A stratum interface interpolation modeling method for intelligent mining

Abstract

This invention discloses a formation interface interpolation modeling method for intelligent mining: Step 1: Organize the data measured at the working face into a set α; Step 2: Grid the set α to form a set β1 of grid points; Step 3: Obtain the coordinates of the insertion points in the insertion point set φ1; Step 4: Set the initial z-coordinate of each point in the point set φ1; Step 5: Set the weight of each point in the point set φ1 and the set β1; Step 6: Define the potential η of each point in the point set φ1 and the set β1. a Step 7: Obtain the union of point set φ1 and point set β1. Step 8: Repeat step 7. Step 9: Replace the set β1 obtained in step 8 with the set β1 obtained in step 2 to obtain a new union. Step 10: Calculate the grid spacing of the union and determine whether the grid spacing reaches the target grid spacing. This invention can fix the grid size according to the realization of the working surface, and at the same time perform local dynamic updates, effectively utilizing real-time exposed data to obtain a more accurate model.

Description

Technical Field

[0001] This invention belongs to the field of coalfield geology and relates to a method for interpolation modeling of stratigraphic interfaces, which is used to realize the functions of stratigraphic interface meshing and dynamic updating of stratigraphic interfaces. Background Technology

[0002] Intelligent and reduced-manpower operations in coal mines are crucial for achieving coal mine safety. Coal mine intelligentization has progressed through four stages: visual remote intervention, automatic face straightening, intelligent coal cutting based on transparent faces, and fully intelligent adaptive mining. Currently, it is at a critical stage of intelligent coal cutting technology using transparent faces. This technology obtains the actual layout of the working face through detection methods such as 3D seismic surveys, tunnel surveys, and borehole surveys. After analyzing this data, a working face model is established using different interpolation methods, and then used to guide the mining machine's operations. A high-precision coal seam model of the working face is key, and interpolation methods are essential for realizing this model.

[0003] Currently, commonly used spatial interpolation methods include function interpolation, kriging interpolation, and smooth discrete interpolation. Due to the principle of smooth discrete interpolation, it cannot form a uniformly sized grid in the X and Y directions, but for a working surface, the size of each cut is basically fixed. Therefore, during model use, the grid size needs to match the size of each cut. Kriging interpolation, due to its algorithmic principle, requires a complete model update, making local updates difficult. For working surface models, which are typically large, a complete model update places higher demands on the system and wastes resources. Achieving local dynamic updates would solve these problems. Function interpolation includes distance-weighted average interpolation (IDW), trend surface methods, and spline function methods. These methods fit data from known points and predict data from unknown points using a function, resulting in lower model accuracy. Summary of the Invention

[0004] The purpose of this invention is to provide a method for interpolation modeling of formation interfaces to overcome the problems of existing technologies, such as the inability to form grids of uniform size and the difficulty in local updates.

[0005] To achieve the above objectives, this invention discloses a method for interpolation modeling of formation interfaces, comprising at least the following steps:

[0006] A formation interface interpolation modeling method for intelligent mining includes the following steps:

[0007] Step 1: Organize the data obtained from the working face measurement into a set α, which contains three columns of data (x, y, z), where x represents the horizontal coordinate, y represents the vertical coordinate, and z represents the elevation;

[0008] Step 2: Grid the set α to form a set β1 of grid points;

[0009] Step 3: Based on the x and y coordinates of set β1, obtain the coordinates of the insertion points in the insertion point set φ1;

[0010] Step 4: Set the initial z-coordinate of each point in the point set φ1;

[0011] Specifically, the average z-coordinate of set α is assigned to the initial z-coordinate of each point in point set φ1;

[0012] Step 5: Set the weight of each point in point set φ1 and set β1;

[0013] Step 6: Define the potential η of each point in the point set φ1 and the set β1. a ;

[0014] Step 7: Let η a =0, according to the definition in step 6, calculate the z of each grid point in the current point set φ1 and set β1. a The value is used to obtain the union of point set φ1 and point set β1.

[0015] Step 8: Repeat step 7 for at least 10 iterations to obtain the union of the current point set φ1 and point set β1.

[0016] Step 9: Replace the union of the current point set φ1 and point set β1 obtained in Step 8 with the set β1 obtained in Step 2. Given set β1, we obtain a new union.

[0017] Step 10: Calculate the union obtained in Step 9 The grid spacing is set, and it is determined whether the grid spacing meets the target grid spacing. If it does, the union set is output. As the final interpolation result; otherwise, the union will be used. Return to step 3 as set β1;

[0018] Furthermore, the gridding process in step 2 employs function fitting calculation or Kriging interpolation.

[0019] Furthermore, in step 3, the formula for calculating the coordinates of the insertion point is:

[0020] x n =(x n+1 +x n-1 ) / 2

[0021] y n =(y n+1 +yn-1 ) / 2

[0022] Where, x n y n x represents the coordinates of an insertion point in the insertion point set φ1; n-1 y n-1 x n+1 y n+1 These are the coordinates of two adjacent points in set β1.

[0023] Furthermore, in step 5, the weight of each point is set to 1.

[0024] Furthermore, in step 6, the potential η of each point in the point set φ1 and the set β1 a Defined as:

[0025]

[0026] Where O(a) represents the set of four grid points (upper, lower, left, and right) that share a common boundary with point a, and i is the index of the four points (upper, lower, left, and right); λ i Let i be the weight of point i;

[0027] If point a is in the middle grid, then the z values ​​of the four grid points a1, a2, a3, and a4 that share an edge with it are included in the calculation;

[0028] If point a is a boundary grid, then the z values ​​of the three grid points a1, a2, and a3 that share a common edge with it are included in the calculation;

[0029] If point a is a corner grid, then the z values ​​of the two grid points a1 and a2 that share a common edge with it are included in the calculation.

[0030] Furthermore, in step 8, the loop is repeated 10 times.

[0031] Compared with the prior art, the method of the present invention has the following technical effects:

[0032] 1. It can set the grid size according to the specific conditions of the working face, and can achieve a fixed grid size. In the intelligent mining process, it can directly provide the cutting curve based on the cutting depth of the coal mining machine without data processing.

[0033] 2. It can perform local dynamic updates and effectively utilize real-time exposed data. As the working face is re-examined, new data is continuously exposed, and the exposed data is dynamically updated within a certain range to obtain a more accurate model. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation examples of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a schematic diagram of the location of grid a, where (a) is the middle grid; (b) is the edge grid; and (c) is the corner grid.

[0036] Figure 2 This is a rendering of the working surface generated by the method of the present invention;

[0037] Figure 3 This is a diagram showing the effect of dynamically updating the working surface according to the present invention. Detailed Implementation

[0038] The formation interface interpolation modeling method for intelligent mining provided in this invention specifically includes the following steps:

[0039] Step 1: Organize the data obtained from the working face measurement into a set α. The set contains three columns of data (x, y, z), which are the coordinate data of the coal seam obtained from the working face measurement. Here, x represents the horizontal coordinate, y represents the vertical coordinate (x and y are commonly expressed in latitude and longitude), and z represents the elevation.

[0040] Table 1. Format of the midpoint of set α

[0041] Serial Number x y z 1 x1 y1 z1 2 x2 y2 z2 3 x3 y3 z3 ... ... ... ...

[0042] Step 2: Grid the set α to form a set β1 of grid points;

[0043] Specifically, there are many methods to transform the discrete data obtained in step 1 into equally spaced grid data, including function fitting calculation and Kriging interpolation. Kriging interpolation can be performed using the PyKrige toolbox.

[0044] Step 3: Based on the x and y coordinates of set β1, obtain the coordinates of the insertion points in the insertion point set φ1;

[0045] x n =(x n+1 +x n-1 ) / 2

[0046] y n =(y n+1 +y n-1 ) / 2

[0047] Where, x n y n x represents the coordinates of an insertion point in the insertion point set φ1; n-1 y n-1 x n+1 y n+1 These are the coordinates of two adjacent points in set β1.

[0048] Step 4: Set the initial z-coordinate of each point in the point set φ1;

[0049] Specifically, the average z-coordinate of set α is assigned to the initial z-coordinate of each point in point set φ1.

[0050] Step 5: Set the weight λ of each point in the point set φ1 and the set β1. i ;

[0051] Specifically, the weights are assigned based on the accuracy of the data. For example, if the measured data is relatively accurate, its weight can be slightly larger; if the error in the 3D seismic interpretation data is large, its weight can be reduced. Generally, the weight of each point is set to 1.

[0052] Step 6: Define the potential η of each point in the point set φ1 and the set β1. a ;

[0053] Specifically, Where O(a) represents the set of four grid points (upper, lower, left, and right) that share a common boundary with point a (i is the index of the four grid points).

[0054] If point a is in the middle grid (e.g.) Figure 1 (a) shows that the z values ​​of the four grid points a1, a2, a3, and a4 that share a common edge with it are included in the calculation;

[0055] If point a is a boundary grid (e.g.) Figure 1 (b) shows that the z values ​​of the three grid points a1, a2, and a3 that share a common edge with it are included in the calculation.

[0056] If point a is a corner grid (e.g.) Figure 1 (b) As shown, the z values ​​of the two grid points a1 and a2 that share a common edge with it are included in the calculation.

[0057] Step 7: Let η a =0, according to the definition in step 6, calculate the z of each grid point in the current point set φ1 and set β1. a The value is used to obtain the union of point set φ1 and point set β1.

[0058] Specifically, let η a =0, z aSince the values ​​are set to unknowns, while the other values ​​in the formula defined in step 6 are known, the z-values ​​of each point in the point set φ1 and set β1 can be calculated. a value.

[0059] Step 8: Repeat step 7 for at least 10 iterations to obtain the union of the current point set φ1 and point set β1.

[0060] This step is an iterative process, continuously calculating to smooth out the parameters of the obtained points. The number of iterations depends on the actual situation. If there are many variations in the formation interface, i.e., the formation has large undulations, the number of iterations needs to be increased; if the formation interface is relatively stable with small undulations, the number of iterations can be appropriately reduced. To make the formation interface smoother, it is recommended that the number of iterations be no less than 10.

[0061] Step 9: Replace the union of the current point set φ1 and point set β1 obtained in Step 8 with the set β1 obtained in Step 2. Given set β1, we obtain a new union.

[0062] Step 10: Calculate the union obtained in Step 9 The grid spacing is set, and it is determined whether the grid spacing meets the target grid spacing. If it does, the union set is output. As the final interpolation result; otherwise, the union will be used. As set β1, return to step 3.

[0063] The formula for calculating grid spacing is as follows:

[0064] Δx=x n -x n-1

[0065] Δy=y n -y n-1

[0066] Where, x n y n x represents the coordinates of the insertion point. n-1 x n+1 , y n-1 y n+1 The coordinates of the grid points are obtained in step 3.

[0067] The target mesh spacing is selected based on the length of each cut made on the working surface in both the x and y directions. The target mesh spacing is considered achieved when the calculated values ​​for both the x and y directions are simultaneously less than the target mesh spacing.

[0068] Example:

[0069] I. Overview of Modeling Coal Seams, i.e., Coal Seam Geological Data

[0070] A working face is 3500m long and 271m wide. The entire working face is considered as the research object, and the roof interface is modeled. The coal seam thickness within the working face ranges from 1.3m to 3.43m, with an average thickness of approximately 2.72m. The coal seam dip angle is relatively large in the middle of the working face.

[0071] The data used for modeling mainly includes: ① realistic data of the tunnels ② realistic data of the cut-off sections after mining.

[0072] The realistic data of the roadway was organized to obtain the interface points of the coal seam floor. Some data are shown in Table 2:

[0073] Table 2

[0074]

[0075]

[0076]

[0077] II. Grid Processing

[0078] This embodiment uses the PyKrige toolbox to implement the gridding of discrete data. The specific steps are as follows: input discrete points, establish an interpolation model; establish a grid, and according to the established interpolation model, obtain the set β1 by calculating the z-values ​​on the grid.

[0079] 3. Calculate the x and y values ​​of the newly inserted point to obtain the coordinates of the inserted point in the insertion point set φ1.

[0080] Set the initial z value for each point in the point set φ1, which is the average z coordinate obtained from each iteration.

[0081] IV. Assigning the initial z coordinates of the point set φ1;

[0082] 5. Assign a weight value of 1 to each point in both the point set φ1 and the set β1;

[0083] VI. Define the potential η of each point in the point set φ1 and the set β1. a ;

[0084] VII. Let η a =0, according to the definition in step six, calculate the z of each grid point in the current point set φ1 and set β1. a The value is used to obtain the union of point set φ1 and point set β1.

[0085] 8. Repeat step 7 for 10 iterations.

[0086] 9. Obtain the new union Calculate the union obtained in step nine The grid spacing is set to the target grid spacing of (0.8m, 0.1m). Steps 3-9 are executed five times. The calculated grid spacing is less than the target grid spacing in both directions. At this point, the target requirement is met, and the union set is output. To obtain the final model, such as Figure 2 As shown.

[0087] 10. Use the union of outputs The working face data is dynamically updated locally, and the update result is as follows: Figure 3 As shown. Figure 3 Eye cutting relative to Figure 2 The incision site was dynamically updated.

[0088] As can be seen from the above, the advantages of the present invention are:

[0089] 1. The grid size can be set according to the specific conditions of the working face. During intelligent mining, the cutting curve can be directly provided based on the cutting depth of the coal mining machine without data processing.

[0090] 2. It can perform local dynamic updates and effectively utilize real-time exposed data. As the working face is re-examined, new data is continuously exposed, and the exposed data is dynamically updated within a certain range to obtain a more accurate model.

[0091] It is obvious that the above description and account are merely illustrative and not intended to limit the disclosure, application, or use of this invention. Although embodiments have been described and illustrated in the accompanying drawings, the invention is not limited to the specific examples exemplified by the drawings and described in the embodiments as currently considered the best mode for carrying out the teachings of the invention. The scope of the invention will include any embodiments falling within the foregoing description and the appended claims.

Claims

1. A method for intelligent exploitation-oriented formation interface interpolation modeling, characterized in that, Specifically, the following steps are included: Step 1: Organize the data obtained from the working face measurement into a set α, which contains three columns of data (x, y, z), where x represents the horizontal coordinate, y represents the vertical coordinate, and z represents the elevation; Step 2: Grid the set α to form a set β1 of grid points; Step 3: Based on the x and y coordinates of set β1, obtain the coordinates of the insertion points in the insertion point set φ1; Step 4: Set the initial z-coordinate of each point in the point set φ1; Specifically, the average z-coordinate of set α is assigned to the initial z-coordinate of each point in point set φ1; Step 5: Set the weight of each point in point set φ1 and set β1; Step 6: Define the potential η of each point in the point set φ1 and the set β1 a ; Step 7: Let η a =0, according to the definition in step 6, calculate the z of each grid point in the current point set φ1 and set β1. a The value is used to obtain the union of point set φ1 and point set β1. Step 8: Repeat step 7 for at least 10 iterations to obtain the union of the current point set φ1 and point set β1. Step 9: Replace the union of the current point set φ1 and point set β1 obtained in Step 8 with the set β1 obtained in Step 2. Given set β1, we obtain a new union. Step 10: Calculate the union obtained in Step 9 The grid spacing is set, and it is determined whether the grid spacing meets the target grid spacing. If it does, the union set is output. As the final interpolation result; otherwise, the union will be used. As set β1, return to step 3.

2. The formation interface interpolation modeling method for intelligent mining as described in claim 1, characterized in that, The gridding process in step 2 is performed using function fitting or kriging interpolation.

3. The formation interface interpolation modeling method for intelligent mining as described in claim 1, characterized in that, In step 3, the formula for calculating the coordinates of the insertion point is: x n = (x n+1 + x n-1 ) / 2 y n = (y n+1 + y n-1 ) / 2 wherein x n , y n is the coordinate value of one insertion point in the insertion point set φ1; x n-1 , y n-1 , x n+1 , y n+1 are the coordinate values of two adjacent points in the set β1, respectively.

4. The formation interface interpolation modeling method for intelligent mining as described in claim 1, characterized in that, In step 5, the weight of each point is set to 1.

5. The formation interface interpolation modeling method for intelligent mining as described in claim 1, characterized in that, In step 6, the potential η of each point in the point set φ1 and the set β1 a is defined as: wherein, O(a) represents the set of the upper, lower, left and right grid points which have common boundary with point a, i is the serial number of the upper, lower, left and right points; λ i is the weight of point i; If point a is in the middle grid, then the z values ​​of the four grid points a1, a2, a3, and a4 that share an edge with it are included in the calculation; If point a is a boundary grid, then the z values ​​of the three grid points a1, a2, and a3 that share a common edge with it are included in the calculation; If point a is a corner grid, then the z values ​​of the two grid points a1 and a2 that share a common edge with it are included in the calculation.

6. The formation interface interpolation modeling method for intelligent mining as described in claim 1, characterized in that, In step 8, the loop is repeated 10 times.

Images