unit can be an area of ​​100 meters by 100 meters, so that the entire target region is divided into 1,000 sub-regions.

[0113] In one embodiment, GIS (Geographic Information System) software can be used for division. A regular grid can be created to cover the entire target area, and the area of ​​each grid cell is the preset basic unit area. If the shape of the target area is irregular, some grid cells may only partially cover the target area. In this case, it is possible to select to retain those grid cells that cover the target area and exceed 50%.

[0114] In another embodiment, if there are obvious geographical or geological boundaries (such as rivers, faults) in the target area, the above-mentioned grid division may ignore the natural boundaries. In this case, the average density and standard deviation of the laser scanning point cloud data in each sub-area can be calculated, and the average density and standard deviation of the laser scanning point cloud data can determine the distribution characteristics of the three-dimensional scanning point cloud data.

[0115] Specifically, first, for each sub-region, all point cloud data points that fall within the region need to be determined. It can usually be checked whether the coordinates of each point are determined within the boundary of the sub-region. The average density of the laser scanning point cloud data can be obtained by dividing the total number of point cloud data points in the sub-region by the area of ​​the sub-region.

[0116] The steps for calculating the standard deviation are as follows: further divide the sub-area into smaller grids (for example, a small grid of 1 meter × 1 meter); calculate the number of points in each small grid to obtain multiple density samples; use the standard deviation calculation formula to calculate the standard deviation of the laser scanning point cloud data.

[0117] The formula for calculating the standard deviation is:

[0118]

[0119] Among them, x is the density of each small grid, μ is the average density, and n is the number of small grids.

[0120] In actual operation, some sub-areas may contain little or no laser scanning point cloud data, which may be caused by occlusion during the scanning process or incomplete data collection. In this case, interpolation or estimation is needed to fill in the laser scanning point cloud data of these sub-areas.

[0121] The preset adaptive calculation formula is used to calculate the grid resolution of each sub-area based on the average density and standard deviation.

[0122] The adaptive calculation formula is:

[0123]

[0124] Among them, k is a scaling factor used to adjust the overall resolution level; α is a preset weight coefficient used to control the influence of the standard deviation on the resolution. In the adaptive calculation formula, the resolution is inversely proportional to the point cloud density, and the standard deviation reflects the unevenness of the density distribution.

[0125] For example, assuming that the average density of a sub-region is 100 points / square meter, the standard deviation is 20 points / square meter, k = 0.5, α = 0.1, then the grid resolution of the sub-region is calculated as follows:

[0126]

[0127] However, in this embodiment, in order to avoid the grid resolution being too small or too large, upper and lower thresholds are set. If the calculated grid resolution is less than the first preset value (for example, 0.05 meters), the grid resolution is set to the first preset value. It is used to prevent too many grids from being generated in high-density areas, resulting in excessive computational burden. Similarly, if the calculated grid resolution is greater than the second preset value (for example, 1 meter), the grid resolution is set to the second preset value. It is used to ensure that a certain model accuracy can still be maintained in low-density areas.

[0128] This implementation first divides the target area into multiple sub-areas, then performs statistical analysis on the laser scanning point cloud data of each sub-area to obtain the average density and standard deviation, and finally dynamically adjusts the grid resolution based on the average density and standard deviation. The grid design can be optimized based on the actual distribution characteristics of the laser scanning point cloud data, improving the computational efficiency while ensuring the model accuracy. By setting upper and lower thresholds, extreme grid sizes can also be avoided to ensure the overall quality of the model. The above method of adaptively adjusting the grid resolution can achieve the best balance between accuracy and efficiency between different sub-areas.

[0129] In one implementation of this embodiment, the second three-dimensional geological model is associated with a pre-constructed big data model, and the current season is obtained. The monitoring data of the current season is obtained according to the big data model, including the following steps:

[0130] S401, constructing an octree spatial index structure of the current season, where the octree spatial index structure includes a plurality of cube subspaces and a plurality of nodes storing boundary coordinates and monitoring data;

[0131] S402, calculating the center point coordinates of each three-dimensional grid cell in the second three-dimensional geological model;

[0132] S403. According to the boundary coordinates, query the target cube subspace including the center point coordinates in the octree space index structure, and determine the node corresponding to the center point coordinates in the target cube subspace, and the monitoring data corresponding to the center point coordinates is the monitoring data stored in the node.

[0133] The octree spatial index structure is a tree data structure in which each node has at most eight child nodes. In three-dimensional space, it organizes data by recursively dividing the space into eight equal cubic subspaces.

[0134] The construction process of the octree spatial index structure starts with a root cube that contains the entire target area. Then, based on predefined conditions (such as data density), the root cube is recursively subdivided into eight smaller sub-cubes. Each sub-cube becomes a new node in the octree. The above process is executed in a loop until a preset termination condition is reached, such as the minimum cube size or the maximum tree depth.

[0135] Two types of key information are stored on each node: boundary coordinates and monitoring data. The boundary coordinates define the range of the cube subspace represented by the node, usually including six values: xmin, xmax, ymin, ymax, zmin, zmax. For example, suppose the target area is a cube with a side length of 1000 meters. The first level of division will produce 8 sub-cubes with a side length of 500 meters. If the data points contained in one of the sub-cubes exceed the preset threshold (such as 1000 points), the sub-cube is further divided into 8 smaller sub-cubes with a side length of 250 meters. The above process is executed in a loop until all sub-cubes meet the termination conditions or no longer contain any data points.

[0136] The octree spatial index structure can adaptively adjust the fineness of the spatial division according to the distribution characteristics of the data. In areas with dense data, the octree will have deeper levels and smaller sub-cubes, while in areas with sparse data, there will be larger sub-cubes, allowing the octree to greatly reduce storage space and improve query efficiency while maintaining high accuracy.

[0137] Next, calculate the center point coordinates of each three-dimensional grid cell in the second three-dimensional geological model. Specifically, in three-dimensional space, each three-dimensional grid cell is usually a hexahedron. The basic principle of calculating the center point coordinates is to determine the geometric center of the hexahedron. For a regular cuboid, the center point coordinates can be calculated in the following way: x_center = (x_min + x_max) / 2; y_center = (y_min + y_max) / 2; z_center = (z_min + z_max) / 2. Among them, x_min, x_max, y_min, y_max, z_min, z_max are the minimum and maximum coordinate values ​​of the grid cell in the three directions of x, y, and z, respectively.

[0138] Finally, according to the boundary coordinates, the target cube subspace including the center point coordinates is queried in the octree spatial index structure, and the node corresponding to the center point coordinates is determined in the target cube subspace. This embodiment starts from the root node of the octree and compares the center point coordinates with the boundary of the cube subspace represented by the current node. If the center point falls within the current cube subspace, the eight child nodes of the current node are continuously checked until a leaf node is reached or the minimum subspace containing the node is found.

[0139] For example, suppose there is a grid cell with a center point coordinate of (105,210,315). The query process can be as follows:

[0140] 1. First check the root node, assuming that the spatial range it represents is (0-1000,0-1000,0-1000), and the center node is within this range.

[0141] 2. Check the 8 child nodes of the root node and assume that a child node with a range of (0-500,0-500,0-500) is found that contains the center node.

[0142] 3. Continue checking the child nodes of the child node with the range (0-500,0-500,0-500), and you may find a child node with the range (0-250,125-375,250-500).

[0143] 4. If the child node in the range (0-250, 125-375, 250-500) is the smallest subspace containing the central node, the query stops.

[0144] After the target cube subspace is determined, the monitoring data stored in the nodes corresponding to the target cube subspace (the nodes corresponding to the coordinates of the center point) can be accessed.

[0145] This implementation method first constructs an octree spatial index structure, then calculates the coordinates of the center point of each grid unit to provide a basis for data matching, and finally uses the octree for spatial query to achieve accurate correspondence between the second three-dimensional geological model and the monitoring data, effectively improving the efficiency of data retrieval and matching, and thus significantly improving the accuracy of the second three-dimensional geological model.

[0146] In one implementation of this embodiment, the deep learning model is a long short-term memory network, which includes an input layer, a first LSTM layer, a second LSTM layer, a third LSTM layer, a fully connected layer, and an output layer. The multidimensional time series is input into the pre-built deep learning model to obtain the water level prediction value of each time unit in the future preset time period, including the following steps:

[0147] S501, input the multidimensional time series through the input layer, and pass through the first LSTM layer, the second LSTM layer, the third LSTM layer and the fully connected layer respectively, and obtain the water level prediction value of each time unit in the future preset time period through the output layer.

[0148] Figure 3 A schematic diagram of the structure of a deep learning model provided in an embodiment of the present application is shown. Figure 3 As shown, the deep learning model includes an input layer, a first LSTM layer, a second LSTM layer, a third LSTM layer, a fully connected layer, and an output layer.

[0149] The multidimensional time series is input through the input layer, and passes through the first LSTM layer, the second LSTM layer, the third LSTM layer and the fully connected layer respectively, and finally the water level prediction value of each time unit in the future preset time period is obtained through the output layer. The long short-term memory network can be used to capture the long-term dependencies in the time series data, thereby realizing accurate prediction of future water levels.

[0150] First, multidimensional time series data enters the network through the input layer. The input layer is used to convert multidimensional time series data into a format suitable for network processing. For example, if there are n features, the input for each time step may be an n-dimensional vector. Assuming that the data of the past 30 days is used to predict the future, the input may be a matrix with a shape of (30,n).

[0151] Next, the multidimensional time series data passes through three LSTM layers in sequence. Each LSTM layer contains multiple LSTM units, which can learn and remember long-term dependencies. LSTM units include forget gates, input gates, and output gates to capture long-term patterns in time series.

[0152] Taking the first LSTM layer as an example, its mathematical expression can be simplified as:

[0153] ft = σ(Wf · [ht-1, xt] + bf) / / forget gate;

[0154] / / Input gate;

[0155] / / Output gate;

[0156] / / Cell status update;

[0157] ht=ot*tanh(ct) / / Hidden state output.

[0158] Where σ is the sigmoid function, * represents the Hadamard product, · represents matrix multiplication, xt is the input of the current time step, ht-1 is the hidden state of the previous time step, and ct-1 is the cell state of the previous time step.

[0159] The second and third LSTM layers work similarly, but their input is the output of the previous layer. Stacking multiple LSTM layers enables deep learning networks to learn more complex temporal dependencies.

[0160] After the LSTM layer, the data passes through the fully connected layer. The fully connected layer is used to map the features extracted by the LSTM layer to the final predicted value. It can be expressed as:

[0161] y=f(Wx+b);

[0162] Here, x is the input vector, W is the weight matrix, b is the bias vector, and f is the activation function (such as ReLU).

[0163] Finally, the final water level prediction value is obtained through the output layer. The output layer usually uses a linear activation function to directly output the predicted water level value. In this embodiment, the long short-term memory network is trained by a back propagation algorithm, and the goal is to minimize the error between the predicted value and the actual water level.

[0164] This implementation method can effectively handle the complex time dependency in water level prediction by adopting a multi-layer LSTM network structure. The input layer receives multi-dimensional time series data, the three LSTM layers gradually extract and learn the long-term and short-term patterns in the time series, the fully connected layer performs feature integration, and finally the output layer gives a specific water level prediction value, thereby providing a more accurate water level prediction.

[0165] In one implementation of this embodiment, the water level prediction value of each time unit in the future preset time period is input into the big data model to perform data correction on the water level prediction value of each time unit in the future preset time period, including the following steps:

[0166] S601. Determine the seasonal index, water level upper limit, water level lower limit and water level trend of the current season according to the big data model;

[0167] S602, smoothing the water level forecast value according to the seasonal index, the water level upper limit value and the water level lower limit value;

[0168] S603. According to the water level trend, the smoothed water level prediction value is trend corrected to obtain a water level prediction value after data correction.

[0169] The seasonal index, upper and lower limits of the water level, and water level trend of the current season are determined based on the big data model. Specifically, the seasonal index reflects the cyclical change pattern of the water level in different seasons. It is calculated by decomposing the time series, breaking down the original data into trend, seasonal, and random components. For example, for monthly data, a 12-month centralized moving average can be calculated, and then the original data can be divided by this moving average to obtain the seasonal factor. Repeat this process for multiple years of data, and then average the seasonal factors for each month to obtain the seasonal index.

[0170] The upper and lower limits of the water level represent the reasonable range of the water level in a particular season. They can be determined by analyzing the distribution of historical data. For example, the 95% quantile can be used as the upper limit and the 5% quantile as the lower limit, which can exclude the influence of extreme outliers while retaining most of the normal fluctuations.

[0171] The water level trend reflects the direction of long-term change. It can be determined by linear regression. For example, when using linear regression, time can be used as the independent variable and water level as the dependent variable to fit a straight line. The slope of the line represents the water level trend.

[0172] Smoothing the water level forecast values ​​according to the seasonal index, upper and lower water level limits can reduce abnormal fluctuations in the forecast and make the forecast results more consistent with seasonal laws and a reasonable range.

[0173] A possible implementation is as follows: First, divide the original forecast value by the corresponding seasonal index to obtain the deseasonalized value: deseasonalized value = original forecast value / seasonal index; then, smooth the deseasonalized value. Use the simple moving average method: smoothed value = (sum of the first n deseasonalized values) / n, where n can be an odd number between 3 and 7; next, multiply the smoothed value by the seasonal index to restore the seasonal characteristics: smoothed value of restored seasonality = smoothed value * seasonal index; finally, check whether the smoothed value of restored seasonality is between the upper and lower limits of the water level. If it is out of range, it is limited to the nearest boundary value.

[0174] According to the water level trend, the smoothed water level forecast value is trend-corrected to obtain the data-corrected water level forecast value, and the steps are as follows: First, calculate the difference between the smoothed forecast value and the trend line: Difference = smoothed forecast value - trend line forecast value, where the trend line forecast value can be the water level trend determined above. Then, according to the size and sign of the difference, adjust the smoothed forecast value: adjusted forecast value = smoothed forecast value + α*(trend line forecast value - smoothed forecast value), where α is an adjustment coefficient between 0 and 1, which is used to control the intensity of trend correction. The larger the α value, the closer the corrected forecast value is to the trend line, and the smaller the α value, the more features of the original forecast are retained; finally, check again whether the adjusted forecast value is within the preset range.

[0175] This implementation method first uses the big data model to determine the key parameters to provide a basis for correction. Then, smoothing is performed based on the seasonal index and upper and lower limits to eliminate abnormal fluctuations. Finally, through trend correction, the prediction results are ensured to be consistent with the long-term change trend, which can effectively improve the accuracy and reliability of water level prediction and make the prediction results more consistent with the actual hydrological characteristics.

[0176] In one implementation of this embodiment, the revised water level prediction value is spatially combined with the second three-dimensional geological model to obtain a dynamic three-dimensional geological model including water level changes in a future preset time period, including the following steps:

[0177] S701, defining the boundary and initial water level of the water body in the second three-dimensional geological model;

[0178] S702, calculating the water level elevation at each time point according to the corrected water level prediction value;

[0179] S703, using a preset bilinear difference interpolation algorithm to diffuse the water level surface elevation at each time point to a preset three-dimensional water level model to obtain a water level surface grid model;

[0180] S704, fusing the water level surface grid model with the second three-dimensional geological model to obtain a fused model;

[0181] S705: In the fusion model, a three-dimensional model of each time point is constructed, and based on the three-dimensional model of each time point, a dynamic three-dimensional geological model including water level changes in a future preset time period is obtained.

[0182] First, the boundary and initial water level of the water body are defined in the second 3D geological model.

[0183] Specifically, first, the boundary of the water body needs to be determined. The boundary of the water body can be identified by analyzing the slope and elevation changes of the terrain. For example, the hydrological analysis tool in the GIS software can be used to automatically extract the outline of the water body by setting an elevation threshold. Next, define the initial water level surface. The initial water level surface represents the state of the water body at the beginning of the simulation. It can be determined by historical observation data. Interpolation methods can be used to expand discrete water level observation points (historical observation data) to the entire water body range. In the second three-dimensional geological model, the water body boundary and water level surface are usually represented as three-dimensional polygons or grids. The water body boundary is a closed polygon composed of a series of three-dimensional coordinate points (x, y, z), and the initial water level surface is a plane or surface composed of a series of three-dimensional points with the same elevation value.

[0184] Calculating the water level surface elevation at each time point based on the revised water level forecast value is a key step in converting the one-dimensional water level forecast data into a three-dimensional spatial representation. First, determine the base elevation of the water level forecast point. Next, convert the revised water level forecast value into the water level surface elevation, that is, add the water level forecast value to the base elevation of the water level forecast point. For example, if the predicted water level at a certain time point is 3 meters, and the base elevation of the forecast point is 50 meters, then the water level surface elevation at that time point is 53 meters.

[0185] The preset bilinear difference interpolation algorithm is used to diffuse the water level surface elevation at each time point to the preset water level three-dimensional model to obtain a water level surface grid model to convert discrete water level data into a continuous surface.

[0186] Bilinear interpolation is a method of interpolation in two-dimensional space, which performs linear interpolation in the x and y directions respectively. Its advantages are fast calculation speed and can produce smooth surfaces.

[0187] The specific implementation steps are as follows: 1. First, create a regular grid in the water area. For example, you can create a 10m*10m grid. 2. For each point in the grid, find the four nearest known water level points around it. These four points form a rectangle, and the point to be interpolated is located in this rectangle. 3. Use the bilinear difference interpolation formula to calculate the water level of the point. 4. Repeat step 3 until all points in the grid are assigned water level values.

[0188] The bilinear difference interpolation formula is as follows:

[0189] f(x,y)≈f(x1,y1)(x2-x)(y2-y)+f(z2,y1)(x-x1)(y2-y)+f(x1,y2)(x2-x)(y-y1)+f(x2,y2)(x-r1)(y-y1);

[0190] Among them, (x, y) is the coordinate of the point to be interpolated, (x1, y1), (x2, y1), (x1, y2), (x2, y2) are the coordinates of the four surrounding known points, and f(x, y) represents the function value (i.e. water level elevation) at point (x, y).

[0191] The water level surface grid model is integrated with the second three-dimensional geological model to obtain a fused model, including the following steps:

[0192] 1. Spatially align the water level grid model with the second three-dimensional geological model. Specifically, the water level grid model and the three-dimensional geological model use the same coordinate system and spatial reference.

[0193] 2. Convert the water surface grid in the water surface grid model into a format compatible with the second three-dimensional geological model. For example, the water surface grid can be converted into a triangular irregular network or directly inserted into the voxel model (the second three-dimensional geological model).

[0194] 3. Through Boolean operations, the strata in the second three-dimensional geological model are connected with the water level surface in the water level surface grid model.

[0195] 4. Add the water surface elevation of the water surface grid model as attribute data to the second three-dimensional geological model.

[0196] In the fusion model, a three-dimensional model is constructed for each time point, and based on the three-dimensional model at each time point, a dynamic three-dimensional geological model including water level changes in the future preset time period is obtained. The specific implementation process is as follows: 1. Create a separate three-dimensional model for each time point in the preset time period. You can choose every hour or every day as a time point. 2. For each time point, determine the specific water level at that time point based on the water level prediction data. 3. For each time point, update the water level information in the fusion model. Specifically, it includes updating the geometry and attribute data of the water body. 4. Link the three-dimensional models of each time point to form a continuous time series. Specifically, it can be achieved by creating a data structure containing all time point models, such as a dictionary or a list, in which each element corresponds to the model state of a time point. In summary, a dynamic three-dimensional geological model including water level changes in the future preset time period can be obtained.

[0197] This implementation method integrates static geological information and dynamic hydrological data, which can show the impact of water level changes on the geological environment in the future and help predict potential coal mine disasters.

[0198] In one implementation of this embodiment, determining the probability of water damage in a coal mine based on a dynamic three-dimensional geological model includes the following steps:

[0199] S801, calculating key parameters corresponding to the three-dimensional model at each time point, the key parameters including the volume proportion of the dangerous area, the water level rise speed, and the carrying capacity of the drainage system;

[0200] S802: Calculate the probability of water damage occurrence based on key parameters using a preset water damage calculation formula.

[0201] First, calculate the volume fraction of the danger zone. The danger zone is usually defined as an area where the water level is above a certain threshold. The volume fraction can be obtained by traversing each voxel (or grid cell) in the 3D model, counting the number of voxels where the water level exceeds the safety threshold, and then dividing it by the total number of voxels.

[0202] The water level rise rate can be calculated by comparing the water level difference between two adjacent time points. For the entire target area, the average rise rate or the maximum rise rate can be calculated.

[0203] The carrying capacity of the drainage system is determined by the performance of the drainage pump, hydraulic losses and hydrostatic pressure.

[0204] For example, suppose there are three dewatering pumps in a coal mine, each with a rated flow of 500 cubic meters per hour and a head of 100 meters. The hydrostatic pressure is 50 meters, and the pipe friction and local losses are calculated to be 30 meters.

[0205] The total theoretical drainage capacity = 500*3 = 1500 cubic meters / hour, the actual working point head = 50+30 = 80 meters. Referring to the performance curve of the pump, the actual flow rate of each pump may drop to 450 cubic meters / hour at a head of 80 meters. The actual total drainage capacity = 450*3 = 1350 cubic meters / hour. Applying a safety factor of 1.2, the carrying capacity of the drainage system = 1350 / 1.2≈1125 cubic meters / hour.

[0206] The formula for calculating the probability of water damage includes: P (water damage) = w1*f1 (volume ratio of dangerous area) + w2*f2 (water level rise rate) + w3*f3 (carrying capacity of drainage system);

[0207] Among them, w1 to w3 are the preset weights of the volume proportion of the dangerous area, the water level rise rate and the carrying capacity of the drainage system, and f1 to f3 are the risk functions corresponding to the volume proportion of the dangerous area, the water level rise rate and the carrying capacity of the drainage system.

[0208] This implementation method converts the complex three-dimensional dynamic geological model into a clear water hazard risk probability by calculating key parameters and applying the water hazard probability calculation formula, providing an intuitive and quantitative decision-making basis for coal mine safety management, helping to timely identify high-risk situations and take preventive measures, effectively improving the accuracy and timeliness of water hazard warnings.

[0209] The embodiment of the present application also provides a coal mine disaster identification device based on a deep learning algorithm, comprising:

[0210] a memory configured to store instructions; and

[0211] The processor is configured to call instructions from the memory and implement the above-mentioned coal mine disaster identification method based on deep learning algorithm when executing the instructions.

[0212] The present application also provides an electronic device, including:

[0213] The above-mentioned coal mine disaster identification device based on deep learning algorithm.

[0214] An embodiment of the present application also provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute the above-mentioned coal mine disaster identification method based on deep learning algorithm.

[0215] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0216] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0217] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0218] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0219] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0220] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0221] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0222] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0223] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A coal mine disaster identification method based on deep learning algorithm, characterized in that: include: Obtain laser scanning point cloud data, geological drilling data and geophysical data of coal mines; Dividing the target area including the coal mine into a plurality of three-dimensional grid units, and interpolating the laser scanning point cloud data, the geological drilling data and the geophysical data into corresponding three-dimensional grid units to obtain a first three-dimensional geological model; Add rock layer parameters and coal layer parameters to the first three-dimensional geological model to obtain a second three-dimensional geological model; Associating the second three-dimensional geological model with a pre-constructed big data model, and obtaining the current season, and obtaining monitoring data of the current season according to the big data model, wherein the monitoring data includes weather parameters, water level data, and environmental data; Extract features of the current season, the weather parameters, the water level data, and the environmental data to obtain a multidimensional time series; Input the multidimensional time series into a pre-built deep learning model to obtain a water level prediction value for each time unit in a future preset time period; Inputting the water level prediction value of each time unit in a future preset time period into the big data model to perform data correction on the water level prediction value of each time unit in a future preset time period; The revised water level prediction value is spatially combined with the second three-dimensional geological model to obtain a dynamic three-dimensional geological model including water level changes in a future preset time period; Based on the dynamic three-dimensional geological model, the probability of water damage occurring in the coal mine is determined.

2. The method according to claim 1, characterized in that The target area including the coal mine is divided into a plurality of three-dimensional grid units, and the laser scanning point cloud data, the geological drilling data and the geophysical data are interpolated to the corresponding three-dimensional grid units to obtain a first three-dimensional geological model, including: Acquire the area of ​​the target area, and determine the density of the laser scanning point cloud data according to the laser scanning point cloud data; Determine the grid resolution according to the area of ​​the target area and the density of the laser scanning point cloud data; Dividing the target area into a plurality of three-dimensional grid units according to the grid resolution; In the case where any one of the three-dimensional grid units contains the laser scanning point cloud data, the geological drilling data and / or the geophysical data, the laser scanning point cloud data, the geological drilling data and / or the geophysical data are interpolated to the three-dimensional grid unit.

3. The method according to claim 2, characterized in that The step of determining the grid resolution according to the area of ​​the target region and the density of the laser scanning point cloud data comprises: Taking a region of a preset area as a basic unit, and dividing the target region into a plurality of sub-regions according to the basic unit; Calculate the average density and standard deviation of the laser scanning point cloud data in each of the sub-areas; A preset adaptive calculation formula is used to calculate the grid resolution of each sub-area according to the average density and the standard deviation, wherein when the grid resolution is less than a first preset value, the grid resolution is set to the first preset value, and when the grid resolution is greater than a second preset value, the grid resolution is set to the second preset value, and the second preset value is greater than the first preset value.

4. The method according to claim 1, characterized in that The step of associating the second three-dimensional geological model with the pre-constructed big data model, obtaining the current season, and obtaining monitoring data of the current season according to the big data model includes: Constructing an octree spatial index structure of the current season, wherein the octree spatial index structure includes a plurality of cubic subspaces and a plurality of nodes storing boundary coordinates and monitoring data; Calculating the center point coordinates of each of the three-dimensional grid cells in the second three-dimensional geological model; According to the boundary coordinates, a target cube subspace including the center point coordinates is queried in the octree space index structure, and a node corresponding to the center point coordinates is determined in the target cube subspace, and the monitoring data corresponding to the center point coordinates is the monitoring data stored in the node.

5. The method according to claim 1, characterized in that The deep learning model is a long short-term memory network, which includes an input layer, a first LSTM layer, a second LSTM layer, a third LSTM layer, a fully connected layer and an output layer. The multidimensional time series is input into the pre-built deep learning model to obtain the water level prediction value of each time unit in the future preset time period, including: The multidimensional time series is input through the input layer, and passes through the first LSTM layer, the second LSTM layer, the third LSTM layer and the fully connected layer respectively, and the water level prediction value of each time unit in the future preset time period is obtained through the output layer.

6. The method according to claim 5, characterized in that The step of inputting the predicted water level value of each time unit in the future preset time period into the big data model to perform data correction on the predicted water level value of each time unit in the future preset time period includes: Determine the seasonal index, water level upper limit, water level lower limit and water level trend of the current season according to the big data model; Smoothing the water level prediction value according to the seasonal index, the water level upper limit value and the water level lower limit value; According to the water level trend, the smoothed water level prediction value is trend corrected to obtain the water level prediction value after data correction.

7. The method according to claim 1, characterized in that The method of spatially combining the corrected water level prediction value with the second three-dimensional geological model to obtain a dynamic three-dimensional geological model including water level changes in a future preset time period includes: defining the boundary and initial water level of the water body in the second three-dimensional geological model; Calculate the water level elevation at each time point according to the revised water level prediction value; Using a preset bilinear difference interpolation algorithm, the water level surface elevation at each time point is diffused to a preset water level three-dimensional model to obtain a water level surface grid model; fusing the water level surface grid model with the second three-dimensional geological model to obtain a fused model; In the fusion model, a three-dimensional model of each time point is constructed, and based on the three-dimensional model of each time point, a dynamic three-dimensional geological model including water level changes in a future preset time period is obtained.

8. The method according to claim 7, characterized in that The determining the probability of water damage occurring in the coal mine based on the dynamic three-dimensional geological model includes: Calculating key parameters corresponding to the three-dimensional model at each time point, the key parameters including the volume proportion of the dangerous area, the water level rise rate and the carrying capacity of the drainage system; The probability of water damage occurring is calculated using a preset water damage calculation formula based on the key parameters.

9. A coal mine disaster identification device based on deep learning algorithm, characterized in that: include: a memory configured to store instructions; as well as A processor is configured to call the instructions from the memory and to implement the coal mine disaster identification method based on deep learning algorithm according to any one of claims 1 to 8 when executing the instructions.

10. An electronic device, characterized in that: include: A device according to claim 9.

Images