Modeling method, device and equipment for multi-source data fusion high-clean coal layer and medium

Abstract

The invention discloses a multi-source data fusion high-clean coal layer modeling method, device, equipment and medium, and relates to the technical field of geological modeling, a large number of high-precision virtual drill holes are extracted by utilizing stratum information recorded in construction records, automatic error correction is carried out according to a self-developed error correction algorithm, and the modeling efficiency is improved. And drill hole data distortion possibly caused by drill hole deviation is removed, so that the data reliability is improved, and the accuracy of tracks, coal-seeing points and coal-penetrating points is ensured. The gas drainage holes are drilled from the lower portion of the coal seam, so that the coal seeing points are used for restraining a coal seam bottom plate, the coal penetrating points are used for restraining a coal seam top plate, and meanwhile, the coal seam form is restrained in an auxiliary mode by combining the contour line of the coal seam bottom plate and a coal seam equal-thickness map; the high-precision coal seam model is constructed only according to various existing production materials under the condition that no extra equipment needs to be installed, and the requirements of a coal mine for a geological model in the aspects of actual production, geological early warning, emergency rescue and the like are met.

Description

Technical Field

[0001] The present application relates to the field of geological modeling technology, and particularly to a modeling method, device, equipment and medium for a high-precision coal seam with multi-source data fusion. Background Technique

[0002] With the rapid development of geological modeling technology, the modeling method based on multi-source data such as geophysical exploration, boreholes, and geology has become a key means to insight into the complex underground geological structure. The modeling method based on multi-source data such as geophysical exploration, boreholes, and geology uses a large amount of low-precision data to make up for the quality problems of boreholes, and can make the geological model close to the actual exposed situation in terms of morphology, but there is still room for improvement in accuracy.

[0003] In the related technology, CN114280676A discloses a method for detecting geological structures of borehole logging in coal mine gas drainage holes. A bedding model is generated by interpolating borehole data and geological section data. The borehole data comes from the original gas drainage holes. Borehole logging technology and logging equipment (such as apparent resistivity acquisition device, natural gamma device, etc.) are used to construct a borehole logging detection system for gas drainage holes in coal mines to obtain characteristic data of physical properties such as strata and coal seams in boreholes in coal mines. However, the applicant realizes that using logging equipment will increase costs and construction complexity. When borehole deviation occurs, it may cause logging data distortion, and the high-precision data collected is less and unevenly distributed and cannot cover the entire work area, which will lead to low accuracy and complex construction of the coal mine three-dimensional geological model, especially the coal seam model. Summary of the Invention

[0004] In view of this, the present application provides a modeling method, device, equipment and medium for a high-precision coal seam with multi-source data fusion, mainly aiming to solve the problems that using logging equipment will increase costs and construction complexity, when borehole deviation occurs, it may cause logging data distortion, and the high-precision data collected is less and unevenly distributed and cannot cover the entire work area, which will lead to low accuracy and complex construction of the coal mine three-dimensional geological model, especially the coal seam model.

[0005] According to the first aspect of the present application, a modeling method for a high-precision coal seam with multi-source data fusion is provided. The method includes:

[0006] Obtain geological-related data, generate ground borehole data by using the drilling comprehensive histogram in the geological-related data, and generate virtual borehole data by using the coal seam contour map, coal seam isopach map, and gas drainage hole construction acceptance record in the geological-related data;

[0007] Extract horizon data from the ground borehole data and the virtual borehole data, and based on the Kriging interpolation algorithm, construct a coal seam bedding model by using the horizon data;

[0008] Construct a fault model using the fault map in the geological-related data, and generate a formation framework grid using the surface borehole data, the virtual drilling data, and the fault model;

[0009] Perform a vertical framework grid construction operation based on the coal seam horizon model and the formation framework grid to obtain a 3D geological model.

[0010] According to a second aspect of the present application, there is provided a modeling device for a multi-source data fusion high-precision coal seam, the device including:

[0011] A data generation module, configured to obtain geological-related data, generate surface borehole data using the comprehensive drilling columnar chart in the geological-related data, and generate virtual drilling data using the coal seam contour map, the coal seam isopach map, and the gas drainage hole construction acceptance record in the geological-related data;

[0012] A first model construction module, configured to extract horizon data from the surface borehole data and the virtual drilling data, and construct a coal seam horizon model using the horizon data based on the Kriging interpolation algorithm;

[0013] A grid generation module, configured to construct a fault model using the fault map in the geological-related data, and generate a formation framework grid using the surface borehole data, the virtual drilling data, and the fault model;

[0014] A second model construction module, configured to perform a vertical framework grid construction operation based on the coal seam horizon model and the formation framework grid to obtain a 3D geological model.

[0015] According to a third aspect of the present application, there is provided a device including a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the method according to any one of the first aspects are implemented.

[0016] According to a fourth aspect of the present application, there is provided a medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method according to any one of the first aspects are implemented.

[0017] By means of the above technical solutions, the technical solutions provided by the embodiments of the present invention at least have the following advantages:

[0018] A modeling method, device, equipment and medium for a multi-source data fusion high-precision coal seam provided by the present application. The present application obtains geological-related data, generates surface borehole data using the comprehensive drilling columnar diagram in the geological-related data, generates virtual drilling data using the coal seam contour map, coal seam isopach map, and gas drainage hole construction acceptance record in the geological-related data, extracts horizon data from the surface borehole data and virtual drilling data, constructs a coal seam surface model using the horizon data based on the Kriging interpolation algorithm, constructs a fault model using the fault map in the geological-related data, generates a stratigraphic framework grid using the surface borehole data, virtual drilling data and fault model, and performs a vertical framework grid construction operation based on the coal seam surface model and the stratigraphic framework grid to obtain a three-dimensional geological model. By using the existing geological boreholes and the construction records of coal mine gas drainage holes, a large number of high-precision virtual boreholes are extracted using the stratigraphic information recorded in the construction records, and at the same time, automatic error correction is performed according to the self-developed error correction algorithm to remove the problem of data distortion caused by borehole offset, thereby improving data reliability and ensuring the accuracy of the trajectory, coal-seeing point, and coal-piercing point. Then, the trajectory, coal-seeing point, and coal-piercing point of each gas drainage hole are calculated. Since the gas drainage hole starts from below the coal seam, the coal-seeing point is used to constrain the coal seam floor, the coal-piercing point is used to constrain the coal seam roof, and at the same time, the coal seam floor contour and coal seam isopach map are combined to assist in constraining the coal seam morphology. Without the need for additional equipment, a high-precision coal seam model is constructed only based on the existing various production materials to meet the needs of the coal mine for geological models in actual production, geological warning, and emergency rescue, etc.

[0019] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specific embodiments of the present application are specifically exemplified. Brief Description of the Drawings

[0020] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present application. And throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0021] Figure 1 Shows a schematic flow chart of a method for modeling a multi-source data fusion high-precision coal seam provided by an embodiment of the present application;

[0022] Figure 2A Shows a schematic flow chart of another method for modeling a multi-source data fusion high-precision coal seam provided by an embodiment of the present application;

[0023] Figure 2BShows the differential change surface provided by the embodiment of the present application;

[0024] Figure 2C Shows the abnormal data points provided by the embodiment of the present application;

[0025] Figure 2D Shows the virtual drilling data provided by the embodiment of the present application;

[0026] Figure 2E Shows the three-dimensional geological model provided by the embodiment of the present application;

[0027] Figure 2F Shows the three-dimensional geological model grid provided by the embodiment of the present application;

[0028] Figure 2G Shows the embedded part of the three-dimensional geological model provided by the embodiment of the present application;

[0029] Figure 2H Shows the schematic flow chart of the multi-source data fusion high-precision coal seam modeling method based on coal mine gas drainage boreholes provided by the embodiment of the present application;

[0030] Figure 3 Shows the schematic structural diagram of the modeling of a multi-source data fusion high-precision coal seam provided by the embodiment of the present application;

[0031] Figure 4 Shows the schematic structural diagram of a device provided by the embodiment of the present application. Detailed implementation manners

[0032] Hereinafter, the exemplary embodiments of the present application will be described in more detail with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present application can be more thoroughly understood and the scope of the present application can be completely conveyed to those skilled in the art.

[0033] The embodiment of the present application provides a method for modeling a multi-source data fusion high-precision coal seam, as Figure 1 shown, the method includes:

[0034] 101. Obtain geological-related data, generate surface borehole data by using the comprehensive drilling columnar diagram in the geological-related data, and generate virtual drilling data by using the coal seam contour map, coal seam isopach map, and gas drainage hole construction acceptance record in the geological-related data.

[0035] In CN114280676A, a horizon model is generated by interpolating drilling data and geological section data. The drilling data comes from the original gas drainage holes. The drilling logging technology is used to construct a drilling logging detection system for coal mine underground gas drainage holes to detect the drilling logging and obtain the characteristic data of the physical properties of the strata and coal seams in the coal mine underground. However, additional logging equipment needs to be installed, such as apparent resistivity acquisition devices, natural gamma devices, etc., which will increase costs and construction complexity. Moreover, drilling deviation may lead to distortion of logging data.

[0036] To solve this problem, this application proposes a modeling method for a multi-source data fusion high-precision coal seam. It does not rely on new equipment and only uses existing surface drill holes and construction records. It can not only reduce costs but also improve technical feasibility. At the same time, a self-developed algorithm is used to automatically correct errors and remove abnormal data, improving data reliability and ensuring the accuracy of the trajectory, coal-seeing points, and coal-penetrating points. Finally, since the roof and floor data of the coal seam directly come from the drilling exposure data and do not go through the process of logging curve interpretation and conversion, it further avoids the errors caused by manual judgment and further reduces the human influence factors to achieve a highly stable modeling accuracy. The coal seam contour and coal seam isopach map are used to assist in constraining the coal seam morphology, making the model highly consistent with the actual exposure morphology, and having both economy and accuracy. At the same time, in some mining areas, the high-precision data is limited and the co-kriging algorithm cannot be used to correct the contour data. Therefore, this application uses the method of fitting the difference trend surface of the drilling contours to correct the contour elevation, so that the contours and geological drill holes will not conflict due to precision differences. The execution subject of this application can be a geological modeling system. The geological modeling system relies on the computing power of the server to provide services for users. The server can be an independent server or can provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery network (Content Delivery Network, CDN), and basic cloud computing servers such as big data and artificial intelligence platforms, so that the geological modeling system can extract a large number of high-precision virtual drillings as formation constraints by making full use of the ordinary production data generated during the production process, and fuse data such as drill holes, faults, formation exposure information, and geophysical exploration to realize the construction of a high-precision three-dimensional geological model.

[0037] In the embodiments of the present application, geological related data includes comprehensive drilling columnar diagrams, fault maps, coal seam contour maps, coal seam isopach maps, and gas drainage hole construction acceptance records. The geological modeling system generates surface borehole data using comprehensive drilling columnar diagrams, and generates virtual drilling data using coal seam contour maps, coal seam isopach maps, and gas drainage hole construction acceptance records. The virtual drilling data can make up for the deficiencies or omissions in the spatial distribution of actual drilling data. By reasonably utilizing the virtual drilling data, while reducing the actual drilling workload and cost, it can improve the accuracy and integrity of the overall understanding of the coal seam, thereby enhancing the data utilization efficiency. It can also better constrain the stratigraphic model and ensure that the geological model is more fitting and realistic with the geological body. The geological modeling system can achieve the construction of a high-precision coal seam model by making full use of existing resources, namely construction records and existing surface borehole data, without relying on additional hardware devices.

[0038] 102. Extract horizon data from surface borehole data and virtual drilling data, and construct a coal seam horizon model based on the Kriging interpolation algorithm.

[0039] In the embodiments of the present application, the geological modeling system extracts horizon data from the surface borehole data generated from surface borehole columnar diagrams and virtual drilling data, which can integrate actual measurement data and virtual data obtained through simulation and other means, expand the data source, make the data basis for model construction richer, and can more comprehensively reflect the geological characteristics of the coal seam. Based on the Kriging interpolation algorithm, a coal seam horizon model is constructed using the horizon data. The principle of the Kriging interpolation algorithm is based on the unbiased and minimum variance conditions in probability statistical estimation theory. During the interpolation process, the spatial correlation of the object to be described is considered, making the interpolation more scientific and closer to the actual situation, accurately reflecting the characteristics such as the shape, undulation, and change trend of the coal seam, providing a reliable basis for subsequent coal seam geological analysis, reserve calculation, and other work.

[0040] 103. Construct a fault model using the fault map in the geological related data, and generate a stratigraphic framework grid using surface borehole data, virtual drilling data, and the fault model.

[0041] In the embodiments of the present application, the geological modeling system constructs a fault model using the fault map in the geological related data. Through the fault map, the geometric shape of the fault and the mutual relationship between faults can be accurately reflected, and the distribution and extension of the fault in three-dimensional space can be accurately simulated, providing an accurate basis for studying geological structures and stratigraphic evolution. Then, the geological modeling system combines surface borehole data, virtual drilling data, and the fault model to generate a stratigraphic framework grid, which can fully consider the actual situation of the strata at the drilling location and the influence of faults on the strata such as faulting, making the generated stratigraphic framework grid more in line with the geological reality and improving the accuracy and reliability of the model.

[0042] 104. Perform a vertical framework grid construction operation based on the coal seam bedding model and the stratigraphic framework grid to obtain a 3D geological model.

[0043] In the embodiment of the present application, the coal seam bedding model can accurately reflect the characteristics of the coal seam such as its morphology and undulation, and the stratigraphic framework grid can represent the structure and distribution of the entire stratum. By combining the two to construct a vertical framework grid and then obtaining a 3D geological model, the geological modeling system can fully integrate the geological information at different levels, avoid the problems of information loss or inaccuracy that may exist in a single model, make the 3D geological model more accurate in depicting the underground geological structure, and be more in line with the actual geological situation.

[0044] The embodiment of the present application provides a modeling method for a multi-source data fusion high-precision coal seam. Compared with the prior art, in the embodiment of the present invention, by obtaining geological-related data, ground borehole data is generated using the comprehensive drilling columnar diagram in the geological-related data, virtual drilling data is generated using the coal seam contour map, coal seam isopach map, and gas drainage hole construction acceptance record in the geological-related data, horizon data is extracted from the ground borehole data and virtual drilling data, a coal seam bedding model is constructed using the horizon data based on the Kriging interpolation algorithm, a fault model is constructed using the fault map in the geological-related data, a stratigraphic framework grid is generated using the ground borehole data, virtual drilling data, and the fault model, and a vertical framework grid construction operation is performed based on the coal seam bedding model and the stratigraphic framework grid to obtain a 3D geological model. By using the existing geological boreholes and the construction records of coal mine gas drainage holes, a large number of high-precision virtual boreholes are extracted using the stratigraphic information recorded in the construction records, and at the same time, automatic error correction is performed according to the self-developed error correction algorithm to remove the problem of borehole data distortion caused by borehole offset, thereby improving data reliability and ensuring the accuracy of the trajectory, coal-seeing point, and coal-penetrating point. Then, the trajectory of each gas drainage hole borehole and the coal-seeing point and coal-penetrating point are calculated. Since the gas drainage hole borehole starts from below the coal seam, the coal-seeing point is used to constrain the coal seam floor, the coal-penetrating point is used to constrain the coal seam roof, and at the same time, the coal seam floor contour and the coal seam isopach map are combined to assist in constraining the coal seam morphology, and a high-precision coal seam model is constructed only based on the existing various production materials without the need to install additional equipment, meeting the needs of the coal mine for geological models in actual production, geological early warning, and emergency rescue.

[0045] Further, as a refinement and extension of the specific implementation manner of the above embodiment, in order to fully illustrate the specific implementation process of this embodiment, the embodiment of the present application provides another modeling method for a multi-source data fusion high-precision coal seam, as Figure 2A shown, this method includes:

[0046] 201. Generate a drilling well position file and a drilling horizon file according to the comprehensive drilling columnar diagram.

[0047] In the embodiment of the present application, the geological modeling system obtains a comprehensive drilling columnar chart, reads the drilling name, plane coordinates, bottom depth, and kelly bushing height of each well in the comprehensive drilling columnar chart, and generates a well location file for multiple wells according to a preset data format using the drilling name, plane coordinates, bottom depth, and kelly bushing height. For example, the drilling name, plane coordinates, bottom depth, and kelly bushing height in the comprehensive drilling columnar chart are processed in the data format of well name, X, Y, Bottom depth, KB to create a well location file a.txt. The well location data specifies the coordinates and depth of the borehole and is an important basis for geological structure analysis, stratigraphic correlation, etc., which helps to reveal the spatial variation law of the underground geological structure.

[0048] Next, the geological modeling system reads the top depth of each well at the target horizon in the comprehensive drilling columnar chart, and generates a well horizon file for multiple wells in tabular form using the top depth of multiple wells at the target horizon. Among them, according to the comprehensive drilling columnar chart, the strata are divided into six horizons: the bottom of the No. 21 coal seam (bottom), the top of the No. 21 coal seam (top), the top of the Upper Carboniferous Taiyuan Formation (C3t), the top of the Lower Permian Shanxi Formation (P1s), the top of the Lower Permian Lower Shihezi Formation (P1x), and the top of the Upper Permian Upper Shihezi Formation (P2s), and the target horizon is the No. 21 coal seam. For example, according to the comprehensive drilling columnar chart, the top depth of the No. 21 coal seam in each well is read, and the top depth of the No. 21 coal seam of each well is sorted in tabular form to generate a well horizon file b.txt.

[0049] 202. Generate a fault file according to the fault map.

[0050] In the embodiment of the present application, the geological modeling system obtains a fault map, corrects the coordinates of the fault map, calibrates each position in the picture to the actual coordinates, and obtains a corrected fault map. Next, the geological modeling system reads the fault dip angle and elevation range in the corrected fault map, and calculates the projection positions of the coal seam top line and the coal seam bottom line in the corrected fault map respectively using the fault dip angle and elevation range, fully considering the spatial geometric characteristics of the fault and the spatial relationship between the coal seam and the fault. Compared with the traditional method that relies solely on experience or simple estimation, this method of determining the projection position based on precise calculation can more accurately reflect the actual positional relationship between the coal seam and the fault in space, greatly improving the accuracy and reliability of geological information. Subsequently, the geological modeling system draws the fault top line and the fault bottom line in the corrected fault map based on the projection positions of the coal seam top line and the coal seam bottom line in the corrected fault map to obtain a target fault map, which can truly present the faulting situation of the coal seam by the fault.

[0051] Then, the geological modeling system performs vectorization processing on the target fault map, automatically tracks pixels of the processed target fault map by setting the background color and target color, obtains multiple first pixel point coordinates, filters the multiple first pixel point coordinates, that is, removes unnecessary pixel points except the head and tail on the straight line, and appropriately reduces the density of arc pixel points, which can ensure that the key features and overall shape of the fault are not damaged while streamlining the data. Finally, a fault file c.txt is generated using the filtered multiple first pixel point coordinates.

[0052] 203. Generate a coal seam contour map file and a coal seam thickness map file based on the coal seam contour map and the coal seam thickness map.

[0053] In the embodiment of the present application, the geological modeling system acquires the coal seam contour map and the coal seam thickness map, performs coordinate correction on the coal seam contour map and the coal seam thickness map to obtain a corrected coal seam contour map and a corrected coal seam thickness map. Then, the geological modeling system performs vectorization processing on the corrected coal seam contour map and the corrected coal seam thickness map, automatically tracks pixels of the processed corrected coal seam contour map and the processed corrected coal seam thickness map by setting the background color and target color, obtains multiple second pixel point coordinates, and filters the multiple second pixel point coordinates to obtain multiple target pixel point coordinates. Since the result after vectorization is coordinate data, in order to ensure the accuracy of the curve shape, a smaller point spacing needs to be set. However, using a smaller point spacing for the straight line is a waste of space. Therefore, unnecessary pixel points except the head and tail on the straight line are removed, and the density of small-curvature arc points is appropriately reduced to save computer resources.

[0054] Subsequently, the geological modeling system constructs a target grid based on the Kriging interpolation algorithm for the multiple target pixel point coordinates, and the obtained target grid can more accurately reflect the regional spatial characteristics. The ground borehole data is acquired, and the calculated elevation value, as well as the x coordinate and y coordinate of each ground borehole, are obtained by substituting the ground borehole data into the target grid. Then, the geological modeling system acquires the actual elevation value of each ground borehole. For each ground borehole, the difference between the calculated elevation value and the actual elevation value of the ground borehole is used as the z coordinate of the ground borehole.

[0055] In this way, the geological modeling system constructs a target three-dimensional grid using the x coordinate, y coordinate, and z coordinate of each ground borehole, and performs Kriging interpolation calculation on the target three-dimensional grid to obtain Figure 2B the differential change surface as shown, to visually display the distribution of the difference between the calculated elevation and the actual elevation. The multiple target pixel point coordinates are substituted into the differential change surface for calculation to obtain elevation data, and the elevation data is used to correct the multiple target pixel point coordinates, effectively correcting the deviation between the calculation result and the actual geological situation.

[0056] Finally, the geological modeling system extracts multiple contour coordinates corresponding to the corrected coal seam contour map from the corrected coordinates of multiple target pixel points, and generates a coal seam contour map file c.txt using the multiple contour coordinates. Multiple isopach coordinates corresponding to the corrected coal seam isopach map are extracted from the corrected coordinates of multiple target pixel points, and a coal seam isopach map file d.txt is generated using the multiple isopach coordinates. The coal seam contour map file and the coal seam isopach map file can display the geological characteristics of the coal seam from different dimensions. The contour map reflects the topographic undulations of the coal seam, and the isopach map shows the thickness changes of the coal seam, providing accurate data on the spatial shape and distribution law of the coal seam for subsequent construction of a three-dimensional geological model.

[0057] 204. Generate gas drainage hole data based on the construction acceptance records of gas drainage holes.

[0058] In the embodiment of the present application, the geological modeling system obtains the construction acceptance records of gas drainage holes. The construction acceptance records of gas drainage holes generated during the production process are a tabular file recording information such as borehole number, coal-seam intersection point, coal-seam penetration length, roof penetration length, hole depth, dip angle, deviation, vertical distance from the highest point of the roadway center line, and horizontal distance from the highest point of the roadway center line. It should be noted that all boreholes in the embodiment of the present application are located in the same column, so the azimuth angle is 0 relative to the origin and is not considered. If the boreholes of other projects are arranged disorderly, the influence of the azimuth angle needs to be considered additionally. Then, the geological modeling system automatically extracts relevant parameters from the construction acceptance records of gas drainage holes using an algorithm, and automatically calculates the actual coordinates and elevation values of each borehole according to trigonometric functions, and generates gas drainage hole data e.xlsx using the construction acceptance records of gas drainage holes, the actual coordinates and elevation values of each borehole.

[0059] 205. Correct the gas drainage hole data using the Isolation Forest algorithm and the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to obtain a corrected gas drainage hole file.

[0060] In the embodiment of the present application, the geological modeling system can automatically correct and remove abnormal data using a self-developed error correction algorithm to improve data reliability. The self-developed error correction algorithm is a combined determination using the Isolation Forest algorithm and the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm. The gas drainage hole data e.xlsx is input into the self-developed error correction model for combined determination of the Isolation Forest algorithm and the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm. At the same time, it will cooperate with the change of the curve slope and the borehole name for comprehensive research and judgment to finally obtain the abnormal point determination result. For example Figure 2CThe abnormal data points shown. The solid line represents the data-fitted coal seam, the dashed line represents the connection line between the coal-seeing points and coal-penetrating points of the gas drainage holes, and the position framed by the rectangle represents the abnormal point, i.e., point 3. Assume that the gas drainage hole data contains n points, and each point is represented by (x,y) coordinates: D = {(x 1 ,y 1 ),(x 2 ,y 2 ),(x 3 ,y 3 ),...,(x n ,y n )}, where (x 3 ,y 3 ) is point 3.

[0061] Specifically, first, the isolation forest algorithm is used for determination. The main feature of the isolation forest algorithm is to randomly build a tree structure to isolate data points, and the abnormality of points is determined by the path length. The shorter the path, the easier the point is to be isolated, that is, the more likely it is an abnormal point.

[0062] Extract multiple coordinate points from the gas drainage hole data, and obtain the isolation forest model included in the isolation forest algorithm. Among them, the isolation forest model is pre-trained with a sample data set and can learn the spatial distribution of the entire data set for abnormal point determination.

[0063] For each coordinate point, input the coordinate point into the isolation forest model, calculate the coordinate point based on the isolation forest model, and output the path length of the coordinate point. Based on the isolation forest model, calculate the anomaly score of the coordinate point using the path length of the coordinate point and the expectation of the path length.

[0064] For example, input point 3 (x 3 ,y 3 ) into the trained isolation forest model. In the isolation forest tree, the path length h(x 3 ) from the root node to point 3 is calculated. Then, the anomaly score s(x 3 ) is calculated through the path length and the normalization constant, and the calculation formula is as follows in Formula 1:

[0065] Formula 1:

[0066] Among them, s(x 3 ) is the anomaly score of point 3, h(x 3 ) is the path length of point 3, and c(N) is the expectation of the path length of point 3. If s(x 3 ) is close to 1, it means it is easier to be isolated and may be an abnormal point; if s(x 3 ) is close to 0, it means it is difficult to be isolated and may be a normal point.

[0067] For gas drainage hole data, Isolation Forest can quickly and efficiently identify those outliers that significantly deviate from the normal data distribution. When dealing with large-scale gas drainage hole data, it can complete the detection of outliers in a relatively short time, improving the efficiency of data processing. Moreover, the data distribution of gas drainage holes may not follow a simple statistical distribution and may be affected by multiple factors (such as geological conditions, the state of drainage equipment, etc.). Isolation Forest can better adapt to this complex data distribution and detect abnormal data caused by equipment failures, geological anomalies, etc.

[0068] Next, a density-based local outlier detection algorithm is determined. The core idea of the density-based local outlier detection algorithm is to judge whether a point is an outlier by comparing the local density of the point with the local density of its neighbors.

[0069] Among multiple coordinate points, multiple other coordinate points except the coordinate point are determined. The Euclidean distances between the coordinate point and the multiple other coordinate points are calculated respectively, and the multiple other coordinate points are sorted in ascending order of the Euclidean distances. The preset number of neighbors included in the density-based local outlier detection algorithm is obtained. Starting from the first other coordinate point among the sorted multiple other coordinate points, the other coordinate points with the preset number of neighbors are extracted, and each of the other coordinate points with the preset number of neighbors is used as a neighbor point to construct a set, obtaining a neighbor point set. The reachable distances between the coordinate point and each neighbor point in the neighbor point set are calculated, obtaining multiple reachable distances. The ratio of the preset number of neighbors to the sum of the multiple reachable distances is used as the local reachability density of the coordinate point. The local reachability density of each neighbor point is calculated, the product value of the preset number of neighbors and the local reachability density of the coordinate point, and the sum value of the local reachability densities of the multiple neighbor points are calculated, and the ratio of the product value to the sum value is used as the local outlier factor of the coordinate point.

[0070] For example, first, the k-neighborhood is determined. For point 3, the parameter k (the number of neighbors) is selected. The default parameter k = 20 provided by the LocalOutlierFactor class in sklearn performs well in this embodiment. Then, the Euclidean distances between point 3 and all other points are calculated, and the calculation formula is as follows in Formula 2:

[0071] Formula 2:

[0072] where distance(x 3 ,x i ) is the Euclidean distance between point 3 and point i, and the coordinates of point i are (x i ,y i)。Find the k nearest neighbor points to point 3, denoted as N k (x 3 )。

[0073] Subsequently, for point 3 and its neighbor point o ∈ N k (x 3 ), calculate the reachable distance between point 3 and each neighbor point respectively. The calculation formula is as follows: Formula 3

[0074] Formula 3: reach-dist k (x 3 , o) = max{k-distance(o), distance(x 3 , o)}

[0075] Among them, reach-dist k (x 3 , o) is the reachable distance from point 3 to neighbor point o, k-distance(o) is the k-distance of neighbor point o, which refers to the distance from neighbor point o to its k-th nearest neighbor point. When calculating the k-distance, it is necessary to calculate the distances from neighbor point o to all other points in the dataset and sort these distances, and select the k-th smallest distance as k-distance(o), distance(x 3 , o) is the Euclidean distance between point 3 and neighbor point o. If the distance between point 3 and neighbor point o is less than k-distance(o), then k-distance(o) is used as the replacement value to prevent outliers (even if they are very close to the neighbor in terms of Euclidean distance) from causing distortion in density calculation.

[0076] Then, calculate the local reachability density (LRD) of point 3. The calculation formula is as follows: Formula 4

[0077] Formula 4:

[0078] Among them, LRD k (x 3 ) is the local reachability density of point 3.

[0079] Calculate the local reachability density of each neighbor point, and calculate the local outlier factor (LOF) using the local reachability density of point 3 and the local reachability densities of multiple neighbor points. The calculation formula is as follows: Formula 5

[0080] Formula 5:

[0081] Among them, LOF k (x 3 ) is the local outlier factor of point 3, LRD k (x 3) is the local reachability density of point 3, LRD k (o) is the local reachability density of neighbor point o. If LOF k (x 3 ) > 1, then point 3 is isolated and may be an outlier.

[0082] For the gas drainage hole data, the gas drainage situation in different regions may have local differences due to factors such as geological structures. LOF can well capture this local anomaly. Moreover, in the gas drainage hole data, there may be data anomalies in the gas drainage holes in a certain local area, which may be due to local equipment failures or special changes in local geological conditions. LOF can accurately detect this local anomaly instead of only considering the global anomaly situation. In addition, through the visualized LOF values, the outlier points can be found more intuitively, and at the same time, the situation around the outlier points can be analyzed, providing a basis for further fault diagnosis and data correction.

[0083] In order to accurately conduct a comprehensive judgment based on the anomaly scores obtained by the isolation forest algorithm and the local anomaly factors obtained by the density-based local outlier detection algorithm, an anomaly score judgment threshold and a local anomaly factor judgment threshold are set. The anomaly score judgment threshold is used to judge the anomaly scores of coordinate points, and at the same time, the local anomaly factor judgment threshold is used to judge the local anomaly factors of coordinate points to obtain the judgment results of coordinate points.

[0084] It should be noted that the isolation forest algorithm allows setting an expected anomaly ratio contamination, such as 0.05, indicating that 5% of the data points are considered as outlier points. When training the model, the isolation forest algorithm will automatically determine the anomaly score judgment threshold based on this ratio. The isolation forest algorithm provides the parameter contamination in libraries such as sklearn for automatically determining the threshold, and this parameter is directly applied in this embodiment.

[0085] If the anomaly score of a coordinate point is greater than the anomaly score judgment threshold and the local anomaly factor of the coordinate point is greater than the local anomaly factor judgment threshold, it means that the judgment results of the two algorithms are consistent, and the coordinate point is an outlier point. Therefore, the outlier point judgment parameter is obtained and used as the judgment result of the coordinate point;

[0086] If the anomaly score of a coordinate point is less than or equal to the anomaly score judgment threshold and the local anomaly factor of the coordinate point is less than or equal to the local anomaly factor judgment threshold, it means that the judgment results of the two algorithms are consistent, and the coordinate point is a normal point. Therefore, the normal point judgment parameter is obtained and used as the judgment result of the coordinate point;

[0087] If the anomaly score of a coordinate point is less than or equal to the anomaly score determination threshold and the local anomaly factor of the coordinate point is greater than the local anomaly factor determination threshold, or the anomaly score of the coordinate point is greater than the anomaly score determination threshold and the local anomaly factor of the coordinate point is less than or equal to the local anomaly factor determination threshold, it indicates that the determination results of the two algorithms are inconsistent, and further determination needs to be combined with the change of the curve slope and the drilling name. Specifically, among multiple coordinate points, determine the first coordinate point and the second coordinate point adjacent to the coordinate point, calculate the slopes of the coordinate point with the first coordinate point and the second coordinate point respectively, obtain the first slope value and the second slope value. When the change rate of the first slope value and the second slope value is greater than the preset change rate, such as the slope suddenly becomes larger, obtain the anomaly point determination parameter, and use the anomaly point determination parameter as the determination result of the coordinate point.

[0088] For example, for a comprehensive study of Point 3, the Isolation Forest algorithm judges the path length by randomly dividing the tree, so the path of Point 3 is short and the anomaly score is high. The Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm compares the local density, and the density of Point 3 is significantly lower than that of the neighbor points, so it is abnormal. Therefore, the results of the two methods are consistent, and Point 3 is determined as an abnormal point and output.

[0089] Based on the above process, the geological modeling system uses the Isolation Forest algorithm and the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to perform determination calculations on each coordinate point, obtains the determination result of each coordinate point. The self-developed error correction large model does not directly modify the data, but adds the determination results of multiple coordinate points to the last column of the gas drainage hole data. 0 represents normal and 1 represents abnormal, and finally obtains the first gas drainage hole data.

[0090] Finally, the geological modeling system deletes the coordinate points with the determination result as the anomaly point determination parameter in the first gas drainage hole data, obtains the second gas drainage hole data, generates the target well position file and the target horizon file according to the preset data format using the second gas drainage hole data, and uses the target well position file and the target horizon file as the gas drainage hole correction file. In the gas drainage hole data, some anomalies may be global anomalies caused by major problems in the entire system, while some anomalies may be caused by local equipment or local geological conditions. Isolation Forest focuses on the detection of anomaly points from a global perspective, while DBSCAN focuses on the local perspective. The combination of the two can comprehensively consider global and local anomalies. Automatically correcting the gas drainage hole data can more accurately detect abnormal data, avoid unnecessary operations caused by misjudging normal data as abnormal, and at the same time prevent abnormal data from being omitted, ensuring the safe and efficient operation of the system.

[0091] It should be noted that the self-developed error correction algorithm of the embodiments of this application can be written in Python language, and the model is trained in advance using data. The self-developed error correction algorithm is packaged into an executable file and can be directly used without programming in the subsequent use process.

[0092] 206. Take the drilling well position file and the drilling horizon file as surface drilling data, and generate virtual drilling data by using the coal seam contour map file, the coal seam isopach map file, and the gas drainage hole correction file.

[0093] In the embodiments of this application, the geological modeling system takes the drilling well position file and the drilling horizon file as surface drilling data, and generates virtual drilling data by using the coal seam contour map file, the coal seam isopach map file, and the gas drainage hole correction file, so as to jointly model by using the coal mine gas drainage hole drill holes, geological drill holes, coal seam contours, and coal seam isopach maps in the subsequent process. The virtual drill holes, like the real drill holes, are drill holes with a single geological attribute after preprocessing and expert judgment using the actually exposed geological information, which can constrain the stratum model and ensure that the geological model is more fitting and realistic with the geological body. The virtual drill holes are directly written into the virtual drilling data in the same format as the real drill hole data. For the convenience of calculation, the Bottom depth value and the KB value are uniformly set. The virtual drilling data is as Figure 2D shown, the dot represents the virtual well of the coal seam roof, and the triangle point represents the surface drill hole. The virtual drilling can improve the stratum information.

[0094] 207. Extract the horizon data from the surface drilling data and the virtual drilling data, and construct the coal seam surface model based on the Kriging interpolation algorithm.

[0095] In the embodiments of this application, the geological modeling system extracts the horizon data from the surface drilling data and the virtual drilling data, and constructs the coal seam surface model by using the horizon data. Through the comprehensive utilization of the data, the distribution and changes of the coal seam in space can be better reflected, the information deviation caused by relying only on a single data source can be avoided, and the quality and reliability of the data can be improved.

[0096] For each horizon, extract multiple horizon coordinate points corresponding to the horizon from the horizon data, calculate the distance values and semi-variances between any two of the multiple horizon coordinate points, and obtain multiple distance values and multiple semi-variances. Obtain the exponential model included in the Kriging interpolation algorithm, and use the exponential model to perform function fitting on the multiple distance values and multiple semi-variances to obtain the objective function. Extract multiple target horizon coordinate points from the multiple horizon coordinate points, and use the objective function and the multiple target horizon coordinate points to construct the first semi-variance coefficient matrix equation, where the target horizon coordinate points are the horizon coordinate points with elevation values among the multiple horizon coordinate points. Extract multiple specified horizon coordinate points from the multiple horizon coordinate points, where the specified horizon coordinate points are the horizon coordinate points without elevation values among the multiple horizon coordinate points.

[0097] For each specified horizon coordinate point, use the objective function to calculate the semi-variances of the specified horizon coordinate point to the multiple target horizon coordinate points respectively, obtain multiple specified semi-variances, and use the multiple specified semi-variances and the objective function to construct the second semi-variance coefficient equation of the specified horizon coordinate point.

[0098] Use the first semi-variance coefficient matrix equation and the second semi-variance coefficient equation of each specified horizon coordinate point to construct the objective equation set, solve the objective equation set, and obtain the weighting coefficients. Extract the elevation values of each target horizon coordinate point from the horizon data, and use the weighting coefficients to perform weighted summation calculation on the elevation values of the multiple target horizon coordinate points to obtain the estimated values of the multiple specified horizon coordinate points.

[0099] Based on the above calculation process, according to the Kriging interpolation algorithm, use the horizon data to calculate each horizon respectively, obtain the elevation values of the multiple target horizon coordinate points corresponding to each horizon and the estimated values of the multiple specified horizon coordinate points, and use the elevation values of the multiple target horizon coordinate points corresponding to each horizon and the estimated values of the multiple specified horizon coordinate points to construct the coal seam floor model. When constructing the coal seam floor model, Kriging interpolation can reasonably estimate the coal seam information in the unknown area according to the spatial positions and attribute values of the known horizon data points, which is more in line with the actual spatial distribution characteristics of the coal seam, making the constructed model more accurately reflect the true shape and structure of the coal seam.

[0100] 208. Use the fault file to construct the fault model, and use the surface borehole data, virtual drilling data, and fault model to generate the formation framework grid.

[0101] In the embodiment of the present application, the fault is the projection of the actual fault on a certain plane. According to the dip angle and dip direction of the fault, the fault normal vector can be calculated, so as to calculate the fault plane equation, and then the coordinates of points on the plane can be obtained. Specifically, the geological modeling system extracts multiple fault coordinate points from the fault file, obtains the preset spacing, and performs digital processing on the multiple fault coordinate points according to the preset spacing to obtain multiple discrete line segments.

[0102] For each discrete line segment, obtain the endpoint coordinates of the discrete line segment to get the first endpoint coordinates and the second endpoint coordinates, and calculate the plane vector and the perpendicular vector corresponding to the discrete line segment using the first endpoint coordinates and the second endpoint coordinates. Then, calculate the target vector that is perpendicular to the discrete line segment and parallel to the fault plane using the plane vector and the perpendicular vector. For example, take any line segment AB with endpoint coordinate values of (x 0 , y 0 , z 0 ), (x 1 , y 1 , z 1 ). Calculate the plane vector v(a, b, 0) and the perpendicular vector v ⊥ (b, -a, 0), and calculate the vector v p (b, -a, tanβ) that is perpendicular to AB and parallel to the fault plane.

[0103] Calculate the product of the plane vector and the target vector to obtain the fault plane normal vector, as shown in Equation 6 below:

[0104] Equation 6:

[0105] where n is the fault plane normal vector.

[0106] Obtain the initial plane equation, and substitute the fault plane normal vector and the first endpoint coordinates into the initial plane equation to obtain the fault plane equation of the discrete line segment, as shown in Equation 7 below:

[0107] Equation 7: m(x - x0) + n(y - y0) + p(z - z0) = 0

[0108] Based on the above process, process each discrete line segment separately to obtain the fault plane equation of each discrete line segment. Obtain the elevation range in the fault file, and construct a fault model using the fault plane equation and the fault plane equations of each discrete line segment. By calculating the dip angle, dip direction of the fault and digitizing it into a discrete point string, the geometric shape of the fault can be accurately represented. The three-dimensional model constructed based on the plane equation can accurately reflect the position and shape of the fault in three-dimensional space, and accurately describe the extension range and spatial distribution of the fault from the perspective of spatial analytic geometry.

[0109] Then, based on the corner point grid model, the geological modeling system generates a formation framework grid using the drilling horizon file and the fault model in the geological related data. Among them, the corner point grid model is a widely used type of structured grid at present. The grid position can be defined by i, j, k. The length and width of the unit grid can be variable. The grid surface connecting the top and bottom grid points vertically can be inclined, and the grid can be distorted, which can conveniently simulate fault lines, boundaries or pinch-out lines.

[0110] Specifically, obtain the fault plane of the fault model, divide the geological body corresponding to the fault model into three planes in the elevation direction, determine the intersection points of the three planes and the fault plane to obtain the intermediate shape points. Obtain the preset coordinate direction and the preset number of grids corresponding to the preset coordinate direction, determine the model boundary according to the surface borehole data, virtual drilling data and the fault model, generate a planar quadrilateral grid using the intermediate shape points, the model boundary, the preset coordinate direction and the preset number of grids corresponding to the preset coordinate direction, obtain the fault strike, the fault top and the fault bottom in the fault model, and construct the planar quadrilateral grid to the fault top and the fault bottom according to the fault strike to obtain a columnar three-dimensional framework network connecting the fault top, the fault middle and the fault bottom, and use the columnar three-dimensional framework network as the formation framework network. Since the length and width of the unit grid of the grid are variable, and the grid surface can be inclined and distorted, it can more realistically represent the shape of the formation in space, especially suitable for areas with complex geological structures. Moreover, for geological bodies with different degrees of complexity, appropriate grids can be generated by adjusting the parameters of the grid, which has strong adaptability.

[0111] 209. Perform a vertical framework grid construction operation based on the coal seam bedding model and the formation framework grid to obtain a three-dimensional geological model.

[0112] In the embodiment of the present application, the geological modeling system imports the coal seam bedding model into the formation framework network, and the intersection points between all grids and the bedding become nodes of the three-dimensional grid, thus obtaining multiple nodes. Obtain the preset grid unit, use the preset grid unit to perform grid construction on each node in the vertical direction to obtain a three-dimensional grid, and adjust the three-dimensional grid using the drilling horizon file to obtain a specified three-dimensional grid. Specifically, more refined vertical grids can be constructed according to the formation thickness and model accuracy, that is, divide into multiple vertical grids between each formation group and optimize and adjust the model using the drilling horizon points. Then fill the specified three-dimensional grid to obtain a three-dimensional geological model.

[0113] After the three-dimensional geological model is generated, affected by the software algorithm used and the complexity of the geological conditions, unreasonable phenomena may occur in the generated high-precision model. It is necessary to use the model editing function of the modeling software to adjust the morphology of individual positions of the model according to actual needs to make the geological model reasonable and consistent with the actual situation. At the same time, the display mode of the model can be modified to meet the system requirements. The three-dimensional high-precision coal seam model is as Figure 2E shown. Figure 2F is the three-dimensional geological model grid. Figure 2G is the embedded part of the three-dimensional geological model.

[0114] In summary, the schematic flow chart of a multi-source data fusion high-precision coal seam modeling method based on coal mine gas drainage boreholes proposed in the embodiment of the present application is as follows:

[0115] AsFigure 2H As shown, modeling is carried out jointly using coal mine gas drainage boreholes, surface boreholes, coal seam isopachs, and coal seam isopach maps. The data of coal mine gas drainage boreholes is derived from the automatic processing of construction records. A large number of high-precision virtual boreholes are extracted using the stratigraphic information recorded in the construction records. At the same time, an in-house error correction algorithm is used for automatic error correction to remove the distortion of borehole data caused by borehole deviation, improving data reliability and ensuring the accuracy of the trajectory, coal-seeing points, and coal-penetrating points. Since the roof and floor data of the coal seam directly come from the borehole exposure data without going through the process of well logging curve interpretation and conversion, it further avoids the errors caused by manual judgment, further reduces the human influence factors, and achieves a highly stable modeling accuracy. The coal seam isopachs and coal seam isopach maps are used to assist in constraining the coal seam morphology, making the model highly consistent with the actual exposure morphology, and having both economy and accuracy. At the same time, due to the limited high-precision data in some mining areas and the inability to use the co-kriging algorithm to correct the isopach data, the method of fitting the difference trend surface of the borehole isopachs is adopted to correct the isopach elevation, so that the isopachs and geological boreholes will not conflict due to accuracy differences, and thus a high-precision coal seam model can be constructed based on the existing production data without the need for additional equipment.

[0116] The embodiment of the present application provides a modeling method for a high-precision coal seam with multi-source data fusion. Compared with the prior art, in the embodiment of the present invention, by obtaining geological-related data, the surface borehole data is generated using the comprehensive drilling columnar diagram in the geological-related data, and the virtual drilling data is generated using the coal seam isopach map, coal seam isopach map, and gas drainage borehole construction acceptance record in the geological-related data. The horizon data is extracted from the surface borehole data and the virtual drilling data. Based on the kriging interpolation algorithm, the coal seam horizon model is constructed using the horizon data, and the fault model is constructed using the fault map in the geological-related data. The formation framework grid is generated using the surface borehole data, virtual drilling data, and fault model. The vertical framework grid construction operation is carried out based on the coal seam horizon model and the formation framework grid to obtain a three-dimensional geological model. The construction records of the existing geological boreholes and coal mine gas drainage boreholes are used, and a large number of high-precision virtual boreholes are extracted using the stratigraphic information recorded in the construction records. At the same time, automatic error correction is carried out according to the in-house error correction algorithm to remove the problem of distortion of borehole data caused by borehole deviation, thereby improving data reliability and then ensuring the accuracy of the trajectory, coal-seeing points, and coal-penetrating points. Then, the trajectory, coal-seeing points, and coal-penetrating points of each gas drainage borehole are calculated. Since the gas drainage borehole starts from below the coal seam, the coal-seeing points are used to constrain the coal seam floor, and the coal-penetrating points are used to constrain the coal seam roof. At the same time, the coal seam floor isopachs and coal seam isopach maps are combined to assist in constraining the coal seam morphology, and a high-precision coal seam model is constructed based on the existing production data without the need for additional equipment, meeting the needs of the coal mine for geological models in actual production, geological warning, and emergency rescue.

[0117] Further, as Figure 1 a specific implementation of the method, an embodiment of the present application provides a modeling device for a high-precision multi-source data fusion coal seam, as Figure 3 shown, the device includes: a data generation module 301, a first model construction module 302, a grid generation module 303, and a second model construction module 304.

[0118] The data generation module 301 is configured to obtain geological-related data, generate surface borehole data by using the comprehensive drilling columnar diagram in the geological-related data, and generate virtual drilling data by using the coal seam contour map, the coal seam isopach map, and the gas drainage hole construction acceptance record in the geological-related data;

[0119] The first model construction module 302 is configured to extract horizon data from the surface borehole data and the virtual drilling data, and construct a coal seam surface model by using the horizon data based on the Kriging interpolation algorithm;

[0120] The grid generation module 303 is configured to construct a fault model by using the fault map in the geological-related data, and generate a formation framework grid by using the surface borehole data, the virtual drilling data, and the fault model;

[0121] The second model construction module 304 is configured to perform a vertical framework grid construction operation based on the coal seam surface model and the formation framework grid to obtain a three-dimensional geological model.

[0122] In a specific application scenario, the data generation module 301 is configured to obtain the comprehensive drilling columnar chart from the geological-related data, read the drilling name, plane coordinates, bottom depth, and kelly bushing height of each well from the comprehensive drilling columnar chart, and generate a well location file for multiple wells in a preset data format using the drilling name, plane coordinates, bottom depth, and kelly bushing height; read the top depth of each well at the target horizon from the comprehensive drilling columnar chart, and generate a drilling horizon file in tabular form using the top depths of multiple wells at the target horizon; use the well location file and the drilling horizon file as the surface borehole data; obtain the coal seam contour map and the coal seam isopach map from the geological-related data, perform coordinate correction on the coal seam contour map and the coal seam isopach map to obtain a corrected coal seam contour map and a corrected coal seam isopach map; perform vectorization processing on the corrected coal seam contour map and the corrected coal seam isopach map, perform pixel automatic tracking on the processed corrected coal seam contour map and the processed corrected coal seam isopach map to obtain multiple second pixel point coordinates, filter the multiple second pixel point coordinates to obtain multiple target pixel point coordinates; construct a target grid based on the Kriging interpolation algorithm for the multiple target pixel point coordinates, obtain the surface borehole data, substitute the surface borehole data into the target grid to calculate the calculated elevation value of each surface borehole, as well as the x coordinate and y coordinate of each surface borehole; obtain the actual elevation value of each surface borehole, and for each surface borehole, use the difference between the calculated elevation value of the surface borehole and the actual elevation value of the surface borehole as the z coordinate of the surface borehole; construct a target three-dimensional grid using the x coordinate, y coordinate, and z coordinate of each surface borehole, perform Kriging interpolation calculation on the target three-dimensional grid to obtain a differential change surface; substitute the multiple target pixel point coordinates into the differential change surface for calculation to obtain elevation data, use the elevation data to correct the multiple target pixel point coordinates, extract multiple contour coordinates corresponding to the corrected coal seam contour map from the corrected multiple target pixel point coordinates, and generate the coal seam contour map file using the multiple contour coordinates; extract multiple isopach coordinates corresponding to the corrected coal seam isopach map from the corrected multiple target pixel point coordinates, and generate the coal seam isopach map file using the multiple isopach coordinates; obtain the construction acceptance record of the gas drainage holes from the geological-related data, calculate the actual coordinates and elevation values of each hole according to the construction acceptance record of the gas drainage holes, and generate the gas drainage hole data using the construction acceptance record of the gas drainage holes, the actual coordinates and elevation values of each hole; obtain the Isolation Forest algorithm and the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm, and use the Isolation Forest algorithm and the DBSCAN algorithm to correct the gas drainage hole data to obtain the corrected gas drainage hole file;Generate the virtual drilling data by using the coal seam contour map file, the coal seam isopach map file, and the gas drainage hole correction file.

[0123] In a specific application scenario, the data generation module 301 is configured to extract multiple coordinate points from the gas drainage hole data, and obtain the isolation forest model included in the isolation forest algorithm; for each of the coordinate points, input the coordinate point into the isolation forest model, calculate the coordinate point based on the isolation forest model, and output the path length of the coordinate point; based on the isolation forest model, calculate the anomaly score of the coordinate point by using the path length of the coordinate point and the expectation of the path length; determine multiple other coordinate points except the coordinate point among the multiple coordinate points, calculate the Euclidean distance between the coordinate point and the multiple other coordinate points respectively, and sort the multiple other coordinate points in ascending order of the Euclidean distance; obtain the preset number of neighbors included in the density-based local outlier detection algorithm, start from the first other coordinate point among the sorted multiple other coordinate points, extract the other coordinate points with the preset number of neighbors, and construct a set with each of the other coordinate points with the preset number of neighbors as a neighbor point to obtain a neighbor point set; calculate the reachable distance between the coordinate point and each neighbor point in the neighbor point set to obtain multiple reachable distances, and use the ratio of the preset number of neighbors to the sum of the multiple reachable distances as the local reachability density of the coordinate point; calculate the local reachability density of each neighbor point, calculate the product value of the preset number of neighbors and the local reachability density of the coordinate point, and the sum value of the local reachability densities of multiple neighbor points, and use the ratio of the product value to the sum value as the local outlier factor of the coordinate point; obtain the anomaly score determination threshold and the local outlier factor determination threshold, use the anomaly score determination threshold to determine the anomaly score of the coordinate point, and at the same time use the local outlier factor determination threshold to determine the local outlier factor of the coordinate point to obtain the determination result of the coordinate point; use the isolation forest algorithm and the density-based local outlier detection algorithm to perform determination calculations on each coordinate point to obtain the determination result of each coordinate point, add the determination results of the multiple coordinate points to the gas drainage hole data to obtain the first gas drainage hole data; delete the coordinate points with the determination result of the outlier determination parameter in the first gas drainage hole data to obtain the second gas drainage hole data, generate a target well position file and a target horizon file by using the second gas drainage hole data in a preset data format, and use the target well position file and the target horizon file as the gas drainage hole correction file.

[0124] In a specific application scenario, the data generation module 301 is configured to: if the anomaly score of the coordinate point is greater than the anomaly score determination threshold and the local anomaly factor of the coordinate point is greater than the local anomaly factor determination threshold, obtain the anomaly point determination parameter and use the anomaly point determination parameter as the determination result of the coordinate point; if the anomaly score of the coordinate point is less than or equal to the anomaly score determination threshold and the local anomaly factor of the coordinate point is less than or equal to the local anomaly factor determination threshold, obtain the normal point determination parameter and use the normal point determination parameter as the determination result of the coordinate point; if the anomaly score of the coordinate point is less than or equal to the anomaly score determination threshold and the local anomaly factor of the coordinate point is greater than the local anomaly factor determination threshold, or the anomaly score of the coordinate point is greater than the anomaly score determination threshold and the local anomaly factor of the coordinate point is less than or equal to the local anomaly factor determination threshold, determine the first coordinate point and the second coordinate point adjacent to the coordinate point among the multiple coordinate points, calculate the slopes of the coordinate point with the first coordinate point and the second coordinate point respectively to obtain the first slope value and the second slope value, and when the change rate of the first slope value and the second slope value is greater than the preset change rate, obtain the anomaly point determination parameter and use the anomaly point determination parameter as the determination result of the coordinate point.

[0125] In a specific application scenario, the first model construction module 302 is configured to, for each horizon, extract multiple horizon coordinate points corresponding to the horizon from the horizon data, calculate the distance values and semi-variances between any two of the multiple horizon coordinate points, and obtain multiple distance values and multiple semi-variances; acquire the exponential model included in the Kriging interpolation algorithm, use the exponential model to perform function fitting on the multiple distance values and the multiple semi-variances, and obtain an objective function; extract multiple target horizon coordinate points from the multiple horizon coordinate points, and use the objective function and the multiple target horizon coordinate points to construct a first semi-variance coefficient matrix equation, where the target horizon coordinate points are the horizon coordinate points with elevation values among the multiple horizon coordinate points; extract multiple specified horizon coordinate points from the multiple horizon coordinate points, where the specified horizon coordinate points are the horizon coordinate points without elevation values among the multiple horizon coordinate points; for each of the specified horizon coordinate points, use the objective function to calculate the semi-variances of the specified horizon coordinate points to the multiple target horizon coordinate points respectively, obtain multiple specified semi-variances, and use the multiple specified semi-variances and the objective function to construct a second semi-variance coefficient equation for the specified horizon coordinate points; use the first semi-variance coefficient matrix equation and the second semi-variance coefficient equation for each of the specified horizon coordinate points to construct an objective equation set, solve the objective equation set, and obtain the weighting coefficients; extract the elevation values of each of the target horizon coordinate points from the horizon data, and use the weighting coefficients to perform weighted summation calculation on the elevation values of the multiple target horizon coordinate points to obtain the estimated values of the multiple specified horizon coordinate points; based on the Kriging interpolation algorithm, use the horizon data to calculate each horizon respectively, obtain the elevation values of the multiple target horizon coordinate points and the estimated values of the multiple specified horizon coordinate points corresponding to each horizon, and use the elevation values of the multiple target horizon coordinate points and the estimated values of the multiple specified horizon coordinate points corresponding to each horizon to construct the coal seam surface model.

[0126] In a specific application scenario, the grid generation module 303 is configured to obtain the fault map from the geological-related data, correct the coordinates of the fault map to obtain a corrected fault map; read the fault dip angle and elevation range in the corrected fault map, calculate the projection positions of the coal seam top line and the coal seam bottom line in the corrected fault map respectively by using the fault dip angle and the elevation range, draw the fault top line and the fault bottom line in the corrected fault map based on the projection positions of the coal seam top line and the coal seam bottom line in the corrected fault map to obtain a target fault map; perform vectorization processing on the target fault map, automatically track pixels of the processed target fault map to obtain a plurality of first pixel point coordinates, filter the plurality of first pixel point coordinates, and generate the fault file by using the filtered plurality of first pixel point coordinates; extract a plurality of fault coordinate points from the fault file, obtain a preset spacing, perform digitization processing on the plurality of fault coordinate points according to the preset spacing to obtain a plurality of discrete line segments; for each of the discrete line segments, obtain the endpoint coordinates of the discrete line segment to obtain a first endpoint coordinate and a second endpoint coordinate, calculate the plane vector and the perpendicular vector corresponding to the discrete line segment by using the first endpoint coordinate and the second endpoint coordinate, and calculate a target vector perpendicular to the discrete line segment and parallel to the fault plane by using the plane vector and the perpendicular vector; calculate the product of the plane vector and the target vector to obtain the fault plane normal vector, obtain the initial plane equation, and substitute the fault plane normal vector and the first endpoint coordinate into the initial plane equation to obtain the fault plane equation of the discrete line segment; process each of the discrete line segments respectively to obtain the fault plane equation of each of the discrete line segments, obtain the elevation range in the fault file, and construct the fault model by using the fault plane equation and the fault plane equations of each of the discrete line segments; obtain the fault plane of the fault model, divide the geological body corresponding to the fault model into three planes in the elevation direction, determine the intersection points of the three planes and the fault plane to obtain intermediate shape points; obtain a preset coordinate direction and the preset number of grids corresponding to the preset coordinate direction, determine the model boundary according to the surface borehole data, the virtual drilling data and the fault model, and generate a planar quadrilateral grid by using the intermediate shape points, the model boundary, the preset coordinate direction and the preset number of grids corresponding to the preset coordinate direction; obtain the fault strike, the fault top and the fault bottom in the fault model, construct the planar quadrilateral grid to the fault top and the fault bottom according to the fault strike to obtain a columnar three-dimensional framework network connecting the fault top, the fault middle and the fault bottom, and use the columnar three-dimensional framework network as the formation framework network.

[0127] In a specific application scenario, the second model construction module 304 is configured to import the coal seam bedding model into the formation framework network to obtain a plurality of nodes, acquire a preset grid unit, use the preset grid unit to construct grids for each of the nodes in the vertical direction to obtain a three-dimensional grid; adjust the three-dimensional grid using the drilling horizon file to obtain a specified three-dimensional grid, and fill the specified three-dimensional grid to obtain the three-dimensional geological model.

[0128] The embodiment of the present application provides a modeling device for high-precision coal seams with multi-source data fusion. Compared with the prior art, in the embodiment of the present invention, by acquiring geological-related data, ground borehole data is generated using the comprehensive drilling columnar diagram in the geological-related data, virtual drilling data is generated using the coal seam contour map, coal seam isopach map, and gas drainage hole construction acceptance records in the geological-related data, horizon data is extracted from the ground borehole data and virtual drilling data, based on the Kriging interpolation algorithm, the coal seam bedding model is constructed using the horizon data, the fault model is constructed using the fault map in the geological-related data, the formation framework grid is generated using the ground borehole data, virtual drilling data, and the fault model, and a vertical framework grid construction operation is performed based on the coal seam bedding model and the formation framework grid to obtain the three-dimensional geological model. Using the existing geological boreholes and the construction records of coal mine gas drainage holes, a large number of high-precision virtual boreholes are extracted using the formation information recorded in the construction records, and at the same time, automatic error correction is performed according to the self-developed error correction algorithm to remove the problem of data distortion caused by borehole offset, thereby improving data reliability and ensuring the accuracy of the trajectory, coal-seeing point, and coal-piercing point. Then, the trajectory of each gas drainage hole borehole, as well as the coal-seeing point and coal-piercing point, are calculated. Since the gas drainage hole borehole starts from below the coal seam, the coal-seeing point is used to constrain the coal seam floor, the coal-piercing point is used to constrain the coal seam roof, and at the same time, the coal seam floor contour and coal seam isopach map are combined to assist in constraining the coal seam morphology. Without the need for additional equipment installation, a high-precision coal seam model is constructed only based on the existing various production materials to meet the needs of the coal mine for geological models in actual production, geological warning, and emergency rescue.

[0129] It should be noted that for other corresponding descriptions of each functional unit provided in the embodiment of the present application for a modeling device for high-precision coal seams with multi-source data fusion, reference can be made to Figure 1 and Figures 2A to 2H the corresponding descriptions therein, which will not be elaborated herein.

[0130] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data that have been authorized by the user or fully authorized by all parties.

[0131] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0132] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

[0133] In an exemplary embodiment, referring to Figure 4 , a device is further provided. The device includes a bus, a processor, a memory, and a communication interface, and may further include an input / output interface and a display device. Among them, each functional unit can complete mutual communication through the bus. The memory stores a computer program, and the processor is used to execute the program stored on the memory to execute the multi-source data fusion high-precision coal seam modeling method in the above embodiments.

[0134] A medium stores a computer program, and when the computer program is executed by a processor, the steps of the multi-source data fusion high-precision coal seam modeling method are implemented.

[0135] Through the description of the above implementation manners, those skilled in the art can clearly understand that the present application can be implemented by hardware or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product. The software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation scenarios of the present application.

[0136] Those skilled in the art can understand that the drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the drawings are not necessarily essential for implementing the present application.

[0137] Those skilled in the art can understand that the modules in the device in the implementation scenario can be distributed in the device in the implementation scenario according to the description of the implementation scenario, or can be correspondingly changed to be located in one or more devices different from the present implementation scenario. The modules in the above implementation scenario can be combined into one module, or further split into multiple sub-modules.

[0138] The above serial numbers of this application are only for description and do not represent the advantages or disadvantages of the implementation scenarios.

[0139] The above are only several specific implementation scenarios of this application. However, this application is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the protection scope of this application.

Claims

1. A modeling method for high-precision coal seams by fusion of multi-source data, characterized in that: include: Acquire geological related data, use the comprehensive drilling column chart in the geological related data to generate ground drilling data, and use the coal seam contour map, coal seam equal thickness map, and gas drainage hole construction acceptance record in the geological related data to generate virtual drilling data; Extracting layer data from the ground drilling data and the virtual drilling data, and constructing a coal seam face model using the layer data based on a Kriging interpolation algorithm; constructing a fault model using the fault map in the geological related data, and generating a stratigraphic framework grid using the ground drilling data, the virtual drilling data and the fault model; A vertical frame grid construction operation is performed based on the coal seam face model and the stratum frame grid to obtain a three-dimensional geological model.

2. The method according to claim 1, characterized in that: The method of using the comprehensive drilling column chart in the geological related data to generate ground drilling data, and using the coal seam contour map, coal seam thickness map, and gas drainage hole construction acceptance record in the geological related data to generate virtual drilling data includes: The drilling comprehensive histogram is obtained from the geological related data, the drilling name, plane coordinates, bottom depth, and core height of each well are read from the drilling comprehensive histogram, and the drilling names, plane coordinates, bottom depths, and core heights of multiple wells are used to generate a drilling well location file according to a preset data format; Reading the top depth of each well in the target layer in the comprehensive drilling histogram, and generating a drilling layer file using the top depths of the multiple wells in the target layer in a table format; Using the drilling well location file and the drilling layer location file as the ground drilling data; Acquire the coal seam contour map and the coal seam isopythickness map from the geological related data, perform coordinate correction on the coal seam contour map and the coal seam isopythickness map, and obtain a corrected coal seam contour map and a corrected coal seam isopythickness map; Performing vector processing on the corrected coal seam contour map and the corrected coal seam isopythickness map, performing automatic pixel tracking on the processed corrected coal seam contour map and the processed corrected coal seam isopythickness map to obtain a plurality of second pixel point coordinates, and filtering the plurality of second pixel point coordinates to obtain a plurality of target pixel point coordinates; Based on the Kriging interpolation algorithm, the coordinates of the multiple target pixel points are grid-constructed to obtain a target grid, and ground drilling data is obtained. The ground drilling data is substituted into the target grid to calculate the calculated elevation value of each ground drilling hole, as well as the x-coordinate and y-coordinate of each ground drilling hole; Acquire the actual elevation value of each ground borehole, and for each ground borehole, use the difference between the calculated elevation value of the ground borehole and the actual elevation value of the ground borehole as the z coordinate of the ground borehole; Using the x-coordinate, y-coordinate and z-coordinate of each ground borehole to construct a target three-dimensional grid, performing Kriging interpolation calculation on the target three-dimensional grid to obtain a difference change surface; Substituting the coordinates of the multiple target pixel points into the difference change surface for calculation to obtain elevation data, using the elevation data to correct the coordinates of the multiple target pixel points, extracting the multiple contour line coordinates corresponding to the corrected coal seam contour map from the corrected multiple target pixel point coordinates, and using the multiple contour line coordinates to generate the coal seam contour map file; Extracting a plurality of isopach coordinates corresponding to the corrected coal seam isopach map from the corrected plurality of target pixel point coordinates, and generating the coal seam isopach map file using the plurality of isopach coordinates; Acquire the gas drainage hole construction acceptance record from the geological related data, calculate the actual coordinates and elevation value of each borehole according to the gas drainage hole construction acceptance record, and generate the gas drainage hole data using the gas drainage hole construction acceptance record, the actual coordinates and elevation value of each borehole; Acquire an isolation forest algorithm and a density-based local anomaly detection algorithm, and use the isolation forest algorithm and the density-based local anomaly detection algorithm to correct the gas drainage hole data to obtain the gas drainage hole correction file; The virtual drilling data is generated by using the coal seam contour map file, the coal seam equal thickness map file, and the gas drainage hole correction file.

3. The method according to claim 2, characterized in that The gas drainage hole data is corrected by using the isolation forest algorithm and the density-based local anomaly detection algorithm to obtain the gas drainage hole correction file, including: Extracting a plurality of coordinate points from the gas drainage hole data to obtain an isolation forest model included in the isolation forest algorithm; For each of the coordinate points, input the coordinate point into the isolation forest model, calculate the coordinate point based on the isolation forest model, and output the path length of the coordinate point; Based on the isolation forest model, the anomaly score of the coordinate point is calculated by using the path length of the coordinate point and the expectation of the path length; Determine a plurality of other coordinate points other than the coordinate point among the plurality of coordinate points, calculate the Euclidean distances between the coordinate point and the plurality of other coordinate points respectively, and sort the plurality of other coordinate points in ascending order of the Euclidean distances; Obtaining a preset number of neighbors included in the density-based local anomaly detection algorithm, extracting the preset number of other coordinate points from the first other coordinate point among the sorted multiple other coordinate points, taking each of the preset number of other coordinate points as a neighbor point and constructing a set to obtain a neighbor point set; Calculate the reachable distance between the coordinate point and each neighbor point in the neighbor point set to obtain multiple reachable distances, and use the ratio of the preset number of neighbors to the sum of the multiple reachable distances as the local reachable density of the coordinate point; Calculating the local reachable density of each of the neighboring points, calculating the product of the preset number of neighbors and the local reachable density of the coordinate point, and the sum of the local reachable densities of multiple neighboring points, and taking the ratio of the product to the sum as the local abnormality factor of the coordinate point; Obtaining an anomaly score determination threshold and a local anomaly factor determination threshold, using the anomaly score determination threshold to determine the anomaly score of the coordinate point, and using the local anomaly factor determination threshold to determine the local anomaly factor of the coordinate point, to obtain a determination result of the coordinate point; Using the isolation forest algorithm and the density-based local anomaly detection algorithm to perform determination calculation on each of the coordinate points, obtaining a determination result of each of the coordinate points, and adding the determination results of the multiple coordinate points to the gas drainage hole data to obtain first gas drainage hole data; In the first gas extraction hole data, the coordinate points whose judgment results are abnormal point judgment parameters are deleted to obtain second gas extraction hole data, and the second gas extraction hole data is used to generate a target well position file and a target layer file according to a preset data format, and the target well position file and the target layer file are used as the gas extraction hole correction file.

4. The method according to claim 3, characterized in that The abnormal score determination threshold is used to determine the abnormal score of the coordinate point, and the local abnormal factor determination threshold is used to determine the local abnormal factor of the coordinate point to obtain the determination result of the coordinate point, including: If the abnormality score of the coordinate point is greater than the abnormality score determination threshold, and the local abnormality factor of the coordinate point is greater than the local abnormality factor determination threshold, then obtaining the abnormal point determination parameter, and using the abnormal point determination parameter as the determination result of the coordinate point; If the abnormal score of the coordinate point is less than or equal to the abnormal score determination threshold, and the local abnormal factor of the coordinate point is less than or equal to the local abnormal factor determination threshold, then obtaining a normal point determination parameter, and using the normal point determination parameter as the determination result of the coordinate point; If the anomaly score of the coordinate point is less than or equal to the anomaly score determination threshold and the local anomaly factor of the coordinate point is greater than the local anomaly factor determination threshold, or the anomaly score of the coordinate point is greater than the anomaly score determination threshold and the local anomaly factor of the coordinate point is less than or equal to the local anomaly factor determination threshold, then a first coordinate point and a second coordinate point adjacent to the coordinate point are determined from the multiple coordinate points, and the slopes of the coordinate point and the first coordinate point and the second coordinate point are calculated respectively to obtain a first slope value and a second slope value. When the rate of change of the first slope value and the second slope value is greater than a preset rate of change, the anomaly point determination parameter is obtained, and the anomaly point determination parameter is used as the determination result of the coordinate point.

5. The method according to claim 1, characterized in that: The method of constructing a coal seam face model using the layer data based on the Kriging interpolation algorithm includes: For each layer, extract multiple layer coordinate points corresponding to the layer in the layer data, calculate the distance value and semi-variance of any two layer coordinate points in the multiple layer coordinate points, and obtain multiple distance values ​​and multiple semi-variances; Obtaining an exponential model included in the Kriging interpolation algorithm, and performing function fitting on the multiple distance values ​​and the multiple semi-variances using the exponential model to obtain a target function; Extracting a plurality of target layer coordinate points from the plurality of layer coordinate points, constructing a first semi-variance coefficient matrix equation using the objective function and the plurality of target layer coordinate points, wherein the target layer coordinate point is a layer coordinate point having an elevation value among the plurality of layer coordinate points; Extracting a plurality of designated layer coordinate points from the plurality of layer coordinate points, wherein the designated layer coordinate points are layer coordinate points that do not have elevation values ​​from the plurality of layer coordinate points; For each of the designated layer coordinate points, the semivariance from the designated layer coordinate point to the multiple target layer coordinate points is calculated using the objective function to obtain multiple designated semivariances, and the second semivariance coefficient equation of the designated layer coordinate point is constructed using the multiple designated semivariances and the objective function; Using the first semivariance coefficient matrix equation and the second semivariance coefficient equation of each of the designated layer coordinate points to construct a target equation group, solving the target equation group to obtain weighted coefficients; Extracting the elevation value of each target layer coordinate point from the layer data, performing weighted sum calculation on the elevation values ​​of the multiple target layer coordinate points using the weight coefficient, and obtaining estimated values ​​of the multiple designated layer coordinate points; The Kriging interpolation algorithm is based on the layer data and is used to calculate each layer respectively, so as to obtain the elevation values ​​of multiple target layer coordinate points corresponding to each layer and the estimated values ​​of multiple specified layer coordinate points, and the coal seam face model is constructed by using the elevation values ​​of multiple target layer coordinate points corresponding to each layer and the estimated values ​​of multiple specified layer coordinate points.

6. The method according to claim 1, characterized in that The method of constructing a fault model by using the fault map in the geological related data and generating a stratigraphic framework grid by using the ground drilling data, the virtual drilling data and the fault model comprises: Acquire the fault map from the geological related data, perform coordinate correction on the fault map, and obtain a corrected fault map; Reading the fault dip angle and elevation range in the corrected fault map, respectively calculating the projection positions of the coal seam top line and the coal seam bottom line in the corrected fault map using the fault dip angle and the elevation range, and drawing the fault top line and the fault bottom line in the corrected fault map based on the projection positions of the coal seam top line and the coal seam bottom line in the corrected fault map to obtain a target fault map; Performing vectorization processing on the target tomogram, automatically tracking pixels on the processed target tomogram to obtain a plurality of first pixel point coordinates, filtering the plurality of first pixel point coordinates, and using the filtered plurality of first pixel point coordinates to generate the tomography file; Extracting a plurality of fault coordinate points from the fault file, obtaining a preset spacing, and digitally processing the plurality of fault coordinate points according to the preset spacing to obtain a plurality of discrete line segments; For each discrete line segment, the endpoint coordinates of the discrete line segment are obtained to obtain a first endpoint coordinate and a second endpoint coordinate, the plane vector and the vertical vector corresponding to the discrete line segment are calculated using the first endpoint coordinate and the second endpoint coordinate, and the target vector perpendicular to the discrete line segment and parallel to the fault plane is calculated using the plane vector and the vertical vector; Calculate the product of the plane vector and the target vector to obtain a fault plane normal vector, obtain an initial plane equation, substitute the fault plane normal vector and the first endpoint coordinates into the initial plane equation, and obtain a fault plane equation of the discrete line segment; Processing each of the discrete line segments respectively to obtain a fault plane equation of each of the discrete line segments, obtaining an elevation range in the fault file, and constructing the fault model using the fault plane equation and the fault plane equation of each of the discrete line segments; Acquire the fault plane of the fault model, divide the geological body corresponding to the fault model into three planes in the elevation direction, determine the intersection points of the three planes and the fault plane, and obtain intermediate shape points; Obtaining a preset coordinate direction and a preset number of grids corresponding to the preset coordinate direction, determining a model boundary according to the ground drilling data, the virtual drilling data and the fault model, and generating a plane quadrilateral grid using the intermediate shape point, the model boundary, the preset coordinate direction and the preset number of grids corresponding to the preset coordinate direction; The fault strike, fault top and fault bottom are obtained in the fault model, and the plane quadrilateral grid is constructed to the fault top and the fault bottom according to the fault strike to obtain a columnar three-dimensional frame network connecting the fault top, the middle of the fault and the fault bottom, and the columnar three-dimensional frame network is used as the stratigraphic frame network.

7. The method according to claim 1, characterized in that The vertical frame grid construction operation is performed based on the coal seam face model and the stratum frame grid to obtain a three-dimensional geological model, including: Importing the coal seam face model into the stratigraphic framework network to obtain a plurality of nodes, obtaining a preset grid unit, and using the preset grid unit to construct a grid for each of the nodes in a vertical direction to obtain a three-dimensional grid; The three-dimensional grid is adjusted using the drilling layer file to obtain a specified three-dimensional grid, and the specified three-dimensional grid is filled to obtain the three-dimensional geological model.

8. A modeling device for high-precision coal seams by fusion of multi-source data, characterized in that: include: A data generation module is used to obtain geological related data, generate ground drilling data using a comprehensive drilling column chart in the geological related data, and generate virtual drilling data using a coal seam contour map, a coal seam equal thickness map, and a gas drainage hole construction acceptance record in the geological related data; A first model building module is used to extract layer data from the ground drilling data and the virtual drilling data, and to build a coal seam face model using the layer data based on a Kriging interpolation algorithm; A grid generation module, used to construct a fault model using the fault map in the geological related data, and to generate a stratigraphic framework grid using the ground drilling data, the virtual drilling data and the fault model; The second model building module is used to perform a vertical frame grid building operation based on the coal seam face model and the stratigraphic frame grid to obtain a three-dimensional geological model.

9. A device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images