Rock mass structural surface roughness measuring method based on three-dimensional laser

Abstract

The invention discloses a rock mass structural surface roughness measuring method based on three-dimensional laser. The method comprises the following steps: firstly, acquiring a rock surface point cloud through high-precision laser scanning, and fitting a reference plane by using a PCA algorithm after denoising classification; secondly, constructing a projection plane perpendicular to the measuring line, and extracting the length L of the measuring line; and finally, adaptively sampling according to the point cloud resolution, calculating a roughness index delta through discrete integration, and converting the roughness index delta into a JRC value. The method solves the problems that manual measurement is high in subjectivity and an existing digital method neglects anisotropy, the precision reaches 0.1 mm level, and the efficiency is improved by 10 times. The method is suitable for stability evaluation of tunnel, slope and other rock mass projects.

Description

Technical Field

[0001] This patent relates to the field of rock engineering investigation, and specifically to a method for measuring the roughness of rock structure surfaces based on three-dimensional laser. Background Art

[0002] Rock mass stability determines the safety and construction costs of geotechnical engineering construction. The roughness coefficient of rock surface is a key parameter that determines the shear strength and stability of the surface. Due to its simplicity and ease of use, it has been widely used in engineering. However, current methods and approaches for characterizing the JRC of rock surface structures are still relatively backward. The most commonly used method is manual precision measurement. This involves collecting rock blocks on-site, depicting the surface roughness on drawings, and then using the four standard surface contour lines in the structural surface correspondence diagram proposed by Patton to determine the approximate range of the JRC. However, this manual measurement method is time-consuming and labor-intensive, and is prone to subjective misjudgment, limiting the accuracy and efficiency of geotechnical engineering construction testing. Various JRC characterization methods have been proposed, including statistical parameter characterization, fractal dimension characterization, and composite parameter methods. However, due to measurement limitations, the results suffer from large precision deviations and low engineering practicality.

[0003] To achieve the detailed characterization and objective measurement of the surface roughness of rock mass structural surfaces, two aspects need to be addressed. The first is the high-precision acquisition of the surface undulation of the structural surface, and the second is the use of quantitative system processes and indicators to characterize the undulating characteristics of JRC. Currently, the accuracy of contact measurement is affected by factors such as the hardness, size, and contact movement speed of the probe exit point. Therefore, it has disadvantages such as low measurement accuracy and structural surface damage caused by probe intrusion. In terms of the statistical process and quantitative evaluation indicators of the JRC system, including the undulation parameter method, fractal dimension method, and three-dimensional layer method, the existing evaluation methods ignore the influence of factors such as sampling direction, sampling scale, and sampling spacing on the roughness evaluation results. The operation process is cumbersome and the repeatability is poor. In addition, most methods only evaluate roughness from the perspective of geometric information transformation and mathematical relationships, ignoring the anisotropy and size effect of roughness. Summary of the Invention

[0004] In response to the above situation, the present invention provides a method for measuring the roughness of rock structure surfaces based on three-dimensional laser. The invention of this application mainly combines the advantages of high-precision acquisition of three-dimensional laser scanning data, and adopts a unified system processing method to obtain evaluation indicators for evaluating JRC, thereby being able to objectively, accurately and truly obtain the roughness of rock structure surfaces.

[0005] In order to achieve the above object, the present invention provides a method for measuring the roughness of a rock structure surface based on three-dimensional laser, which is characterized by comprising the following steps:

[0006] S1, using a 3D laser scanner to scan the rock mass structure surface to be measured and obtain point cloud data containing spatial coordinates;

[0007] S2, performing denoising, classification, and cropping on the point cloud data in sequence to extract valid point cloud coordinates of the structural surface to be measured;

[0008] S3, based on the valid point cloud coordinates, determine the longest survey line direction or the orthogonal survey line direction;

[0009] S4, using principal component analysis to fit the reference plane of the structural surface, and obtaining the normal vector and boundary coordinates of the fitting plane;

[0010] S5, generating a projection surface perpendicular to the fitting plane and containing the survey line, and establishing an equation of the projection surface;

[0011] S6, solve the intersection line of the fitting plane and the projection plane, and determine the length L of the curve projection line;

[0012] S7, based on the point cloud resolution, select points within a certain range near the projection surface as sampling points;

[0013] S8, calculate the horizontal distance X from the sampling point to the survey line n and vertical distance Z n ;

[0014] S9, calculate the roughness δ by the rate of change of horizontal distance and vertical distance;

[0015] S10, substituting δ into the preset fitting formula to calculate the rock mass structural surface roughness coefficient JRC.

[0016] Furthermore, the three-dimensional laser scanner in step S1 is a ground-based, handheld or mobile device with millimeter-level accuracy. The scanning parameters include a resolution of ≤5 mm and a scanning range covering the structural surface and its surrounding area of ​​≥10 cm.

[0017] Furthermore, the specific steps of step S2 are as follows: first, the point cloud denoising algorithm is used to denoise the point cloud; second, the ground point classification is completed by using the progressive encryption triangulation filtering algorithm for the denoised point cloud; then, according to the extracted ground points, the point cloud classification results of rock mass, vegetation, ground, etc. are extracted respectively according to the elevation interpolation, and finally, the surface point cloud coordinates of the structural surface to be measured are obtained by clipping them respectively according to the position of the structural surface to be measured.

[0018] Furthermore, the direction of the longest survey line in step S3 can be calculated using the following formula:

[0019]

[0020] L max =max{L1,1 ,L 1,2 ,…L ij} (2)

[0021] L max =L ab (3)

[0022] p1=(x a ,y a ,z a ) (4)

[0023] p2=(x b ,y b ,z b ) (5)

[0024]

[0025] where x i ,y i ,z i is the x, y, z coordinates of the i-th point cloud; y j ,y j ,z j is the x, y, z coordinates of the j-th point cloud; L max is the longest line length connecting all point clouds, p1 and p2 are the coordinates of the point clouds connecting the longest line length, is the direction vector of p1, p2, is the unit vector of the longest survey line, then the three-dimensional coordinate equation of the longest survey line can be obtained by formula 8:

[0026]

[0027] Furthermore, the directions of the orthogonal survey lines in S3 can be obtained by the following method: the longest survey line direction is selected as the north-south direction, and the other three directions of east-west, northeast-southwest, and northwest-southeast are measured at intervals of 45°, which are called the four orthogonal survey line directions.

[0028] Furthermore, the specific steps of step S4 are as follows: first, the covariance matrix of all point clouds of the structural surface to be measured must be obtained, and then the three eigenvalues ​​and corresponding three eigenvectors of the covariance matrix must be obtained; the eigenvalues ​​and eigenvectors are obtained by singular value decomposition, and the eigenvectors corresponding to the first two principal components are taken to construct the orthogonal direction vectors of the fitting plane; the convex hull boundary of the two-dimensional projection of the point cloud is extracted using the Alphashape algorithm, and then the two-dimensional convex hull boundary coordinates are restored and converted into three-dimensional coordinates to obtain the three-dimensional coordinate boundary of the fitting plane.

[0029] Furthermore, the method for establishing the projection surface equation in step S5 includes:

[0030] Using the two end points p1(xa ,y a ,z a ), p2(x b ,y b ,z b ) and the coordinates of the fitted plane normal vector u3(x u3 ,y u3 ,z u3 ) to perform cross multiplication to obtain the projection surface normal vector, and the projection surface normal vector F can be calculated by formula (17): norm ;

[0031] F norm =[(x a -x b ) (y a -y b ) (z a -z b )]×[x u3 y u3 z u3 ] (17)

[0032]

[0033] By projecting the surface normal vector F norm (x f ,y f ,z f ) and the point p1(x a ,y a ,z a ) The plane equation of the projection surface is calculated using formula (18).

[0034] Furthermore, the specific steps of step S6 are to obtain the equations by fitting the plane equation parameters A1, B1, C1, D1 and the projection plane equation parameters A2, B2, C2, D2. The specific calculation process is shown in formula (19-20):

[0035] A1,B1,C1=x u3 ,y u3 ,z u3 (19)

[0036] D1=-(x u3 ×x a +y u3 ×y a +z u3 ×z a ) (20)

[0037] A2,B2,C2=x f ,y f ,z f (twenty one)

[0038] D2=-(x f ×x a +y f ×y a +z f ×z a ) (twenty two)

[0039]

[0040] Set z=z a and z=z b Substitute them into equation (23) and solve them to obtain the two endpoints p3(x c ,y c ,z a ) and p4(x d ,y d ,z b ); the projected measurement line length L can be solved by formula (24).

[0041] Furthermore, in step S8, the horizontal distance X from the sampling point to the survey line is calculated. n and vertical distance Z n Specifically:

[0042] The endpoint of the projection line p3(x c ,y c ,z a ) and p4(x d ,y d ,z b ) as the reference line, and calculate the horizontal distance X of the nth sampling point cloud on the reference orientation. n and vertical distance Z n , where n∈λ, then calculate X according to formula (28-33) n ,Z n .

[0043] S1,S2,S3=(x c -x d ),(y c -y d ),(z a -z b ) (28)

[0044]

[0045] Among them (α n , β n , γ n ) represents the projection coordinates of the nth point cloud on the projection line.

[0046] Furthermore, the calculation method of the roughness δ is:

[0047]

[0048] Where L is the curve projection length, N is the number of points on the line segment, and x n is the horizontal distance of the nth sampling point, Z n is the vertical distance of the nth sampling point;

[0049] The fitting formula of the body surface roughness coefficient JRC in step S10 is:

[0050] JRC=31.28+33.54log 10 δ.

[0051] The present invention proposes a method for measuring the roughness of rock mass structural surfaces based on three-dimensional laser, which has the following advantages:

[0052] (1) The present invention constructs an efficient measurement method for rock surface roughness based on three-dimensional laser scanning. Through outdoor high-precision point cloud acquisition and indoor full-digital rapid analysis, the subjective error caused by manual measurement can be reduced. At the same time, since the entire indoor processing process is spatial geometric transformation calculation, except for the accuracy error of the scanner itself, the rest of the process has no random error, which can not only improve the efficiency of field work, but also improve the accuracy of structural surface roughness measurement.

[0053] (2) The roughness calculation method proposed in this invention fully considers the optimal survey line direction, fitting reference plane and sampling distance under different survey line conditions, so that the measurement results are more in line with the actual situation, and can provide a technical basis for actual engineering surveys. It has important practical significance for further promoting the digital construction and intelligent evaluation of geological surveys.

[0054] (3) The present invention uses three-dimensional laser point cloud as basic data to realize the roughness measurement of rock structure surface. Compared with traditional manual precision measurement method, photogrammetry and other technologies, it can maximize the accuracy of data acquisition and can be used under conditions without light source. In particular, it provides a new means for measuring the roughness of structure surface in complex and light-free underground spaces such as lanes, tunnels, and exploration adit. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a block diagram of the method principle of the present invention.

[0056] Figure 2 This is a workflow diagram of the ground three-dimensional laser scanning of the present invention.

[0057] Figure 3 is a schematic diagram of the point cloud preprocessing results.

[0058] Figure 4 is a schematic diagram of the survey line layout direction.

[0059] FIG5 is a schematic diagram of structural surface plane fitting based on PCA principal component analysis.

[0060] FIG6 is a schematic diagram of obtaining a projection surface in which the measurement lines are coplanar and perpendicular to the fitting plane.

[0061] Figure 7 It is a schematic diagram for obtaining the curve projection measurement line length L.

[0062] Figure 8 The blue points shown are the schematic diagram of the point cloud selected for roughness calculation. DETAILED DESCRIPTION

[0063] The present invention will be further described in detail below with reference to specific embodiments and accompanying drawings. The following embodiments are not intended to limit the present invention but are merely intended to illustrate the present invention.

[0064] In the specific embodiments, steps, material selections, and numerical parameters that are not described in detail are all conventional choices in the prior art or any currently disclosed prior art.

[0065] The technical problem to be solved by the present invention is to provide a method for measuring the roughness of rock structure surfaces based on three-dimensional laser. Based on the precise acquisition of three-dimensional laser scanning data, the coefficients obtained can accurately reflect the roughness of the rock structure surfaces and the stability of the rock mass. The roughness of the rock structure surfaces can be objectively, accurately and truly obtained, providing a basis for rock engineering surveys and subsequent related designs.

[0066] In order to solve the above technical problems, the technical solution adopted in this implementation is a rock structure surface roughness measurement method based on three-dimensional laser, a rock surface coordinate information collection method based on three-dimensional laser scanning, and an indicator acquisition method of the vertical transformation standard deviation of the PCA fitting plane method. It solves the problem of quantitative evaluation of roughness based on three-dimensional point cloud and adopts a unified system processing method to obtain the evaluation index of JRC.

[0067] See also Figure 1, this embodiment includes two major processing steps: Step 1: 3D laser scanning data acquisition (outdoor acquisition), Step 2: Roughness characteristic index calculation (indoor processing and analysis). The processing order of each module is progressive, wherein: Step 1 first selects the position of the structural surface to be measured in the test area through a 3D laser scanner, and then performs point cloud denoising, point cloud classification, and point cloud cropping preprocessing on the collected 3D point cloud data, and inputs the processed point cloud data into Step 2; In Step 2, based on the coordinate information of the structural surface point cloud to be measured, first select the optimal survey line direction and use it to count the degree of structural surface undulation near the survey line, then use PCA principal component analysis to obtain the fitting plane of all structural surface point clouds, and then obtain the projection surface equation perpendicular to the fitting plane and coplanar with the survey line, and then obtain the intersection line of the projection surface and the fitting surface to determine the survey line length L, and further select the appropriate sampling point distance, and count the horizontal distance x from the sampling point to the survey line. n and vertical distance z n Finally, the roughness evaluation index δ is calculated and brought into the JCR fitting formula to obtain the roughness index of the rock mass structural surface to be tested.

[0068] The structural surface roughness measurement method of this invention is carried out in sequence according to the above two steps and 11 modules. The specific detailed steps are as follows (unless otherwise noted in the detailed operation process of this embodiment, the rest are processed using the Python programming language).

[0069] See also Figure 2 , in 3D laser scanning data acquisition:

[0070] ① 3D laser scanner selection. 3D laser scanners can be categorized by operating method: ground-based, airborne, handheld, and mobile. Ground-based 3D laser scanners mostly use pulsed ranging, offering the highest accuracy and ease of operation, making them more suitable for the data acquisition tasks of this invention. If conditions prohibit, handheld or mobile laser scanners can be used as substitutes. However, please note that any laser scanner hardware must achieve mm-level acquisition accuracy to be used for roughness measurement.

[0071] ② Selection of test rock surface. For test rock surfaces, try to select a large and generally flat surface, and a fresh surface. Avoid uneven or fragmented structures caused by weathering and unloading. This way, the collected data can best reach the theoretical measurement value of the roughness of the structural surface.

[0072] ③ Collection of point cloud data on the surface of rock mass structure. The work of ground-based 3D laser scanning can be mainly divided into: scanning preparation, scanning site selection, target layout, etc.

[0073] (1) Site survey and preparation of the survey area (2) Site selection (3) Centering, leveling and scanning parameter setting

[0074] Except for the specific instructions in the above steps, other technical requirements shall refer to the relevant requirements of CH / Z 3017-2015 "Technical Specifications for Terrestrial 3D Laser Scanning Operations".

[0075] ④Point cloud data preprocessing

[0076] 3D point cloud (Points) preprocessing: The overall processing flow is "filtering denoising - point cloud classification - point cloud clipping". First, the point cloud denoising algorithm is used to denoise the point cloud. Second, the ground point classification is completed using the progressively dense triangulated network filtering algorithm on the denoised point cloud. Then, based on the extracted ground points and using elevation interpolation, the point cloud classification results for rock mass, vegetation, and ground are extracted. Finally, the surface point cloud of the structural surface to be measured is clipped according to the location of the structural surface to be measured. The final result is shown in Figure 3.

[0077] In the calculation of roughness characteristic index:

[0078] Select the survey line direction. After obtaining the three-dimensional point cloud coordinates of the rock surface to be measured in step 1, obtain survey lines located in different directions in the point cloud. Generally, the line connecting the two point clouds with the longest distance is selected as the survey line, which is called the longest survey line direction, as shown in Figure 5-a. The longest survey line direction of the structural surface roughness can be obtained by the following formula (1-7):

[0079]

[0080] L max =max{L 1,1 ,L 1,2 ,...L ij} (2)

[0081] L max =L ab (3)

[0082] p1=(x a ,y a ,z a ) (4)

[0083] p2=(x b ,y b ,z b ) (5)

[0084]

[0085] where x i ,y i ,z i is the x, y, z coordinates of the i-th point cloud; x j ,y j ,z jis the x, y, z coordinates of the j-th point cloud. max is the longest line length connecting all point clouds, p1 and p2 are the coordinates of the point clouds connecting the longest line length, is the direction vector of p1, p2, is the unit vector of the longest survey line. Then the three-dimensional coordinate equation of the longest survey line can be obtained by formula 8:

[0086]

[0087] In addition, because the roughness parameters of structural surfaces are anisotropic and random, for highly nonuniform structural surfaces, the longest survey line direction can be selected as the north-south direction, and three other directions, east-west, northeast-southwest, and northwest-southeast, can be measured at intervals of 45°. These are referred to as the four orthogonal survey line directions. Steps ②-⑧ are repeated for each survey line to measure its roughness index, as shown in Figure 4. In the survey line selection method of the present invention, due to the nonuniformity of the rock surface, the roughness measurement results of small areas may be inaccurate. On a complete rock surface, the longer the sampling line and the more samples of the undulating surface are collected, the more noise-resistant and accurate the measurement results are.

[0088] ②Point cloud fitting plane based on PCA

[0089] This step mainly uses the PCA principal component analysis method to realize the plane fitting of the structural surface point cloud and obtain the normal vector and range coordinates of the fitting plane.

[0090] First, we need to obtain the covariance matrix of all point clouds of the structural surface to be measured, and then obtain the three eigenvalues ​​of the covariance matrix and the corresponding three eigenvectors. The specific calculation is shown in the following formula:

[0091]

[0092] x ∈ ,y ∈ ,z ∈ =max(x i )-min(x i ),max(y i )-min(y i ),max(z i )-min(z i ) (10)

[0093]

[0094] A=U·S·V T (13)

[0095] Where U is an m×m orthogonal matrix (left singular vector matrix), S is an m×n matrix, the elements on its diagonal are called singular values, and the rest of the elements are zero, V T is an n×n orthogonal matrix (right singular vector matrix); m, n are the number of point clouds and the dimension (n=3); x ∈ ,y ∈ ,z ∈ The extreme value differences of the point cloud in the x, y, and z coordinates are obtained from formula 13. The three eigenvalues ​​S (arranged from large to small) of the structure surface point cloud and the corresponding eigenvector U = [u1, u2, u3] can be obtained. Therefore, the two orthogonal direction vectors U of the point cloud fitting plane can be obtained by taking only the first two columns of U. s .

[0096] U s =[u1,u2] (14)

[0097] Secondly, it is necessary to obtain the two-dimensional projection coordinates of all three-dimensional point cloud coordinates on the fitting plane, which can be obtained by formula (15):

[0098] {(x 2d ,y 2d )}=φ·U s (15)

[0099] Get the two-dimensional projection of the point cloud of the fitting plane {(x 2d ,y 2d After the point cloud is triangulated, the convex hull boundary of this set of points can be obtained using the Alpha shape algorithm. This algorithm is based on the triangulation and the introduction of a parameter α. For each edge in the point cloud triangulation, if the radius of the circumscribed circle of the corresponding triangle is less than α, the edge belongs to the convex hull boundary of the point cloud. In this invention, α is generally set to 2. The obtained convex hull boundary is shown in Figure 5. This step can be implemented using the open-source Python library "alphashape".

[0100] Finally, the two-dimensional convex hull boundary coordinates can be restored and converted into three-dimensional coordinates using formula (16) to obtain the three-dimensional coordinate point boundary of the fitting plane.

[0101]

[0102] where x 3d ,y 3d ,z 3d are the boundary coordinate points of the convex hull of the transformed structural surface, x 2di y 2di are the horizontal and vertical coordinates of the i-th two-dimensional convex hull coordinate boundary point respectively.

[0103] ③ Obtain a projection surface that is coplanar with the measurement line and perpendicular to the fitting plane

[0104] This step mainly uses the fitting plane and the spatial coordinates of the longest survey line to generate a projection plane that is coplanar with the survey line and perpendicular to the fitting plane. The specific implementation method is as follows:

[0105] Determine the normal vector of the projection surface. Use the two end points p1(x a ,y a ,z a ), p2(x b ,y b ,z b ) and the coordinates of the fitted plane normal vector u3(x u3 ,y u3 ,z u3 ) to perform cross multiplication to obtain the projection surface normal vector, and the projection surface normal vector F can be calculated by formula (17): norm .

[0106] F norm =[(x a -x b ) (y a -y b ) (z a -z b ])×[x u3 y u3 z u3 ] (17)

[0107] Get the plane equation of the projection surface. Use the projection surface normal vector F norm (x f ,y f ,z f ) and the point p1(x a ,y a ,z a ) Calculate the plane equation of the projection surface, which can be calculated using formula (18).

[0108]

[0109] ④ Obtain the curve projection line length L

[0110] The length of the projected curve is the intersection of the projection plane and the fitted plane. This can be obtained by using the simultaneous equations of the fitted plane parameters A1, B1, C1, and D1 and the projection plane parameters A2, B2, C2, and D2. The specific calculation process is shown in Formulas (19-20).

[0111] A1,B1,C1=x u3 ,y u3 ,z u3 (19)

[0112] D1=-(xu3 ×x a +y u3 ×y a +z u3 ×z a ) (20)

[0113] A2,B2,C2=x f ,y f ,z f (twenty one)

[0114] D2=-(x f ×x a +y f ×y a +z f ×z a ) (twenty two)

[0115]

[0116] Set z=z a and z=z b Substitute them into equation (23) and solve them to obtain the two endpoints p3(x c ,y c ,z a ) and p4(x d ,y d ,z b The projected line length L can be solved by formula (24). The obtained curve projected line length L is as follows: Figure 7 shown.

[0117]

[0118] ⑤Select sampling points

[0119] Select the point cloud on the projected survey line for roughness calculation. Since rock mass roughness refers to the degree of undulation on the rock surface, it is necessary to obtain the coordinates of the point cloud that floats above and below the fitting plane and lies on the survey line. However, due to the acquisition accuracy principle of the laser scanner, not all point clouds actually lie completely on the straight line where the survey line lies. Therefore, it is necessary to select points within a certain distance h near the projection surface as the data for calculating the roughness of the survey line. In this invention, the sampling point selection is to capture the point cloud coordinates on the survey line. However, the actual point cloud has a certain resolution, so it is impossible for all the measured point clouds to be exactly located on the survey line. Therefore, a certain distance from the survey line is selected to capture the point cloud near the survey line. If this capture distance is too large, the selected point cloud will not be near the survey line and will not represent the degree of undulation near the survey line. If it is too small, sufficient point cloud samples of the survey line cannot be collected, resulting in a high degree of randomness in the test results. Therefore, a point cloud resolution of 0.001m is selected to capture the point cloud near the survey line. If the point cloud resolution is 0.001m, h = 0.001m in this invention. To this end, first calculate the distance between all the point clouds to be measured and the projection surface according to formula (25-26).

[0120]

[0121] where x i ,y i ,z i is the x, y, z coordinates of the i-th point cloud, i represents the i-th point cloud to be measured, i=1,2....m, m is the number of all point clouds. ξ i is the absolute value of the distance between the i-th point cloud and the projection surface. Then the point cloud sequence dataset λ selected as the sampling point can be expressed as shown in formula 27, as shown in Figure 8 The blue points shown are the point clouds selected for roughness calculation.

[0122] λ={{i|ξ i ≤h}} (27)

[0123] ⑥Calculate the horizontal distance X from the sampling point to the survey line n and vertical distance Z n

[0124] The endpoint of the projection line p3(x c ,y c ,z a ) and p4(x d ,y d ,z b ) as the reference line, and calculate the horizontal distance X of the nth sampling point cloud on the reference orientation. n and vertical distance Z n , where n∈λ. First, calculate X according to formula (28-33) n ,Zn .

[0125] S1,S2,S3=(x c -x d ),(y c -y d ),(z a -z b ) (28)

[0126]

[0127]

[0128] Among them (α h , β n , γ n ) represents the projection coordinates of the nth point cloud on the projection line.

[0129] ⑦Calculate the roughness δ

[0130] The undulation δ represents the degree of concavity and convexity of the rock surface, which can be expressed by the rate of change of the vertical distance in the horizontal direction. The X obtained in step ⑥ is n With Z n , the fluctuation δ is calculated using formula (34).

[0131]

[0132] Where L is the curve projection length, N is the number of points on the line segment, and x n is the horizontal distance of the nth sampling point, Z n is the vertical distance of the nth sampling point.

[0133] ⑧ Obtain rock mass structural surface roughness JRC

[0134] The JRC fitting formula for the fluctuation index δ has been extensively researched by scholars. The fitting formula obtained by the applicant of the present invention through indoor experiments and comparative analysis of point cloud roughness collected at a resolution of 1 mm is as follows:

[0135] JRC=31.28+33.54log 10 δ (35)

[0136] Substituting the δ calculated in step 7 into formula (35) can obtain the roughness JRC value on the measuring line.

[0137] The above detailed description is a specific description of one feasible embodiment of the present invention. This embodiment is not intended to limit the patent scope of the present invention. Any equivalent implementation or modification that does not depart from the present invention should be included in the scope of the technical solution of the present invention.

Claims

1. A method for measuring the roughness of rock structure surface based on three-dimensional laser, characterized in that: The following steps are involved: S1, using a 3D laser scanner to scan the rock mass structure surface to be measured and obtain point cloud data containing spatial coordinates; S2, performing denoising, classification, and cropping on the point cloud data in sequence to extract valid point cloud coordinates of the structural surface to be measured; S3, based on the valid point cloud coordinates, determine the longest survey line direction or the orthogonal survey line direction; S4, using principal component analysis to fit the reference plane of the structural surface, and obtaining the normal vector and boundary coordinates of the fitting plane; S5, generating a projection surface perpendicular to the fitting plane and containing the survey line, and establishing an equation of the projection surface; S6, solve the intersection line of the fitting plane and the projection plane, and determine the length L of the curve projection line; S7, based on the point cloud resolution, select points within a certain range near the projection surface as sampling points; S8, calculate the horizontal distance X from the sampling point to the survey line n and vertical distance Z n ; S9, calculate the roughness δ by the rate of change of horizontal distance and vertical distance; S10, substituting δ into the preset fitting formula to calculate the rock mass structural surface roughness coefficient JRC.

2. The method according to claim 1, characterized in that The three-dimensional laser scanner in step S1 is a ground-based, handheld or mobile device with millimeter-level accuracy. The scanning parameters include a resolution of ≤5 mm and a scanning range covering the structural surface and its surrounding area of ​​≥10 cm.

3. The method according to claim 1, characterized in that The specific steps of step S2 are as follows: first, the point cloud is denoised using a standard deviation denoising algorithm; second, the ground point classification is completed using a progressive encryption triangulation filtering algorithm for the denoised point cloud; then, based on the extracted ground points, elevation interpolation is performed to extract the point cloud classification results of rock mass, vegetation, ground, etc., and finally, the surface point cloud coordinates of the structural surface to be measured are obtained by clipping them separately according to the position of the structural surface to be measured.

4. The method according to claim 1, wherein The direction of the longest survey line in step S3 can be calculated using the following formula: L max =max{L 1,1 ,L 1,2 ,…L ij } (2) L max =L ab (3) p1=(x a ,y a ,from a ) (4) p2=(x b ,y b ,z b ) (5) where x i ,y i ,z i is the x, y, z coordinates of the i-th point cloud; x j ,y j ,z j is the x, y, z coordinates of the j-th point cloud; L max is the longest line length connecting all point clouds, p1 and p2 are the coordinates of the point clouds connecting the longest line length, is the direction vector of p1, p2, is the unit vector of the longest survey line, then the three-dimensional coordinate equation of the longest survey line can be obtained by formula 8:

5. The method according to claim 4, characterized in that The directions of the orthogonal survey lines in S3 can be obtained by the following method: the longest survey line direction is selected as the north-south direction, and the other three directions of east-west, northeast-southwest, and northwest-southeast are measured at intervals of 45°. These are called the four orthogonal survey line directions.

6. The method according to claim 1, characterized in that The specific steps of step S4 are as follows: first, the covariance matrix of all point clouds of the structural surface to be measured is obtained, and then the three eigenvalues ​​and corresponding three eigenvectors of the covariance matrix are obtained; the eigenvalues ​​and eigenvectors are obtained by singular value decomposition, and the eigenvectors corresponding to the first two principal components are taken to construct the orthogonal direction vectors of the fitting plane; the convex hull boundary of the two-dimensional projection of the point cloud is extracted by the Alpha shape algorithm, and then the two-dimensional convex hull boundary coordinates are restored and converted into three-dimensional coordinates to obtain the three-dimensional coordinate boundary of the fitting plane.

7. The method according to claim 1, characterized in that The method for establishing the projection surface equation in step S5 includes: Using the two end points p1(x a ,y a ,z a ), p2(x b ,y b ,z b ) and the coordinates of the fitted plane normal vector u3(x u3 ,y u3 ,z u3 ) to perform cross multiplication to obtain the projection surface normal vector, and the projection surface normal vector F can be calculated by formula (17): norm ; F norm =[(x a -x b )(y a -y b )(z a -z b )]×[x u3 y u3 z u3 ] (17) By projecting the surface normal vector F norm (x f ,y f ,z f ) and the point p1(x a ,y a ,z a ) The plane equation of the projection surface is calculated using formula (18).

8. The method according to claim 1, characterized in that The specific steps of step S6 are to obtain the equations by fitting the plane equation parameters A1, B1, C1, D1 and the projection plane equation parameters A2, B2, C2, D2. The specific calculation process is shown in formulas (19-24): A1,B1,C1=x u3 ,y u3 ,from u3 (19) D1=-(x u3 ×x a +y u3 ×y a +z u3 ×z a ) (20) A2,B2,C2=x f ,y f ,z f (21) D2=-(x f ×x a +y f ×y a +z f ×z a ) (22) Set z=z a and z=z b Substitute them into equation (23) and solve them to obtain the two endpoints p3(x c ,y c ,z a ) and p4(x d ,y d ,z b ); the projected measurement line length L can be solved by formula (24).

9. The method according to claim 1, characterized in that In step S8, the horizontal distance X from the sampling point to the survey line is calculated. n and vertical distance Z n Specifically: The endpoint of the projection line p3(x c ,y c ,z a ) and p4(x d ,y d ,z b ) as the reference line, and calculate the horizontal distance X of the nth sampling point cloud on the reference orientation. n and vertical distance Z n , where n∈λ, then calculate X according to formula (28-33) n ,Z n . <h2 style=";text-align:left;direction:ltr">S1,S2,S3=(x<h2 style=";text-align:left;direction:ltr"> c <h2 style=";text-align:left;direction:ltr"> -x<h2 style=";text-align:left;direction:ltr"> d <h2 style=";text-align:left;direction:ltr"> ),(y<h2 style=";text-align:left;direction:ltr"> c <h2 style=";text-align:left;direction:ltr"> -y<h2 style=";text-align:left;direction:ltr"> d <h2 style=";text-align:left;direction:ltr"> ),(z<h2 style=";text-align:left;direction:ltr"> a <h2 style=";text-align:left;direction:ltr"> -z<h2 style=";text-align:left;direction:ltr"> b <h2 style=";text-align:left;direction:ltr"> ) (28) Among them (α n , β n , γ n ) represents the projection coordinates of the nth point cloud on the projection line.

10. The method according to claim 1, characterized in that The calculation method of the roughness δ is: Where L is the curve projection length, N is the number of points on the line segment, and x n is the horizontal distance of the nth sampling point, Z n is the vertical distance of the nth sampling point; The fitting formula of the body surface roughness coefficient JRC in step S10 is: JRC=31.28+33.54log 10 d.

Images