Grounding grid underground laying structure identification method, device and storage device

Abstract

This invention provides a method, device, and storage device for identifying the underground laying structure of a grounding grid. S1: Acquire detection data of the grounding grid, normalize it, and perform biharmonic sample interpolation to fill in scattered and missing data into uniformly distributed response data. S2: Perform wavelet noise reduction on the response data to reduce the noise impact of the complex underground environment and stray currents. S3: Obtain structural data points of different shapes based on the characteristics of the underground laying structure of the grounding grid. S4: Based on the obtained structural data points, remove scattered structural data points by setting a threshold for connected regions to obtain the final identification result of the underground laying structure of the grounding grid. A grounding grid underground laying structure identification device and storage device are also provided to implement a grounding grid underground laying structure identification method. The beneficial effects of this invention are: it solves the problem of complex grounding grid corrosion discrimination algorithms and realizes the identification of the underground laying location of grounding grids with arbitrary laying structures.

Description

Technical Field

[0001] This invention relates to the field of substation grounding grid structure identification, and in particular to a method, device and storage device for identifying the underground laying structure of grounding grid based on data completion and structural data point selection. Background Technology

[0002] Detecting and analyzing the grounding grid is the primary method for inspecting its underground structure. Non-contact detection methods based on Faraday's law of electromagnetic induction can obtain complete response data of the grounding grid without excavation or interruption of substation operation. Existing technologies often employ transient electromagnetic imaging diagnostics, locating grounding conductors by inverting the magnetic field or analyzing the electromagnetic response characteristics of the grounding grid under different defect states. However, these methods directly image the collected data for analysis, lacking targeted data processing, resulting in complex algorithms and high computational demands. Summary of the Invention

[0003] To address the aforementioned problems, this invention provides a method, device, and storage device for identifying the underground laying structure of a grounding grid, thereby resolving the complexity of grounding grid corrosion discrimination algorithms and enabling the identification of the underground laying location of grounding grids with arbitrary laying structures. A method for identifying the underground laying structure of a grounding grid mainly includes:

[0004] S1: Obtain the detection data of the grounding grid, normalize it and perform biharmonic sample interpolation to fill in the scattered and missing data into uniformly distributed response data;

[0005] S2: Perform wavelet noise reduction on the response data;

[0006] S3: Obtain structural data points of different shapes from the data after processing in steps S1-2;

[0007] S4: Based on the obtained structural data points, scattered structural data points are eliminated by setting a threshold for the connected region, and the final identification result of the underground laying structure of the grounding grid is obtained.

[0008] Further, in step S1, the probe data is normalized using the following maximum / minimum normalization formula to obtain the processed data z':

[0009]

[0010] Where z represents the original detection data.

[0011] Furthermore, in step S1, the process of performing biharmonic sample interpolation using the biharmonic spline interpolation method is as follows:

[0012] First, each detection data includes three variables: X-direction position X = [x1, x2, ..., x...]. n ] T Position in the Y direction: Y = [y1, y2, ..., y n ] T Magnetic induction intensity detection data at the corresponding location Z=[z1,z2,...,z n ] T Calculate the Green's function matrix G between n detection points:

[0013]

[0014]

[0015]

[0016] Where, d ij The Green's function represents the relationship between each detection point, (x i ,y i ), (x j ,y j ) represent the coordinates of two different detection points, r ij Representing point (x) i ,y i ) and point (x) j ,y j The Euclidean distance of ), where n represents the number of probe points;

[0017] Then, the interpolation weights W are obtained from the Green's function matrix:

[0018] W=G -1 Z

[0019] Finally, the interpolation point data Z_grid(x,y) is obtained based on the interpolation weights:

[0020] Z_grid(x,y) = W*g(x,y)

[0021] Where g(x,y) is the Green's function at the interpolation point.

[0022] g(x,y)=[d1,d2,...,d n ]

[0023] d i =[ln(r i )-1]*r i 2

[0024] Where, r i d represents the Euclidean distance from the interpolation point to each probe point. iLet i = 1, 2, ..., n represent the Green's function between the interpolation point and each probe point.

[0025] Furthermore, in step S3, based on the structural characteristics of the underground grounding grid, “I”, “L”, “T”, and “X” shapes are proposed, and structural data points are identified for each shape. The process is as follows:

[0026] 1) Regarding the underground structure of a certain part of the "I"-shaped grounding grid

[0027] If a certain structural data point P(i,j) satisfies any of the following formulas, it means that the response data of this point is greater than or equal to the response data of the two points above and below or the two points to the left and right, and does not satisfy other shape conditions, then this point is an "I" shaped structural data point of the underground laying structure of the grounding grid:

[0028] P(i,j)≥P(i+1,j)、P(i-1,j)

[0029] P(i,j)≥P(i,j+1), P(i,j-1)

[0030] Where P(i-1,j) and P(i+1,j) represent the structural data points adjacent to the left and right of structural data point P(i,j), respectively, and P(i,j+1) and P(i,j-1) represent the structural data points adjacent to the top and bottom of structural data point P(i,j), respectively.

[0031] 2) Regarding the underground structure of a certain part of the "L"-shaped grounding grid

[0032] If a certain structural data point P(i,j) satisfies any one of the following formulas, it means that the response data of this point is greater than or equal to the response data of any two points in any adjacent direction among the four points (up, down, left, and right) and does not satisfy other shape conditions, then this point is an "L"-shaped structural data point of the underground laying structure of the grounding grid:

[0033] P(i,j)≥P(i+1,j), P(i,j-1)≥P(i-1,j), P(i,j+1)

[0034] P(i,j)≥P(i+1,j), P(i,j+1)≥P(i-1,j), P(i,j-1)

[0035] P(i,j)≥P(i-1,j), P(i,j-1)≥P(i+1,j), P(i,j+1)

[0036] P(i,j)≥P(i-1,j), P(i,j+1)≥P(i+1,j), P(i,j-1)

[0037] 3) Regarding a certain underground section of the "T"-shaped grounding grid structure

[0038] If a structural data point P(i,j) satisfies any one of the following formulas, it means that the response data of this point is greater than or equal to the response data of any three of the four points (up, down, left, and right) and does not satisfy other shape conditions, then this point is...

[0039] Data points for the "T"-shaped structure of the underground grounding grid:

[0040] P(i,j)≥P(i+1,j), P(i,j-1), P(i-1,j)≥P(i,j+1)

[0041] P(i,j)≥P(i+1,j), P(i,j-1), P(i,j+1)≥P(i-1,j)

[0042] P(i,j)≥P(i+1,j), P(i,j+1), P(i-1,j)≥P(i,j-1)

[0043] P(i,j)≥P(i,j+1), P(i,j-1), P(i-1,j)≥P(i+1,j)

[0044] 4) Regarding the underground structure of a certain part of the "X"-shaped grounding grid

[0045] If a certain structural data point P(i,j) satisfies the following formula, it means that the response data of this point is greater than or equal to the response data of the four points above, below, left, and right. Then this point is a data point of the "X" shaped structure of the underground grounding grid:

[0046] P(i,j)≥P(i+1,j), P(i,j-1), P(i-1,j), P(i,j+1).

[0047] Furthermore, in step S4, the process of eliminating scattered structural data points is as follows:

[0048] First, determine all structural data points. If a point has structural data points in more than one of its eight directions, then that point is a connected point.

[0049] Then, determine whether the number of connected points in each connected region is greater than or equal to the set threshold. If yes, all connected points in the connected region are still structured data points; otherwise, all connected points in the connected region are no longer structured data points.

[0050] Furthermore, the method also includes an evaluation of the recognition results:

[0051] min(dist(P(i,j),Loc))≤d

[0052] Where dist() represents the distance between the structure data point P(i,j) and the true structure Loc, and d is the confidence range size. If the structure data point is within the confidence range, then the structure data point is a reliable structure data point.

[0053] A storage device that stores instructions and data for implementing a method for identifying the underground laying structure of a grounding grid.

[0054] A grounding grid underground laying structure identification device includes: a processor and a storage device; the processor loads and executes instructions and data in the storage device to implement a grounding grid underground laying structure identification method.

[0055] The beneficial effects of the technical solution provided by this invention are as follows: This invention acquires grounding grid detection data, performs normalization and biharmonic sample interpolation processing on it, and completes the scattered and missing data into uniformly distributed response data; it performs wavelet denoising processing on the response data; it obtains structural data points of different shapes from the processed data; based on the obtained structural data points, it removes scattered structural data points by setting a threshold for connected regions, and obtains the final identification result of the underground laying structure of the grounding grid. Normalization and biharmonic sample interpolation are used to complete the detection data from discrete and missing to spatially uniformly distributed data; wavelet denoising is used to reduce the noise of the detection data; and a method for detecting structural data points with four different structural shapes, including "I", "L", "T", and "X" shapes, is proposed; a threshold for connected regions is set to remove scattered structural data points, thus achieving the screening of structural data points; finally, a method for evaluating the accuracy of underground laying structure identification of grounding grids based on the proportion of structural data points within the confidence interval of the actual laying structure is obtained, solving the problem of complex grounding grid corrosion discrimination algorithms and realizing the identification of the underground laying location of grounding grids with arbitrary laying structures. Attached Figure Description

[0056] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings:

[0057] Figure 1 This is a flowchart of a method for identifying the underground laying structure of a grounding grid according to an embodiment of the present invention.

[0058] Figure 2 This is a schematic diagram of the underground laying structure features of the grounding grid in an embodiment of the present invention.

[0059] Figure 3 This is a schematic diagram illustrating the principle of selecting connected regions in an embodiment of the present invention.

[0060] Figure 4 This is a schematic diagram of the confidence range of the grounding grid laying structure in an embodiment of the present invention.

[0061] Figure 5 This is a schematic diagram of the construction of a non-uniform underground grounding grid structure model in an embodiment of the present invention.

[0062] Figure 6 This is a schematic diagram of the simulation results of the response data in an embodiment of the present invention.

[0063] Figure 7 This is a schematic diagram of the non-uniformly distributed raw data in an embodiment of the present invention.

[0064] Figure 8 This is a schematic diagram of uniformly distributed completion data in an embodiment of the present invention.

[0065] Figure 9 This is a schematic diagram of the response data after data completion in an embodiment of the present invention.

[0066] Figure 10 This is a schematic diagram of the structural data point selection results in an embodiment of the present invention.

[0067] Figure 11 This is a schematic diagram of the structural data point filtering results in an embodiment of the present invention.

[0068] Figure 12 This is a schematic diagram comparing the identified structural data points with the actual laying positions in an embodiment of the present invention.

[0069] Figure 13 This is a schematic diagram of the hardware device working in an embodiment of the present invention. Detailed Implementation

[0070] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0071] The embodiments of the present invention provide a method, device and storage device for identifying the underground laying structure of grounding grid based on data completion and structural data point selection.

[0072] Please refer to Figure 1 , Figure 1 This is a flowchart of a method for identifying the underground laying structure of a grounding grid according to an embodiment of the present invention, specifically including:

[0073] S1: Obtain the detection data of the grounding grid, normalize it and perform biharmonic sample interpolation to fill in the scattered and missing data into uniformly distributed response data;

[0074] S2: When detecting underground grounding grids, the detection data will be affected by the complex underground environment and stray currents, resulting in noise. To avoid this effect, wavelet noise reduction processing is required on the response data to reduce the noise impact of the complex underground environment and stray currents on the response data.

[0075] S3: In the data after processing steps S1-S2, structural data points of different shapes are obtained based on the underground laying structure characteristics of the grounding grid;

[0076] S4: Based on the obtained structural data points, scattered structural data points are eliminated by setting a threshold for the connected region, and the final identification result of the underground laying structure of the grounding grid is obtained.

[0077] After processing the detection data, the problems of uneven spatial distribution of the detection data (step S1) and noise (step S2) were solved. After obtaining these processed data, grounding grid structure data points were selected from them, and then the underground laying structure of the grounding grid was identified.

[0078] This embodiment uses the COMSOL Multiphysics simulation platform to build a grounding grid model of a non-uniform underground grounding grid laying structure. Simulation results are used to obtain detection response data for experimental verification, validating the effectiveness of the method provided by this invention. According to national standards, for example, the length of the grounding grid conductor buried in the ground must be greater than or equal to 15m, and the burial depth of the grounding grid must be greater than or equal to 0.8m, to build a non-uniform underground grounding grid laying structure model. The model parameters are as follows: the grounding grid is laid 1 meter underground, with a 1-meter-thick soil layer; the conductor is an 8mm diameter round steel bar; the detection instrument is 0.08 meters above ground, with a 0.08-meter-thick air layer; the detection current intensity is 5A; the current is injected at (3,23) and flows out at (27,3). The specific structure is as follows... Figure 5 As shown. After the model was built, model simulation was performed. The simulation data was then used for experimental verification. The simulation results are as follows. Figure 6 As shown.

[0079] (1) Normalization process:

[0080] Because the range of values ​​for the detection data z is variable, influenced by the underground structure of the grounding grid and the strength of the injected current during detection, it is necessary to normalize the detection data to ensure the scheme has universal applicability.

[0081] This invention employs existing technology: extreme value normalization processing, which normalizes the original detection data to obtain the processed data z', as shown in the formula.

[0082]

[0083] In this embodiment, the minimum value of the simulation data is 0 and the maximum value is 1. The data is proportionally mapped to values ​​between 0 and 1.

[0084] (2) Data completion:

[0085] Observing the distribution of the detection data reveals that the grounding grid detection data is discrete and even incomplete. The reasons for this can be attributed to two main factors: firstly, the manual propulsion of the instrument cannot guarantee a uniform speed; secondly, the influence of the substation equipment on the ground forces the detection instrument to stop or detour at certain points. Therefore, when processing the raw detection data, it is necessary to perform data completion processing to obtain complete and orderly response data within the operator's detection area.

[0086] In this embodiment, biharmonic spline interpolation is used to complete the response data obtained after normalization. Compared with other spline interpolation methods, biharmonic spline interpolation can eliminate corner characteristics while maintaining smoothness and second-order continuity. First, each detection data includes three variables (X-direction position X = [x1, x2, ..., x...). n ] T Position in the Y direction: Y = [y1, y2, ..., y n ] T Magnetic induction intensity detection data at the corresponding location Z=[z1,z2,...,z n ] T If the magnetic field strength detection data at position (x1, y1) is z1, calculate the Green's function matrix G between the n detection points:

[0087]

[0088]

[0089]

[0090] Among them, (x i ,y i ), (x j ,y j ) represent the coordinates of two different locations, r ij Representing point (x) i ,y i ) and point (x) j ,y j The Euclidean distance of ), where n represents the number of probe points;

[0091] Then, the interpolation weight W is determined by the original data values ​​and the Green's function matrix:

[0092] W=G -1 Z

[0093] Finally, the value of each interpolation point Z_grid(x,y) is determined by the interpolation weights and the Green's function at that point. The data of the interpolation point can be represented as Z_grid(x,y):

[0094] Z_grid(x,y) = W*g(x,y)

[0095] Where g(x,y) is the Green's function at the interpolation point, and r i is the Euclidean distance from the interpolation point to each original detection point.

[0096] g(x,y)=[d1,d2,...,d n ]

[0097] d i =[ln(r i )-1]*r i 2

[0098] Where, d i The Green's function, r, represents the relationship between the interpolation point and each original probe point. i The interpolation point represents the Euclidean distance between the interpolation point and each original detection point, Z represents the original data value matrix, and d represents the Euclidean distance between the interpolation point and each original detection point. ij The Green's function, r, represents the relationship between the original detection points. ij This represents the Euclidean distance between each original detection point.

[0099] By interpolating the original detection data, scattered and missing data are filled in to form uniformly distributed response data. For example... Figure 7 , Figure 8 As shown, comparing the distribution of the original data and the completed data reveals that data completion transforms the non-uniform original data into uniformly distributed completed data. The response data after data completion is as follows: Figure 9 As shown.

[0100] (3) Data noise reduction:

[0101] Because the underground conductors of the substation grounding grid are buried in a complex soil environment year-round, the detection data obtained by operators during the detection process will contain noise components due to the influence of the underground environment, stray currents, and various short-circuit currents. Therefore, it is necessary to perform noise reduction processing on the detection data.

[0102] This embodiment employs existing technology: wavelet denoising for noise reduction. Wavelet transform is used to decompose the response data into various scales. At each scale, smaller wavelet coefficients (noise) that belong to noise are deleted, while larger wavelet coefficients (signal) that belong to the signal are retained and enhanced. Finally, the signal is reconstructed through inverse wavelet transform.

[0103] (4) Selection of structural data points:

[0104] By observing examples of underground grounding grid installations, it is observed that grounding grids generally employ a mesh structure during installation. Based on the structural characteristics of underground grounding grid installations, this invention classifies the various local areas of the mesh structure into four shapes: "I" shape, "L" shape, "T" shape, and "X" shape. Each shape has different data characteristics, and different identification methods are designed. The data processed in steps S1-S2 is filtered to determine which data matches the data characteristics corresponding to each shape, thus obtaining the structural data points for these four shapes. Figure 2 As shown, a method is provided to select the structural data points corresponding to each shape and their locations, based on different shapes. Finally, the structural data points of different shapes are integrated to identify the underground laying structure of the grounding grid. The algorithm for identifying structural data points for various shapes is as follows:

[0105] 1) For a certain underground part of the "I"-shaped grounding grid structure, if a certain data point P ij If any one of the formulas is satisfied, that is, the response data of the point is greater than or equal to the response data of the two points above and below or the two points to the left and right, and does not satisfy other shape conditions, then the point is an "I" shaped structure data point of the underground laying structure of the grounding grid.

[0106] P(i,j)≥P(i+1,j)、P(i-1,j)

[0107] P(i,j)≥P(i,j+1), P(i,j-1)

[0108] Where P(i-1,j) and P(i+1,j) represent the structural data points adjacent to the left and right of structural data point P(i,j), respectively, and P(i,j+1) and P(i,j-1) represent the structural data points adjacent to the top and bottom of structural data point P(i,j), respectively.

[0109] 2) For a certain underground part of the "L"-shaped grounding grid structure, if a certain data point P ij If any one of the formulas is satisfied, that is, the response data of the point is greater than or equal to the response data of any two points in any adjacent direction among the four points (up, down, left, and right) and does not satisfy other shape conditions, then the point is an "L" shaped structure data point of the underground grounding grid laying structure.

[0110] P(i,j)≥P(i+1,j), P(i,j-1)≥P(i-1,j), P(i,j+1)

[0111] P(i,j)≥P(i+1,j), P(i,j+1)≥P(i-1,j), P(i,j-1)

[0112] P(i,j)≥P(i-1,j), P(i,j-1)≥P(i+1,j), P(i,j+1)

[0113] P(i,j)≥P(i-1,j), P(i,j+1)≥P(i+1,j), P(i,j-1)

[0114] 3) For a certain underground part of the "T"-shaped grounding grid structure, if a certain data point P ij If any one of the formulas is satisfied, that is, the response data of the point is greater than or equal to the response data of any three of the four points (up, down, left, and right) and does not satisfy other shape conditions, then the point is a data point of the "T" shaped structure of the underground grounding grid.

[0115] P(i,j)≥P(i+1,j), P(i,j-1), P(i-1,j)≥P(i,j+1)

[0116] P(i,j)≥P(i+1,j), P(i,j-1), P(i,j+1)≥P(i-1,j)

[0117] P(i,j)≥P(i+1,j), P(i,j+1), P(i-1,j)≥P(i,j-1)

[0118] P(i,j)≥P(i,j+1), P(i,j-1), P(i-1,j)≥P(i+1,j)

[0119] 4) For a certain underground part of the "X"-shaped grounding grid structure, if a certain data point P ij If the formula is satisfied, meaning the response data at this point is greater than or equal to the response data at the four points above, below, left, and right, then this point is a data point for the "X" shaped structure of the underground grounding grid.

[0120] P(i,j)≥P(i+1,j), P(i,j-1), P(i-1,j), P(i,j+1)

[0121] In this invention, data points whose values ​​are greater than the values ​​of the four points (up, down, left, and right) are designated as "X"-shaped data points; data points whose values ​​are greater than any three of the four points and do not meet other shape conditions are designated as "T"-shaped data points; data points whose values ​​are greater than any two of the four adjacent points and do not meet other shape conditions are designated as "L"-shaped data points; and data points whose values ​​are greater than either the two points above or below or the two points to the left and right and do not meet other shape conditions are designated as "I"-shaped data points. The results of the data point selection are as follows: Figure 10 As shown.

[0122] (5) Filtering of structural data points:

[0123] Since there are no small, isolated conductors during the substation grounding grid installation process, the integrated structural data points should be connected and free of scattered points. Therefore, to prevent scattered points from causing misjudgments of the structure and affecting the accuracy of identifying the underground grounding grid structure, this invention provides a method for setting a threshold for the size of the connected region to eliminate scattered points. The principle is as follows:

[0124] First, after obtaining all structural data points, the connected regions are divided. Then, each structural data point is considered connected if it has structural data points in more than one of its eight directions. For example... Figure 3 As shown, the gray areas are connected regions, while the black areas are scattered regions.

[0125] Then, it is determined whether the number of connected vertices in each connected region is greater than or equal to the set threshold. If so, all connected vertices in that region are still structured data points; otherwise, all connected vertices in that region are no longer structured data points. The structured data point filtering results are as follows: Figure 11 As shown.

[0126] (6) Accuracy of identification results for underground grounding grid laying structure:

[0127] Through the aforementioned solutions to the problems in identifying the underground laying structure of the grounding grid and the corresponding processing algorithms designed in this invention, data points and locations containing structural information were obtained, leading to the identification results of the underground laying structure of the substation grounding grid. Then, it is essential to evaluate the accuracy of the results. Therefore, this invention designs an evaluation index: the proportion of structural data points within the confidence interval of the actual laying structure. The principle is as follows:

[0128] like Figure 4 As shown, the black dots represent structural data points, the black lines represent the underground laying locations of the grounding grid, and the gray area represents the confidence interval of the actual laying structure. In other words, if a structural data point falls within the confidence interval, as shown in the formula, then that point is a reliable structural data point. By calculating the proportion of reliable structural data points among all structural data points, the accuracy of the identification result of the underground laying structure of the grounding grid can be obtained.

[0129] min(dist(P(i,j),Loc))≤d

[0130] Where dist() represents the distance between the structured data point P(i,j) and the true structure Loc, and d is the confidence range size.

[0131] Calculate the proportion of structural data points within the confidence interval of the actual laid structure. For example... Figure 12As shown, dots indicate the identification position of conductors with non-uniform grid laying structure, and straight lines indicate the actual position of conductors with non-uniform grid laying structure. The accuracy of the identification results reaches 92%.

[0132] Please see Figure 13 , Figure 13 This is a schematic diagram of the hardware device in operation according to an embodiment of the present invention. The hardware device specifically includes: a grounding grid underground laying structure identification device 401, a processor 402, and a storage device 403.

[0133] A grounding grid underground laying structure identification device 401: The grounding grid underground laying structure identification device 401 implements the grounding grid underground laying structure identification method.

[0134] Processor 402: The processor 402 loads and executes the instructions and data in the storage device 403 to implement the method for identifying the underground laying structure of the grounding grid.

[0135] Storage device 403: The storage device 403 stores instructions and data; the storage device 403 is used to implement the method for identifying the underground laying structure of the grounding grid.

[0136] The beneficial effects of this invention are as follows: This invention acquires grounding grid detection data, performs normalization and biharmonic sample interpolation processing on it, and completes the scattered and missing data into uniformly distributed response data; it performs wavelet denoising processing on the response data; it obtains structural data points of different shapes from the processed data; based on the obtained structural data points, it removes scattered structural data points by setting a threshold for connected regions, and obtains the final identification result of the underground laying structure of the grounding grid. Normalization and biharmonic sample interpolation are used to complete the detection data from discrete and missing to spatially uniformly distributed; wavelet denoising is used to reduce the noise of the detection data; and a method for detecting structural data points with four different structural shapes, including "I", "L", "T", and "X" shapes, is proposed; a threshold for connected regions is set to remove scattered structural data points, thus achieving the screening of structural data points; finally, a method for evaluating the accuracy of underground laying structure identification of grounding grids based on the proportion of structural data points within the confidence interval of the actual laying structure is obtained, solving the problem of complex grounding grid corrosion discrimination algorithms and realizing the identification of the underground laying location of grounding grids with arbitrary laying structures.

[0137] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for identifying an underground structure of a ground grid, characterized by: include: S1: Obtain the detection data of the grounding grid, normalize it and perform biharmonic sample interpolation to fill in the scattered and missing data into uniformly distributed response data; S2: Perform wavelet noise reduction on the response data; S3: Obtain structural data points of different shapes from the data after processing steps S1-S2; S4: Based on the obtained structural data points, scattered structural data points are eliminated by setting a threshold for the connected region to obtain the final identification result of the underground laying structure of the grounding grid. In step S1, the process of performing biharmonic sample interpolation using the biharmonic spline interpolation method is as follows: First, each detection data includes three variables: position in the X direction. Y-direction position Magnetic induction intensity detection data at corresponding locations Calculate the Green's function matrix G between n detection points: Where, dij represents the Green's function between each detection point, (xi, yi) and (xj, yj) represent the coordinates of two detection points at different locations, rij represents the Euclidean distance between points (xi, yi) and (xj, yj), and n represents the number of detection points; Then, the interpolation weights W are obtained from the Green's function matrix: Finally, the interpolation point data Z_grid(x,y) is obtained based on the interpolation weights: Where g(x,y) is the Green's function at the interpolation point. Where ri is the Euclidean distance from the interpolation point to each detection point, di represents the Green's function between the interpolation point and each detection point, and i=1,2,...,n; In step S3, based on the characteristics of the underground grounding grid structure, "I" shape, "L" shape, "T" shape, and "X" shape are proposed, and structural data points are identified for each shape. The process is as follows: Regarding a certain underground section of the "I"-shaped grounding grid structure, If a structural data point P(i,j) satisfies any of the following formulas, indicating that the response data of this point is greater than or equal to the response data of the two points above and below or the two points to the left and right, and does not satisfy other shape conditions, then this point is an "I" shaped structural data point of the underground grounding grid: Where P(i-1, j) and P(i+1, j) represent the structural data points adjacent to the left and right of structural data point P(i,j), respectively, and P(i, j+1) and P(i, j-1) represent the structural data points adjacent to the top and bottom of structural data point P(i,j), respectively. For a certain underground part of the "L"-shaped grounding grid structure If a certain structural data point P(i,j) satisfies any one of the following formulas, it means that the response data of this point is greater than or equal to the response data of any two points in any adjacent direction among the four points (up, down, left, and right) and does not satisfy other shape conditions, then this point is an "L"-shaped structural data point of the underground laying structure of the grounding grid: Regarding a certain underground section of the "T"-shaped grounding grid structure, If a certain structural data point P(i,j) satisfies any one of the following formulas, it means that the response data of this point is greater than or equal to the response data of any three of the four points (up, down, left, and right) and does not satisfy other shape conditions, then this point is a "T" shaped structural data point of the underground laying structure of the grounding grid: Regarding a certain underground section of the "X"-shaped grounding grid structure, If a certain structural data point P(i,j) satisfies the following formula, it means that the response data of this point is greater than or equal to the response data of the four points above, below, left, and right. Then this point is a data point of the "X" shaped structure of the underground grounding grid: In step S4, the process of removing scattered structural data points is as follows: First, determine all structural data points. If a point has structural data points in more than one of its eight directions, then that point is a connected point. Then, determine whether the number of connected points in each connected region is greater than or equal to the set threshold. If yes, all connected points in the connected region are still structured data points; otherwise, all connected points in the connected region are no longer structured data points.

2. The method of claim 1, wherein: In step S1, the probe data is normalized using the maximum / minimum value normalization formula shown below to obtain the processed data. : Where z represents the original detection data.

3. The method of claim 1, wherein: In step S2, the wavelet noise reduction process is to reduce the noise impact of the complex underground environment and stray currents on the response data.

4. The method of claim 1, wherein: The method also includes an evaluation of the recognition results: Where dist() represents the distance between the structure data point P(i, j) and the true structure Loc, and d is the confidence range size. If the structure data point is within the confidence range, then the structure data point is a reliable structure data point.

5. A storage device, characterized in that: The storage device stores instructions and data for implementing the grounding grid underground laying structure identification method according to any one of claims 1 to 4.

6. A ground net underground structure identification device characterized by comprising: include: A processor and a storage device; the processor loads and executes instructions and data in the storage device to implement the grounding grid underground laying structure identification method according to any one of claims 1 to 4.

Images