Multi-layer transformer substation plant measurement and calculation method based on multi-view projection and analytic hierarchy process

Abstract

The invention provides a multi-layer transformer substation plant measurement and calculation method based on multi-view projection and analytic hierarchy process, and belongs to the technical field of transformer substation plant information measurement, and the method comprises the steps: collecting point cloud data of a transformer substation plant, and carrying out the preprocessing of the collected point cloud data; performing hierarchical analysis of the point cloud on the point cloud data through a gradient segmentation method to obtain single-layer point cloud data; according to the obtained single-layer point cloud data, segmenting the single-layer point cloud data into a top plane and a side plane through normal vector analysis; projecting the top plane and the side plane obtained after segmentation to generate a multi-view projection image of the point cloud data of the transformer substation; and based on the multi-view projection image, obtaining the polygonal boundary of the transformer substation plant through contour extraction, and measuring and integrating the size information of each layer of the transformer substation plant. According to the method, efficient extraction and accurate measurement and calculation of the multi-layer transformer substation plant are realized, and the problems of incomplete noise removal, detail loss caused by down-sampling, inaccurate height threshold segmentation and the like in the existing method are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of substation plant information measurement, and particularly to a method for measuring multi-layer substation plants based on multi-view projection and hierarchical analysis. Background Art

[0002] With the continuous progress of laser scanning technology, 3D point cloud data, as a high-precision spatial information expression method, has been widely used in many fields, especially in the measurement and analysis of buildings. Laser scanning can generate high-density and high-resolution 3D point cloud data, providing comprehensive spatial information, which provides a basis for the geometric structure modeling, dimension measurement and status monitoring of buildings. However, in the face of complex building structures, especially large-scale and multi-level buildings such as substation plants, traditional point cloud data processing methods often fail to meet the actual needs in terms of efficiency and accuracy.

[0003] When dealing with building point cloud data, existing technologies usually adopt conventional methods such as denoising based on KD-tree and height threshold segmentation. These methods perform well when dealing with relatively simple single-layer buildings or regular geometric structures, but they expose obvious limitations when facing complex multi-layer buildings.

[0004] First of all, the denoising method based on KD-tree accelerates data query by establishing a spatial index of the point cloud and judges noise points through the statistical characteristics of the local neighborhood. This method has a certain ability to process noise in a small range, but in the face of an environment with a complex structure and dense building elements such as a substation plant, it is often difficult to effectively distinguish noise points from useful data points. Non-building elements (such as cables, brackets, equipment, etc.) in the substation environment are highly mixed with the structural characteristics of the plant. The denoising method based on KD-tree will be interfered by these complex elements, resulting in incomplete noise removal or misidentifying useful building data as noise, thus affecting subsequent building extraction and analysis.

[0005] Secondly, the existing height threshold segmentation method shows great limitations in the processing of multi-layer buildings. The height threshold segmentation method usually divides each layer of the building by setting different height thresholds based on the height information of the building. However, in a multi-layer substation plant, the height distribution of different floors is often irregular, and sometimes the height difference between floors is not significant. Since this method relies on height information, in practical applications, it is easily affected by local height changes of the building, resulting in inaccurate floor segmentation. In addition, this method cannot handle the complex geometric structures inside and outside the building, such as the intersection of external walls, internal walls, stairs and other structures, which leads to problems such as unclear stratification and incomplete extraction.

[0006] In summary, to deal with the complex point cloud structure and dense noise in substations, it is necessary to introduce multi-angle projections to disassemble the building structure. At the same time, due to the characteristics of multiple building levels and unclear boundaries, point cloud hierarchical analysis is also required to measure the precise geometric data of each floor of the plant building. Summary of the Invention

[0007] In order to overcome the deficiencies of the prior art, the purpose of the present invention is to provide a method for measuring multi-layer substation plant buildings based on multi-view projection and hierarchical analysis, which realizes the efficient extraction and precise measurement of multi-layer substation plant buildings, and solves the problems existing in the prior methods, such as incomplete noise removal, detail loss caused by downsampling, inaccurate height threshold segmentation, etc.

[0008] To achieve the above purpose, the present invention provides the following solutions:

[0009] A method for measuring multi-layer substation plant buildings based on multi-view projection and hierarchical analysis, comprising the following steps:

[0010] S1. Collect the point cloud data of the substation plant building, and preprocess the collected point cloud data;

[0011] S2. Perform hierarchical analysis of the point cloud on the preprocessed point cloud data by the gradient segmentation method to obtain single-layer point cloud data;

[0012] S3. According to the obtained single-layer point cloud data, divide the single-layer point cloud data into a top plane and a side plane by normal vector analysis;

[0013] S4. Project the top plane and the side plane obtained after segmentation to generate multi-view projection images of the substation point cloud data;

[0014] S5. Based on the multi-view projection images, obtain the polygonal boundary of the substation plant building by contour extraction, and measure and integrate the size information of each floor of the substation plant building.

[0015] Preferably, in step S1, the preprocessing of the collected point cloud data includes:

[0016] Denoise the point cloud data of the substation plant building by a top-down method;

[0017] Perform ground plane correction on the denoised point cloud data of the substation plant building;

[0018] After the correction is completed, perform coordinate system conversion on the point cloud data of the substation plant building to facilitate height calculation.

[0019] Preferably, denoising the point cloud data of the substation plant building by a top-down method includes:

[0020] Calculate the height range of the point cloud data:

[0021] z min = min(z i );

[0022] z max = max(z i );

[0023] where z min is the minimum height, z max is the maximum height, and z i is the height coordinate of the i-th point in the point cloud data;

[0024] Subsequently, set the height threshold z threshold as a reference, and mark the points above this threshold as possible noise points:

[0025] z threshold = z max - λ(z max - z min );

[0026] where λ is an adjustment coefficient, 0.01 ≤ λ ≤ 0.05;

[0027] Calculate the number of points N i in each layer of the substation building point cloud data, and set the density threshold N threshold . Points below this density are regarded as noise points. The formula is:

[0028]

[0029] where n is the total number of layers of the point cloud data, α is an adjustment coefficient, 0.5 ≤ α ≤ 0.8; Subsequently, remove the low-density points, that is, points with N i < N threshold are marked as noise and removed.

[0030] Preferably, perform ground plane correction on the substation building point cloud data after denoising, including:

[0031] Extract the ground points of the substation building point cloud data by RANSAC, and determine the equation of the ground plane through the plane fitting method. Set the fitted plane equation as:

[0032] ax + by + cz + d = 0;

[0033] Solve the plane parameters a, b, c, and d by the least squares method. The solution process is:

[0034] Select the ground point set {(x i , y i , z i)}, and construct matrix A and vector b;

[0035]

[0036] Solve for the plane parameters [a, b, d] by the least squares method:

[0037]

[0038] where c = -1, thereby performing ground plane correction on the point cloud data of the substation building.

[0039] Preferably, after the correction is completed, perform coordinate system conversion on the point cloud data of the substation building, including:

[0040] Rotate the normal vector [a, b, c] of the ground plane to the Z-axis direction, and define the rotation axis as the angle between the normal vector of the ground plane and the Z-axis. The rotation axis is obtained by cross product calculation, and the rotation angle θ is the angle that needs to be rotated, calculated by the angle between the normal vector and the Z-axis:

[0041]

[0042] Subsequently, rotate the point cloud data around the rotation axis to align the ground plane to the xy plane, completing the coordinate system conversion.

[0043] Preferably, in step S2, perform hierarchical analysis of the point cloud on the preprocessed point cloud data by the gradient segmentation method to obtain single-layer point cloud data, including:

[0044] Calculate the gradient of the point density distribution of the point cloud data in the height direction. The formula is:

[0045]

[0046] where G(z) is the gradient, dz is the change in height; D(z) is the point density at height z, calculated by the following formula:

[0047]

[0048] Determine the region with large gradient change through the calculated gradient G(z), and set a threshold T g as the floor demarcation point standard; when |G(z)| > T g then it is considered that a density mutation occurs at height z, defined as the hierarchical boundary of the substation building, and record its height z i as the floor demarcation point;

[0049] According to the obtained floor demarcation point z i , cut the point cloud data along the Z-axis to obtain single-layer point cloud data.

[0050] Preferably, in step S3, according to the obtained single-layer point cloud data, the single-layer point cloud data is segmented into a top plane and a side plane through normal vector analysis, including:

[0051] According to the single-layer point cloud data, for each point in each layer, a neighborhood with a radius of r is selected, and all the point cloud data within the neighborhood is regarded as the neighborhood point set of this point. Then, for each point p i the neighborhood point set is P i :

[0052] P i ={p j |d(p j ,p i )≤r};

[0053] where p j is any point in the point cloud data, and d(p j ,p i ) is the distance from point p j to point p i ;

[0054] For the neighborhood point set N i of each point, calculate the covariance matrix C:

[0055]

[0056] where μ is the centroid of the point cloud neighborhood point set N i , and the calculation formula is:

[0057]

[0058] Subsequently, perform eigenvalue decomposition on the covariance matrix to obtain eigenvalues and eigenvectors. The eigenvector corresponding to the smallest eigenvalue is the normal vector;

[0059] Determine the plane to which the calculated point belongs by calculating the angle between the normal vector and the Z-axis. The angle θ between the normal vector n = [n x ,n y ,n z and the Z-axis:

[0060]

[0061] If the normal vector is less than the threshold, the threshold is set to 10°, then this point belongs to the top surface; if the angle θ is approximately 90°, then this point belongs to the side plane.

[0062] Preferably, in step S4, project the top plane and side planes obtained after segmentation to generate multi-view projection images of the substation point cloud data, including: project the top plane and side planes obtained after segmentation in the directions of the three views and perform network sampling to obtain the top view, front view, and left view of the single-story substation building respectively.

[0063] Preferably, in step S5, based on the multi-view projection images, obtain the polygonal boundary of the substation building through contour extraction, measure and integrate the dimensional information of each layer of the substation building, including:

[0064] Based on the top view, front view, and left view obtained in step S4, measure the front view and left view through the improved Douglas-Peucker. Set the point set P = {p 1 , p 2 ,..., p n} on the plane, identify and extract all points approximately located at y = y min , take two endpoints of them as the initial point set, denoted as P base ;

[0065] For other points in P that do not belong to P base , apply the Douglas-Peucker algorithm to calculate the straight-line distance of the perpendicular distance from each point to the line p 0 p n :

[0066]

[0067] where × represents the cross product of vectors, and ||·|| represents the modulus of the vector. Find the point with the maximum distance; then determine the maximum distance point p max , and set the threshold ∈. If the distance d max of this point is greater than the set threshold ∈, then divide the polygon edge into p 0 p max , p max p n two parts, and then perform recursive processing on these two parts respectively; obtain the final point set P simplified ;

[0068] Finally, perform standard polygonal edge extraction on the top view of the building, measure according to the edges obtained from multiple views and verify each other, obtain the final measurement result of the single-layer point cloud data and output it.

[0069] According to the specific embodiments provided by the present invention, compared with the prior art, the present invention discloses the following technical effects:

[0070] (1) The present invention has high precision and multi-level discrimination ability: Through multi-view projection technology, the present invention can project and analyze point cloud data from multiple perspectives, effectively solving the problem of data confusion caused by complex structures and significantly improving the ability to distinguish different floors. Combining the analytic hierarchy process, the present invention can also gradually extract the geometric information of the building at different scales, accurately identify the structural details of each floor. This way of layer-by-layer analysis and segmentation, especially when dealing with multi-level buildings such as substation workshops, can avoid the problem of inaccurate layer division in traditional methods and achieve precise measurement.

[0071] (2) The method provided by the present invention has gradually improved the efficiency of processing complex buildings and also solved problems existing in existing methods, such as incomplete noise removal, detail loss caused by downsampling, and inaccurate height threshold segmentation. By optimizing the processing steps, the present invention ensures the efficient extraction and accurate measurement of multi-level substation workshops, providing more reliable technical support for fields such as automated inspection, facility maintenance, and engineering survey of substation workshops. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0073] Figure 1 It is a flowchart of a method for measuring multi-level substation workshops based on multi-view projection and analytic hierarchy process of the present invention;

[0074] Figure 2 It is a schematic diagram provided in Embodiment 1 of the present invention;

[0075] Figure 3 It is a gradient hierarchical analysis data diagram of substation workshop point cloud data provided in Embodiment 1 of the present invention;

[0076] Figure 4 It is a schematic diagram of separating the top plane and side plane of a substation workshop provided in Embodiment 1 of the present invention;

[0077] Figure 5 It is a measurement result diagram of the dimensions of a substation workshop provided in Embodiment 1 of the present invention; where, Figure 5 (a) is the dimension diagram of the substation workshop under the front view, Figure 5 (b) is the dimension diagram of the substation workshop under the top view, Figure 5 (c) is the dimension diagram of the substation workshop under the left view. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0078] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0079] To make the objectives, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0080] Embodiment 1

[0081] As Figure 1 and Figure 2 shown, the present invention provides a multi-layer substation plant measurement method based on multi-view projection and hierarchical analysis, including the following steps:

[0082] S1. Collect the point cloud data of the substation plant and preprocess the collected point cloud data;

[0083] S2. Perform hierarchical analysis of the point cloud on the preprocessed point cloud data by the gradient segmentation method to obtain single-layer point cloud data;

[0084] S3. According to the obtained single-layer point cloud data, divide the single-layer point cloud data into a top plane and a side plane through normal vector analysis;

[0085] S4. Project the top plane and the side plane obtained after segmentation to generate a multi-view projection image of the substation point cloud data;

[0086] S5. Based on the multi-view projection image, obtain the polygonal boundary of the substation plant through contour extraction, and measure and integrate the dimension information of each layer of the substation plant.

[0087] Among them, in step S1, the preprocessing of the collected point cloud data includes:

[0088] Denoise the point cloud data of the substation plant by a top-down method;

[0089] Perform ground plane correction on the denoised point cloud data of the substation plant;

[0090] After the correction is completed, perform coordinate system conversion on the point cloud data of the substation plant to facilitate height calculation.

[0091] In this embodiment, denoising the point cloud data of the substation plant by a top-down method includes:

[0092] Calculate the height range of the point cloud data:

[0093] z min = min(z i );

[0094] z max = max(z i );

[0095] where z min is the minimum height, z max is the maximum height, and z i is the height coordinate of the i-th point in the point cloud data;

[0096] Subsequently, set the height threshold z threshold as a reference, and mark the points above this threshold as potential noise points:

[0097] z threshold = z max - λ(z max - z min );

[0098] where λ is an adjustment coefficient, 0.01 ≤ λ ≤ 0.05;

[0099] Calculate the number of points N i in each layer of the point cloud data of the substation building, and set the density threshold N threshold . Points below this density are regarded as noise points. The formula is:

[0100]

[0101] where n is the total number of layers of the point cloud data, and α is an adjustment coefficient, 0.5 ≤ α ≤ 0.8; Subsequently, remove the low-density points, that is, points with N i < N threshold are marked as noise and removed.

[0102] Secondly, perform ground plane correction on the point cloud data of the substation building after denoising, including:

[0103] Extract the ground points of the point cloud data of the substation building by RANSAC, and determine the equation of the ground plane through the plane fitting method. Set the fitted plane equation as:

[0104] ax + by + cz + d = 0;

[0105] Solve the plane parameters a, b, c, and d by the least squares method. The solving process is:

[0106] Select the ground point set {(x i , y i , z i )}, and construct the matrix A and the vector b;

[0107]

[0108] Solve for the plane parameters [a, b, d] by the least squares method:

[0109]

[0110] where c = -1, thus performing ground plane correction on the point cloud data of the substation building.

[0111] In addition, after the correction is completed, perform coordinate system conversion on the point cloud data of the substation building, including:

[0112] Rotate the normal vector [a, b, c] of the ground plane to the Z-axis direction, and define the rotation axis as the angle between the ground plane normal vector and the Z-axis. The rotation axis is obtained by cross product calculation, and the rotation angle θ is the angle to be rotated, calculated by the angle between the normal vector and the Z-axis:

[0113]

[0114] Subsequently, rotate the point cloud data around the rotation axis to align the ground plane to the xy plane, completing the coordinate system conversion.

[0115] In step S2, perform hierarchical analysis of the point cloud on the preprocessed point cloud data by the gradient segmentation method to obtain single-layer point cloud data, including:

[0116] Calculate the gradient of the point density distribution of the point cloud data in the height direction. The formula is:

[0117]

[0118] where G(z) is the gradient, dz is the change in height; D(z) is the point density at height z, calculated by the following formula:

[0119]

[0120] Determine the region with large gradient change through the calculated gradient G(z), and set a threshold T g as the floor demarcation point standard; when |G(z)| > T g then it is considered that a density mutation occurs at height z, defined as the hierarchical boundary of the substation building, and record its height z i as the floor demarcation point;

[0121] According to the obtained floor demarcation point z i , cut the point cloud data along the Z-axis to obtain single-layer point cloud data, and the obtained result is as Figure 3 shown.

[0122] In step S3, according to the obtained single-layer point cloud data, the single-layer point cloud data is segmented into a top plane and a side plane through normal vector analysis, including:

[0123] According to the single-layer point cloud data, for each point in each layer, a neighborhood with a radius of r is selected, and all the point cloud data within the neighborhood is regarded as the neighborhood point set of this point. Then, for each point p i the neighborhood point set is P i :

[0124] P i = {p j |d(p j , p i ) ≤ r};

[0125] where p j is any point in the point cloud data, and d(p j , p i ) is the distance from point p j to point p i ;

[0126] For the neighborhood point set N i of each point, calculate the covariance matrix C:

[0127]

[0128] where μ is the centroid of the point cloud neighborhood point set N i , and the calculation formula is:

[0129]

[0130] Subsequently, perform eigenvalue decomposition on the covariance matrix to obtain eigenvalues and eigenvectors. The eigenvector corresponding to the smallest eigenvalue is the normal vector;

[0131] Judge the plane to which the calculated point belongs by calculating the angle between the normal vector and the Z-axis. The angle θ between the normal vector n = [n x , n y , n z and the Z-axis:

[0132]

[0133] If the normal vector is less than the threshold, and the threshold is set to 10°, then this point belongs to the top surface; if the angle θ is approximately 90°, then this point belongs to the side plane. The obtained result is as Figure 4 shown.

[0134] In step S4, project the top plane and side planes obtained after segmentation to generate multi-view projection images of the substation point cloud data, including: project the top plane and side planes obtained after segmentation in the direction of the three-view drawings and perform network sampling to obtain the top view, front view, and left view of the single-story substation building respectively.

[0135] According to step S4, in step S5, based on the multi-view projection images, obtain the polygonal boundary of the substation building through contour extraction, measure and integrate the dimensional information of each layer of the substation building, including:

[0136] Based on the top view, front view, and left view obtained in step S4, measure the front view and left view through the improved Douglas-Peucker. Set the point set P = {p 1 , p 2 ,..., p n} on the plane, identify and extract all points approximately located at y = y min , take two endpoints among them as the initial point set, denoted as P base ;

[0137] For other points in P that do not belong to P base , apply the Douglas-Peucker algorithm to calculate the straight-line distance of the perpendicular distance from each point to the straight line p 0 p n :

[0138]

[0139] where × represents the cross product of vectors, and ||·|| represents the modulus of the vector. Find the point with the maximum distance; then determine the maximum distance point p max , and set the threshold ∈. If the distance d max of this point is greater than the set threshold ∈, then divide the polygon edge into p 0 p max , p max p n two parts, and then perform recursive processing on these two parts respectively; obtain the final point set P simplified ;

[0140] Finally, perform standard polygon edge extraction on the top view of the building, measure according to the edges obtained from the multi-views and verify each other, obtain the final measurement results of the single-layer point cloud data and output. The obtained dimensional information results are as Figure 5 shown. Referring to Figure 5 (a), the dimensions of the substation building in the front view are 24.336 m in length and 10.65 m in height. Referring to Figure 5 (b), the dimensions of the substation building in the top view are 24.461 m in length and 3.223 m in width. Referring to Figure 5(c), the dimensions of the substation building in the left view are 10.897 m in height and 3.041 m in width. According to the final results, the differences in each dimension are not significant, thus achieving the accurate measurement of the multi-story substation building.

[0141] Therefore, by adopting the above method for measuring the multi-story substation building based on multi-view projection and hierarchical analysis, the efficient extraction and accurate measurement of the multi-story substation building are realized, and the problems existing in the existing methods, such as incomplete noise removal, detail loss caused by downsampling, and inaccurate height threshold segmentation, are solved.

[0142] In this article, specific examples are used to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, there will be changes in the specific implementation manner and application scope according to the idea of the present invention. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A multi-layer substation plant building calculation method based on multi-view projection and hierarchical analysis, characterized in that: The following steps are involved: S1. Collecting point cloud data of the substation plant and preprocessing the collected point cloud data; S2, performing a hierarchical analysis of the point cloud data after preprocessing by a gradient segmentation method to obtain single-layer point cloud data; S3. According to the obtained single-layer point cloud data, the single-layer point cloud data is divided into a top plane and a side plane by normal vector analysis; S4, projecting the top plane and the side plane obtained after segmentation to generate a multi-view projection image of the substation cloud data; S5. Based on the multi-view projection images, the polygonal boundary of the substation building is obtained by contour extraction, and the size information of each layer of the substation building is measured and integrated.

2. A multi-layer substation plant building measurement method based on multi-view projection and hierarchical analysis according to claim 1, characterized in that: In step S1, preprocessing the collected point cloud data includes: A top-down approach is used to denoise the point cloud data of the substation building; Perform ground plane correction on the substation plant point cloud data after denoising; After the calibration is completed, the coordinate system of the substation building point cloud data is converted to facilitate the calculation of the height.

3. A multi-layer substation plant building measurement method based on multi-view projection and hierarchical analysis according to claim 2, characterized in that: The top-down method is used to denoise the substation plant point cloud data, including: Calculate the height range of point cloud data: With min =min(of i ); With max =max(from i ); Among them, z min is the minimum height, z max is the maximum height, z i is the height coordinate of the i-th point in the point cloud data; Then set the height threshold z threshold As a benchmark, points above this threshold are marked as possible noise points: With threshold =from max -λ(z max -With min ); Among them, λ is the adjustment coefficient, 0.01≤λ≤0.05; Calculate the number of point clouds N in each layer of the substation plant point cloud data i , and set the density threshold N threshold , points below this density are considered noise points, and the formula is: Among them, n is the total number of point cloud data layers, α is the adjustment coefficient, 0.5≤α≤0.8; then the low-density points are removed, that is, N i <N threshold The points are marked as noise and removed.

4. A multi-layer substation plant building measurement method based on multi-view projection and hierarchical analysis according to claim 3, characterized in that: The ground plane correction is performed on the point cloud data of the substation building after denoising, including: The ground points of the substation building point cloud data are extracted by RANSAC, and the equation of the ground plane is determined by the plane fitting method. The fitting plane equation is set as: ax+by+cz+d=0; The plane parameters a, b, c and d are solved by the least squares method. The solution process is: Select the ground point set {(x i ,y i , z i )}, and construct matrix A and vector b; Solve for the plane parameters [a, b, d] by the least squares method: Wherein, c=-1, so that the ground plane correction is performed on the point cloud data of the substation building.

5. A multi-layer substation plant building measurement method based on multi-view projection and hierarchical analysis according to claim 4, characterized in that: After the calibration is completed, the coordinate system of the substation plant point cloud data is converted, including: The normal vector [a, b, c] of the ground plane is rotated to the Z axis direction, and the rotation axis is defined as the angle between the normal vector of the ground plane and the Z axis. The rotation axis is calculated by the cross product, and the rotation angle θ is the angle to be rotated, which is calculated by the angle between the normal vector and the Z axis: Subsequently, the point cloud data is rotated around the rotation axis to align the ground plane to the xy plane, completing the conversion of the coordinate system.

6. A multi-layer substation plant building measurement method based on multi-view projection and hierarchical analysis according to claim 1, characterized in that: In step S2, the pre-processed point cloud data is subjected to a layered analysis of the point cloud by using a gradient segmentation method to obtain single-layer point cloud data, including: The gradient of the point density distribution of the point cloud data in the height direction is calculated using the formula: Where G(z) is the gradient, dz is the change in height, and D(z) is the point density at height z, calculated by the following formula: By calculating the gradient G(z), we can determine the area with large gradient changes and set the threshold T g As the floor demarcation point standard; when |G(z)|>T g When , it is considered that a density mutation occurs at height z, which is defined as the hierarchical boundary of the substation building, and its height z is recorded. i As a floor dividing point; According to the obtained floor dividing point z i , cut the point cloud data along the Z axis to obtain a single layer of point cloud data.

7. A multi-layer substation plant building measurement method based on multi-view projection and hierarchical analysis according to claim 1, characterized in that: In step S3, according to the obtained single-layer point cloud data, the single-layer point cloud data is segmented into a top plane and a side plane by normal vector analysis, including: According to the single-layer point cloud data, for each point in each layer, a neighborhood with a radius of r is selected, and all point cloud data in the neighborhood are regarded as the neighborhood point set of the point. Then, for each point p i The neighborhood point set is P i : P i ={p j |d(p j ,p i )≤r}; Among them, p j is any point in the point cloud data, d(p j , p i ) is point p j To point p i distance; For each point's neighborhood point set N i , calculate the covariance matrix C: Among them, μ is the point cloud neighborhood point set N i The centroid of is calculated as: Then, the covariance matrix is ​​decomposed into eigenvalues ​​to obtain eigenvalues ​​and eigenvectors, where the eigenvector corresponding to the smallest eigenvalue is the normal vector; The plane to which the calculation point belongs is determined by calculating the angle between the normal vector and the Z axis. The normal vector n = [n x , n y , n z ] and the Z axis: If the normal vector is smaller than a threshold value, which is set to 10°, the point belongs to the top surface; if the angle θ is approximately 90°, the point belongs to the side plane.

8. The method for calculating the plant building of a multi-layer substation based on multi-view projection and hierarchical analysis according to claim 1 is characterized in that: In step S4, the top plane and side plane obtained after segmentation are projected to generate a multi-view projection image of the substation cloud data, including: projecting and network sampling the top plane and side plane obtained after segmentation in the three-view direction to obtain the top view, front view and left view of the single-layer substation building respectively.

9. A multi-layer substation plant building measurement method based on multi-view projection and hierarchical analysis according to claim 8, characterized in that: In step S5, based on the multi-view projection images, the polygonal boundary of the substation building is obtained by contour extraction, and the size information of each layer of the substation building is measured and integrated, including: Based on the top view, front view and left view obtained in step S4, the front view and left view are measured by the improved Douglas-Peucker, and the point set P on the plane is set to be {p1, p2, ..., p n }, identify and extract all the min points, take two of them as the initial point set, denoted as P base ; For P that does not belong to P base For other points, apply the Douglas-Peucker algorithm to calculate the straight line p0p point by point n The straight-line distance of the vertical distance: Among them, × represents the cross product of the vector, and ||·|| represents the modulus of the vector. Find the point with the largest distance; then determine the maximum distance point p max , and set the threshold ∈, if the point is away from d max If it is greater than the set threshold ∈, the polygon edge is split into p0p max , p max p n Two parts, then recursively process these two parts respectively; get the final point set P simplified ; Finally, standard polygon edge extraction is performed on the top view of the factory building, and the edges of multiple views are measured and verified with each other to obtain the final measurement results of the single-layer point cloud data and output them.

Images