A method for replicating the three-dimensional morphology of rock mass structural surfaces

Abstract

This invention discloses a method for replicating the three-dimensional morphology of rock mass structural surfaces. The method includes collecting rock samples with different types of structural surfaces, cleaning and cutting the structural surfaces of the rock samples into planes, and then scanning to obtain a three-dimensional digital model of the structural surfaces. Based on the dip angle of data points on the three-dimensional digital model, a roughness parameter R based on the Lyapunov index is calculated. λ and roughness parameter R based on fractal dimension D ; Select the roughness parameter R from multiple rock samples that meets the preset conditions. λ and roughness parameter R D The corresponding three-dimensional digital model is used as the three-dimensional replica structural surface and imported into the engraving machine; the cutting method of the structural surface is selected and input into the engraving machine; the engraving machine generates the three-dimensional engraving path of the structural surface according to the input three-dimensional digital model and cutting method; according to the three-dimensional engraving path, the engraving machine is used to replicate the selected structural surface on the splicing surface of two rock samples, and then the rock samples are spliced ​​to form a test specimen with a three-dimensional replica structural surface.

Description

Technical Field

[0001] This invention relates to the field of rock mechanics experimental technology, specifically to a method for replicating the three-dimensional morphology of rock mass structural surfaces. Background Technology

[0002] In geotechnical engineering, engineering rock masses consist of intact rock blocks and structural planes. The influence of external factors such as the geological environment and excavation disturbance on the mechanical properties of engineering rock masses is mainly manifested in the combined effect on fractured structural planes and intact rock. The shear mechanical properties of structural planes under complex environments have gradually become a focus of attention for scholars. However, due to the influence of the three-dimensional morphology, infill, and stress conditions of the structural plane surface, the mechanical properties of structural planes are complex and variable. Replicating structural planes with consistent three-dimensional morphology and conducting shear tests under different stress and geological environments is crucial for accurately understanding the shear mechanical properties of structural planes under complex environments.

[0003] Existing methods for preparing structural surfaces can be broadly categorized into four types: in-situ acquisition, artificial splitting / shearing, artificial casting, and 3D printing. While in-situ acquisition can reflect the original three-dimensional morphology of the structural surface after geological processes, it is difficult to acquire large batches of structural surfaces with consistent morphology and mechanical properties. Structural surfaces obtained through artificial splitting / shearing can reflect the morphological characteristics of shear and tension-type structural surfaces, but the morphological characteristics of structural surfaces obtained through artificial splitting / shearing using the same lithology and stress are not entirely consistent. Artificial casting and 3D printing can effectively obtain large batches of structural surfaces with identical morphological characteristics; however, due to the differences in mechanical properties between the materials used in casting and 3D printing and natural rock, the shear mechanical properties of the structural surfaces tested differ significantly from their inherent mechanical characteristics.

[0004] Therefore, it is urgent to propose a three-dimensional morphology replication technology for rock mass structural surfaces suitable for different testing conditions, to prepare a large batch of rock mass structures with the same three-dimensional morphological characteristics, and to truly obtain the shear mechanical properties of natural structural surfaces. Summary of the Invention

[0005] In view of the above-mentioned shortcomings in the prior art, the three-dimensional morphology replication method of rock mass structure surface provided by the present invention solves the problem that the existing replication methods cannot obtain a large number of structural surfaces with consistent morphological features and mechanical properties.

[0006] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows:

[0007] A method for replicating the three-dimensional morphology of rock mass structural surfaces is provided, comprising the following steps:

[0008] S1. Collect rock samples of different types of structural surfaces at the construction site, clean and cut the structural surfaces of the rock samples into planes, and then scan to obtain a three-dimensional digital model of the structural surfaces.

[0009] S2. Based on the chaotic bifurcation theory and the tilt angle of the data points on the 3D digital model, calculate the roughness parameter R based on the Lyapunov exponent. λ and roughness parameter R based on fractal dimension D ;

[0010] S3. Select the roughness parameter R from multiple rock samples that meets the preset conditions. λ and roughness parameter R D The corresponding 3D digital model is used as the structural surface for 3D replication and is imported into the engraving machine;

[0011] S4. Based on the test method of the structural surface, select the tool path mode of the structural surface and input it into the engraving machine; the engraving machine generates the three-dimensional engraving path of the structural surface based on the input three-dimensional digital model and tool path mode.

[0012] S5. Based on the three-dimensional carving path, the selected structural surface is replicated on the splicing surface of the two rock samples using a carving machine. Then, the rock samples are spliced ​​together to form a test specimen with a three-dimensional replicated structural surface.

[0013] Furthermore, step S2 further includes:

[0014] S21. Sort the apparent tilt angles of the data points according to their spatial location to form a spatial sequence {θ1, θ2, ..., θ...} N}, θ1, θ2 and θ N These are the 1st, 2nd, and Nth tilt angles in the spatial sequence, where N is the total number of data points;

[0015] S22. Based on the spatial sequence, determine the average mutual information corresponding to multiple different intervals τ using the average mutual information method, and select the interval τ corresponding to the minimum average mutual information; the expression for the average mutual information method is:

[0016]

[0017] Where I(τ) is the average mutual information; p i and p j p represents the probability density of two data points spaced τ apart in a spatial sequence. ij (τ) is the joint probability density of two data points separated by τ in the spatial sequence;

[0018] S23. For each data point in the spatial sequence, the nearest neighbor method is used to determine the embedding dimension m of the spatial sequence. Based on the embedding dimension and the selected interval τ, the reconstructed phase space vector is obtained:

[0019] X k =[θ k θ k+τ θ k+2τ,...,θ k+(m-1)τ ]

[0020] Where k = 1, 2, ..., N-(m-1)τ are variables; θ k θ k+τ θ k+2τ and θ k+(m-1)τ These are the data points with serial numbers k, k+τ, k+2τ, and k+(m-1)τ, respectively.

[0021] S24. For each phase space vector X k Find its nearest neighbor X k′ :

[0022]

[0023] Where ||·|| is the Euclidean distance; W is the exclusion window; and k′ is the phase space vector X. k The nearest neighbor number;

[0024] S25. According to the phase space vector X k and its nearest neighbor X k′ Calculate the initial distance d k (0) and tracking distance evolution d k (n):

[0025] d k (0)=||X k -X k′ ||,d k (n)=||X k+n -X k′+n ||

[0026] Among them, X k+n and X k′+n They are phase space vectors X k and its nearest neighbor X k′ The spatial vector obtained at discrete time step n;

[0027] S26. Calculate the maximum Lyapunov exponent λ based on the evolution of the initial distance and tracking distance. max :

[0028]

[0029] Where Δt is the sampling time interval;

[0030] S27. Calculate the correlation dimension D2 based on the phase space vectors:

[0031]

[0032] Where C(r) is the association score; r is the distance threshold; H(r-||X) i -X j ||) is the Heaviside function, which takes the value 1 if the input is greater than zero, and 0 otherwise; X i and X j These are the i-th and j-th phase space vectors, respectively;

[0033] S28. Using the maximum Lyapunov exponent λ max R as roughness parameter λ The correlation dimension D2 is used as the roughness parameter R. D .

[0034] Furthermore, the testing method includes a direct shear test and a triaxial test for a structural surface with an inclination angle; in the direct shear test, the tool moves back and forth along the horizontal plane of the structural surface to carve; in the triaxial test, the tool moves back and forth along the oblique cut surface to carve.

[0035] Furthermore, when replicating the rock sample blocks that underwent the direct shear test, after one rock sample block was carved according to the generated carving path, the carving path was mirrored in 3D and 2D to obtain a new path, and the new path was used to carve the other rock sample block.

[0036] When replicating rock samples used in triaxial tests, after one rock sample is carved according to the generated carving path, the carving path is mirrored in 3D and then in 2D. Then, the path is reversed to obtain a new path, which is then used to carve another rock sample.

[0037] Furthermore, when replicating the selected structural surfaces on the rock sample block, a fixing clamp is used to hold the rock sample block so that the splicing surface of the rock sample block faces the engraving machine's cutting tool; the fixing clamp for fixing the rock sample block for triaxial testing includes:

[0038] A base on which a support plate is fixed, and the support plate has a mounting hole, and an annular limiting platform is provided around the edge of the mounting hole;

[0039] The rock sample clamping part includes a main clamping block and a secondary clamping block that are detachably connected by fastening bolts. Both the main clamping block and the secondary clamping block are provided with receiving grooves that cooperate with each other to clamp the rock sample block. The end of the main clamping block is provided with a circular rotating part that passes through the mounting hole.

[0040] The limiting part includes a limiting hole provided on the annular limiting platform and the circular rotating part, and a limiting member inserted into the limiting hole to limit the rotation of the main clamping block relative to the support plate.

[0041] Furthermore, the circular rotating part includes a circular boss passing through the mounting hole, and an annular knob that surrounds the annular limiting platform in the opposite direction is fixed to the end of the circular boss. The limiting hole of the circular rotating part is opened on the annular knob, and an angle mark is provided on the annular knob.

[0042] Furthermore, the method for replicating the three-dimensional morphology of rock mass structural surfaces also includes a support block disposed on the main clamping block and / or the secondary clamping block for supporting the rock sample block at the end where the splicing surface is located.

[0043] Furthermore, the cross-section of the limiting hole is semi-circular or triangular.

[0044] Compared with existing methods for preparing structural surfaces, this method proposes a three-dimensional morphology replication method for rock mass structural surfaces, which has the following advantages:

[0045] This solution involves collecting various types of structural surfaces on-site, obtaining 3D digital models of these surfaces using 3D laser scanning technology, and then calculating the roughness parameter R. λ and roughness parameter R D Select the structural surface to be replicated, with roughness parameter R. λ and roughness parameter R D It can capture complex nonlinear features that traditional methods cannot describe, and can describe the self-similarity and complexity of structural surfaces at different scales.

[0046] The method provided in this scheme is used to select structural surfaces. The resulting structural surfaces have relatively small anisotropy and roughness that is neither too large nor too small. When conducting experiments, the results will not be too discrete. This allows for a better highlighting of the influence of the environment on the rock when studying the impact of the environment on shear strength.

[0047] This solution provides a fixing fixture for rock samples used in triaxial testing. The angle of the rock sample can be adjusted through the clamping part so that the splicing surface of the rock sample faces the cutting tool of the engraving machine, transforming the bevel engraving into a flat engraving, which simplifies the engraving difficulty and reduces the risk of chipping.

[0048] The rock fixture proposed in this invention, compared with existing methods for preparing structural surfaces, can achieve three-dimensional replication of structural surfaces applicable to different experimental methods. Attached Figure Description

[0049] Figure 1 This is a flowchart of a method for replicating the three-dimensional morphology of rock mass structural surfaces.

[0050] Figure 2 This is a schematic diagram of a rock mass structure taken from a construction site according to an embodiment of the present invention.

[0051] Figure 3 for Figure 2 A schematic diagram of the three-dimensional digital model of the rock mass structure surface after scanning.

[0052] Figure 4 The diagrams show the tool path for different test methods; (a) is a schematic diagram of the engraving path for the direct shear test; (b) is a schematic diagram of the engraving path for the triaxial test.

[0053] Figure 5 The front view of the fixture for fixing rock samples for triaxial testing, omitting the base and support block.

[0054] Figure 6 Left view of the fixture used to hold a rock sample block for triaxial testing.

[0055] Figure 7 Right view of the fixture used to hold a rock sample block for triaxial testing.

[0056] Figure 8 This is a schematic diagram of the cutting method applicable to elliptical oblique cut surfaces.

[0057] Figure 9 A schematic diagram of a test specimen applicable to direct shear testing of structural surfaces according to an embodiment of the present invention is shown.

[0058] Figure 10 A schematic diagram of a test specimen applicable to triaxial testing of a tilted structural surface according to an embodiment of the present invention is shown.

[0059] Figure 11 A comparison diagram of the actual structural surface and the replicated structural surface according to an embodiment of the present invention is shown; (a) is a schematic diagram of the actual structural surface; (b) is a schematic diagram of the replicated structural surface.

[0060] Figure 12 A comparison diagram of the actual structural surface traces and the replicated structural surface traces according to an embodiment of the present invention is shown.

[0061] The components include: 1. Fixing clamp; 11. Base; 12. Rock sample clamping part; 121. Main clamping block; 1211. Receiving groove; 1212. Circular rotating part; 12121. Circular boss; 12122. Annular knob; 122. Secondary clamping block; 13. Limiting hole; 14. Support block; 15. Support plate; 151. Annular limiting platform; 2. Engraving path; 3. Rock sample block; 4. Test specimen. Detailed Implementation

[0062] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0063] refer to Figure 1 , Figure 1 A method for replicating the three-dimensional morphology of rock mass structural surfaces is shown; the method includes steps S1 to S5.

[0064] In step S1, rock samples of different types of structural surfaces are collected at the construction site. The structural surfaces of the rock samples are cleaned and cut into planes, and then scanned to obtain a three-dimensional digital model of the structural surface; specifically:

[0065] Collect shear-type or tension-type slightly weathered or unweathered rock mass structural surfaces at the construction site. Use a soft brush to properly clean the surface of the rock mass structural surfaces. Cut irregular structural surface samples into 100mm × 100mm planar dimensions (e.g., Figure 2 As shown), to meet the requirements of 3D laser scanning of structural surfaces. A high-precision 3D laser scanner is used to perform detailed scanning of the structural surfaces, resulting in a 3D digital model of the structural surfaces (such as...). Figure 3 (As shown).

[0066] The scanned 3D digital models vary in size. Before proceeding to step S2, the 3D digital models are appropriately trimmed using a Matlab algorithm, with the center point of the structural surface as a reference. Considering the errors during the structural surface replication process and the needs of testing, the 3D digital models are trimmed to a planar size of 52mm × 50mm.

[0067] In step S2, based on the chaotic bifurcation theory and the tilt angle of the data points on the three-dimensional digital model, the roughness parameter R based on the Lyapunov exponent is calculated. λ and roughness parameter R based on fractal dimension D .

[0068] In one embodiment of the present invention, step S2 further includes:

[0069] S21. Sort the apparent tilt angles of the data points according to their spatial location to form a spatial sequence {θ1, θ2, ..., θ...} N}, θ1, θ2 and θ N These are the 1st, 2nd, and Nth tilt angles in the spatial sequence, where N is the total number of data points;

[0070] S22. Based on the spatial sequence, determine the average mutual information corresponding to multiple different intervals τ using the average mutual information method, and select the interval τ corresponding to the minimum average mutual information; the expression for the average mutual information method is:

[0071]

[0072] Where I(τ) is the average mutual information; p i and p j p represents the probability density of two data points spaced τ apart in a spatial sequence. ij (τ) is the joint probability density of two data points separated by τ in the spatial sequence;

[0073] S23. For each data point in the spatial sequence, the nearest neighbor method is used to determine the embedding dimension m of the spatial sequence. Based on the embedding dimension and the selected interval τ, the reconstructed phase space vector X is obtained. k :

[0074] X k =[θ k θ k+τ θ k+2τ ,...,θ k+(m-1)τ ]

[0075] Where k = 1, 2, ..., N-(m-1)τ are variables; θ k θ k+τ θ k+2τ and θ k+(m-1)τ These are the data points with serial numbers k, k+τ, k+2τ, and k+(m-1)τ, respectively.

[0076] S24. For each phase space vector X k Find its nearest neighbor X k′ :

[0077]

[0078] Where ||·|| is the Euclidean distance; W is the exclusion window; and k′ is the phase space vector X. k The nearest neighbor number;

[0079] S25. According to the phase space vector X k and its nearest neighbor X k′ Calculate the initial distance d k (0) and tracking distance evolution d k (n):

[0080] d k (0)=||X k -X k′ ||,d k (n)=||Xk+n -X k′+n ||

[0081] Among them, X k+n and X k′+n They are phase space vectors X k and its nearest neighbor X k′ The spatial vector obtained at discrete time step n;

[0082] S26. Calculate the maximum Lyapunov exponent λ based on the evolution of the initial distance and tracking distance. max :

[0083]

[0084] Where Δt is the sampling time interval;

[0085] S27. Calculate the correlation dimension D2 based on the phase space vectors:

[0086]

[0087] Where C(r) is the association score; r is the distance threshold; H(r-||X) i -X j ||) is the Heaviside function, which takes the value 1 if the input is greater than zero, and 0 otherwise; X i and X j These are the i-th and j-th phase space vectors, respectively;

[0088] S28. Using the maximum Lyapunov exponent λ max R as roughness parameter λ The correlation dimension D2 is used as the roughness parameter R. D .

[0089] Chaotic bifurcation theory can reveal the nonlinear dynamic behavior of a system. Applying it to the evaluation of surface roughness can capture complex nonlinear characteristics that traditional methods cannot describe. Furthermore, chaotic characteristic parameters (such as fractal dimension) can describe the self-similarity and complexity of the surface at different scales. This is particularly important for rock surfaces with multi-scale roughness characteristics.

[0090] In step S3, roughness parameters R that meet preset conditions are selected from multiple rock samples. λ and roughness parameter R D The corresponding 3D digital model is used as the structural surface for 3D replication and is imported into the engraving machine;

[0091] In step S4, the tool path mode of the structural surface is selected according to the test method of the structural surface and input into the engraving machine; the engraving machine generates the three-dimensional engraving path 2 of the structural surface according to the input three-dimensional digital model and tool path mode, specifically using JD Paint software in the engraving machine.

[0092] In implementation, the preferred testing methods of this scheme include direct shearing test and triaxial test for structural surfaces with inclination angle; in the direct shearing test, the tool path is to carve back and forth along the horizontal plane of the structural surface; in the triaxial test, the tool path is to carve back and forth along the oblique cut surface.

[0093] Depending on the testing method, the carving method for the structural surface varies, and the method for generating the 3D carving path 2 for the structural surface also differs. To facilitate understanding of the 3D carving path 2, this solution provides a schematic diagram of the carving path 2 for direct shearing tests and triaxial tests. For details, please refer to [the relevant documentation / references]. Figure 4 (a) and (b) in the example.

[0094] In step S5, according to the three-dimensional carving path 2, the selected structural surface is copied on the splicing surface of the two rock sample blocks 3 using a carving machine. Then, the rock sample blocks 3 are spliced ​​together to form a test specimen 4 with a three-dimensional copied structural surface.

[0095] In practice, when replicating the rock sample 3 that underwent the direct shear test, this scheme preferably involves carving one rock sample 3 according to the generated carving path 2, then performing 3D mirroring and 2D mirroring operations on the carving path 2 to obtain a new path, and using the new path to carve the other rock sample 3.

[0096] When replicating the rock sample 3 used in the triaxial test, after one of the rock sample 3 is carved according to the generated carving path 2, the carving path 2 is mirrored in 3D and then mirrored in 2D. Then, the path is reversed to obtain a new path, and the new path is used to carve the other rock sample 3.

[0097] This scheme employs the above method to carve the two mating rock sample blocks 3, ensuring a complete fit of the structural surfaces at the joint, more closely resembling real-world structural surfaces. After the two rock sample blocks 3 are carved, they are assembled to obtain a test specimen 4 suitable for direct shear testing of the structural surface, as shown below. Figure 9 It is applicable to test specimens with inclined structural surfaces undergoing triaxial testing. Figure 10 As shown.

[0098] like Figures 5-7 As shown, when replicating the selected structural surface on the rock sample block 3, the rock sample block 3 is clamped by the fixing clamp 1 so that the splicing surface of the rock sample block 3 faces the engraving machine cutter; the fixing clamp 1 for fixing the rock sample block 3 for triaxial testing includes a base 11, a rock sample block clamping part 12 and a limiting part.

[0099] A support plate 15 is fixed on the base 11. The support plate has a mounting hole and an annular limiting platform 151 surrounding the mounting hole is provided at the edge of the mounting hole.

[0100] The rock sample clamping part 12 includes a main clamping block 121 and a secondary clamping block 122 that are detachably connected by fastening bolts. Both the main clamping block 121 and the secondary clamping block 122 are provided with receiving grooves 1211 that cooperate with each other to clamp the rock sample block 3. The end of the main clamping block 121 is provided with a circular rotating part 1212 that passes through the mounting hole.

[0101] The limiting part includes a limiting hole 13 provided on the annular limiting platform 151 and the circular rotating part 1212, and a limiting member inserted into the limiting hole 13 to limit the rotation of the main clamping block 121 relative to the support plate 15. The limiting member is not shown in the figure.

[0102] Refer again Figure 6 The circular rotating part 1212 includes a circular boss 12121 passing through the mounting hole. The end of the circular boss 12121 is fixed with an annular knob 12122 that surrounds the annular limiting platform 151 in the opposite direction. The limiting hole 13 of the circular rotating part 1212 is opened on the annular knob 12122. Angle marks are provided on the annular knob 12122.

[0103] like Figure 6 and Figure 7 As shown, the fixing clamp also includes a support block 14 disposed on the main clamping block 121 and / or the auxiliary clamping block 122 for supporting the rock sample block 3 at the end where the splicing surface is located. The cross-section of the limiting hole 13 is semi-circular or triangular. When it is set as a triangle, the same type of clamp can be adapted to clamp rock sample blocks of different diameters, making it more practical.

[0104] Combination Figures 5-7 The fixed fixture shown rotates the beveled surface to keep it horizontal, and the tool path is the same as... Figure 4 (a) shows consistency; considering that the elliptical bevel is not convenient for locating the initial point of carving path 2, a circular carving path 2 can be generated using the "circular equidistant" method (e.g. Figure 8 (As shown).

[0105] A rock sample was collected from the construction site. After cleaning and cutting its structural surface, a high-precision 3D laser scanner was used to perform a detailed scan of the structural surface. The scan results are for reference. Figure 11 (a) in the middle; followed by Figure 4 The carving path shown in (a) is used to carve rock sample 3. The scanning results of the resulting structural surfaces are referenced. Figure 11 (b) in the middle, in contrast Figure 11 As can be seen from (a) and (b) in the figure, the replicated structural surface basically simulates and replicates the rough features of the actual structural surface. It can be seen that the engraving method can replicate structural surfaces with consistent morphological features and mechanical properties in large batches.

[0106] exist Figure 11 For comparison, a trace is selected at the corresponding position of the structural surface shown in (a) and (b), as shown in the figure. Figure 12 As shown, the replicated structural surface traces are quite close to the real structural surface traces. Under the same y-value conditions, the difference in z-value is less than 0.2 mm. It can be seen that the engraving method can basically restore the real structural surface. It can be seen that the replicated structural surface can basically reflect the original three-dimensional morphological features of the structural surface after geological processes.

Claims

1. A method for reproducing three-dimensional topography of a rock mass structural plane, characterized in that, The method comprises the steps of: S1, collecting rock samples of different types of structural surfaces at a construction site, cleaning and cutting the structural surfaces of the rock samples into planes, and then scanning to obtain a three-dimensional digital model of the structural surface; S2. Calculate the roughness parameter based on Lyapunov exponent according to the chaos bifurcation theory and the viewing angle of the data points on the three-dimensional digital model R λ and the roughness parameter based on fractal dimension R D ; the step S2 further comprises: S21, ordering the view inclination angles of the data points according to the spatial positions to form a spatial sequence , θ 1、 θ 2and θ N for the first, second and third view inclination angles in the spatial sequence, N N is the total number of data points.​ S22、According to the spatial sequence, the average mutual information method is used to determine a plurality of different intervals τ The corresponding average mutual information is selected, and the interval corresponding to the minimum average mutual information is selected τ The expression of the average mutual information method is: in, I ( τ () represents average mutual information; p i and p j Spacing in the spatial sequence τ The probability density of two data points; Spacing in the spatial sequence τ The joint probability density of two data points; S23. For each data point in the spatial sequence, determine the embedding dimension of the spatial sequence using a nearest neighbor method m , according to the embedding dimension and the selected interval τ , obtain the reconstructed phase space vector : wherein k = 1, 2,... N ( m 1) τ is a variable; θ k 、 θ k+τ 、 θ k+2τ and θ k+（m-1）τ are data points of order k 、 k+ τ 、 k+ 2 τ 、 k+ (m - 1) τ . S24, for each phase space vector , find its nearest neighbor : wherein, is the Euclidean distance; W is the exclusion window; is the phase space vector is the order of the nearest neighbor of S25, according to phase space vector X k and its nearest neighbors , compute initial distances and track distance evolution : ， wherein, X k+n and are phase space vectors X k and their nearest neighbors at discrete time steps n ; S26、According to the initial distance and the tracking distance evolution, the maximum Lyapunov index is calculated λ max : wherein Δt is a sampling time interval; S27. Calculate the correlation dimension from the phase space vectors D 2: ， wherein, C r ) is a correlation score; r is a distance threshold; is a Heaviside function that takes 1 when the input is greater than zero and 0 otherwise; X i and X j are the first i and the first j phase space vectors, respectively;​ S28, using the largest Lyapunov exponent λ max as a roughness parameter R λ , using the correlation dimension D2 D as a roughness parameter R D ; S3, selecting roughness parameters in the plurality of rock samples that satisfy the predetermined condition R λ and the roughness parameter R D the corresponding three-dimensional digital model as a three-dimensionally replicated structural plane, and importing it into the engraving machine; S4, selecting a tool path of the structural surface according to the test method of the structural surface, and inputting the tool path into the engraving machine; the engraving machine generates a three-dimensional engraving path of the structural surface according to the input three-dimensional digital model and the tool path; S5, according to the three-dimensional engraving path, the selected structural surface is reproduced on the splicing surface of the two rock sample blocks by using the engraving machine, and then the rock sample blocks are spliced to form a test sample with a three-dimensional reproduced structural surface.

2. The method according to claim 1, wherein, The test method comprises a direct shear test and a triaxial test for a structural surface with an inclination angle; in the direct shear test, the tool path is engraved along the horizontal plane of the structural surface; in the triaxial test, the tool path is engraved along the inclined surface.

3. The method according to claim 2, wherein, When the rock sample blocks for the direct shear test are reproduced, one of the rock sample blocks is engraved according to the generated engraving path, then a new path is obtained by performing 3D mirroring and 2D mirroring on the engraving path, and the other rock sample block is engraved using the new path; When the rock sample blocks for the triaxial test are reproduced, one of the rock sample blocks is engraved according to the generated engraving path, then a new path is obtained by performing 3D mirroring, 2D mirroring and reverse operation on the engraving path, and the other rock sample block is engraved using the new path.

4. The method according to claim 2, wherein, When the selected structural surface is reproduced on the rock sample blocks, a fixed clamp is used to clamp the rock sample blocks, so that the splicing surface of the rock sample blocks faces the engraving tool of the engraving machine; the fixed clamp for fixing the rock sample blocks for the triaxial test comprises: a base, a support plate is fixed on the base, an installation hole is formed in the support plate, and an annular limiting table surrounding the installation hole is arranged at the edge of the installation hole; a rock sample block clamping part, which comprises a main clamping block and a secondary clamping block connected by a fastening bolt, and a receiving groove for clamping the rock sample block is arranged on the main clamping block and the secondary clamping block; a circular rotating part penetrating through the installation hole is arranged at the end of the main clamping block; a limiting part, which comprises a limiting hole arranged on the annular limiting table and the circular rotating part, and a limiting piece inserted into the limiting hole to limit the rotation of the main clamping block relative to the support plate.

5. The method according to claim 4, wherein, The circular rotating part comprises a circular boss penetrating through the installation hole, an annular knob surrounding the annular limiting table in the opposite direction is fixed at the end of the circular boss, the limiting hole of the circular rotating part is formed in the annular knob, and an angle scale is arranged on the annular knob.

6. The method according to claim 4 or 5, wherein, A supporting block is arranged on the main clamping block and / or the secondary clamping block to support the rock sample block at the end of the splicing surface.

7. The method according to claim 4 or 5, wherein, The cross section of the limiting hole is semicircular or triangular.

Images