Underground engineering surrounding rock surface and inner structure identification method

Abstract

The invention discloses an underground engineering surrounding rock surface and inner structure identification method, and relates to the technical field of geotechnical engineering reconnaissance, and the method comprises the steps: obtaining a while-drilling parameter based on an on-site surrounding rock digital drilling process, and analyzing an inversion change curve of a rock mass mechanical parameter along with the drilling depth; the method comprises the following steps: constructing a parameter discriminant analysis model and an image processing model according to multidirectional surrounding rock surface and internal crack images, and establishing a surrounding rock'surface-inside 'structure characteristic parameter set; the method comprises the steps of extracting fracture feature points and fracture contour lines by constructing a fracture image preprocessing neural network model; according to the method, surface and internal fracture feature point coordinates are fused on the basis of an SURF function calculation method, a surrounding rock fracture structure three-dimensional model is established, and surface-inside structure in-situ identification of surrounding rock is achieved by fusing while-drilling parameter inversion and a digital image processing technology and combining a neural network three-dimensional modeling technology. The technical problems that surrounding rock surface and internal fracture structure parameters are difficult to synchronously obtain and surface and internal structure characteristic collaborative analysis is insufficient are solved.

Description

Technical Field

[0001] This invention relates to the field of geotechnical engineering investigation technology, specifically to a method for identifying the surface and internal structure of surrounding rock in underground engineering. Background Technology

[0002] With the acceleration of urbanization, underground space engineering projects such as traffic tunnels and mine roadways have developed rapidly, and their construction scale has continued to increase. The stability of the surrounding rock in underground engineering is affected by the rock mass structure and properties, especially by rock fissures. During the construction of underground engineering projects, areas with well-developed fissures are prone to engineering disasters such as large deformation of the surrounding rock, roof falls, and collapses. Accurately identifying the "surface-interior" structure of the surrounding rock, understanding the development of "surface-interior" fissures, and obtaining the characteristic parameters of the "surface-interior" structure of the surrounding rock can provide key basis for the design and construction of underground engineering projects.

[0003] Existing methods for identifying surrounding rock structures commonly employ drilling sampling, acoustic testing, and computer numerical simulation imaging. Drilling sampling can obtain characteristic parameters of the internal structure of the surrounding rock, but it only provides information about the local surrounding rock near the borehole, making it difficult to reflect the situation over a large area. Acoustic testing can achieve non-destructive testing of the surrounding rock structure, but its detection depth is limited and it is easily affected by the surrounding environment and the complexity of the internal rock mass, leading to insufficient reliability and accuracy of the test results. Computer numerical simulation imaging can obtain characteristic parameters of the surrounding rock surface structure relatively accurately, but it is highly dependent on the quality of the input parameters, the process is cumbersome, and obtaining characteristic parameters of the internal structure of the surrounding rock is difficult. In summary, among existing methods for identifying the surface and internal structure of surrounding rock, the difficulty in obtaining characteristic parameters of internal fracture structures makes it difficult to effectively combine the identification processes of the surface and internal structures, resulting in a lack of comprehensive and accurate identification of the surface and internal structure of a large area of ​​surrounding rock. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by providing a method for identifying the surface and internal structures of surrounding rock in underground engineering. This method overcomes the difficulty in obtaining characteristic structural parameters of internal fissures in surrounding rock and solves the problem of combining the identification of large-scale surface-internal fissure structural characteristics of surrounding rock. It enables accurate and comprehensive identification of the surface and internal structures of surrounding rock, providing a basis for on-site analysis of surrounding rock conditions.

[0005] To solve the above problems, the following solution is adopted: Methods for identifying the surface and internal structure of surrounding rock in underground engineering include: Conduct on-site digital drilling tests of the surrounding rock to obtain drilling parameters of the rock mass and collect images of fractures on the surface and inside the surrounding rock. Structural feature parameters of fractures are extracted by analyzing drilling parameters of rock mass, and geometric feature parameters of fractures are obtained by analyzing fracture images. The structural feature parameters and geometric feature parameters are combined to form a set of fracture feature parameters. After preprocessing the fracture images, the fracture feature points and fracture contour lines are extracted from the fracture images, the coordinates of the three-dimensional key points of the fracture structure are calculated, and a three-dimensional model of the surrounding rock fracture structure is constructed by combining the fracture feature parameter set. Identify the surface and internal structure of the surrounding rock based on a three-dimensional model of the surrounding rock fracture structure.

[0006] Furthermore, the structural characteristic parameters of fractures extracted from the drilling parameters of the analyzed rock mass include: Based on the drilling parameters of the rock mass, the inversion curve of the compressive strength of the rock mass with the drilling depth is obtained; A discriminant analysis model for the characteristic parameters of surrounding rock fractures was constructed using the inversion variation curves. The drilling parameters of the rock mass were input into the discriminant analysis model for the characteristic parameters of surrounding rock fractures to extract the structural characteristic parameters of the fractures.

[0007] Furthermore, the structural characteristic parameters include crack density, crack width, and crack orientation.

[0008] Furthermore, the step of analyzing the crack image to obtain the geometric feature parameters of the crack includes: using an image processing model to label the geometric feature parameters in the crack image, and converting the crack image information into crack image label data.

[0009] Furthermore, the geometric feature parameters include the fracture location, fracture area, and fracture length.

[0010] Furthermore, the crack images are acquired through drilling and external camera acquisition.

[0011] Furthermore, the preprocessing of the crack image includes: The images of cracks on the surface and inside the surrounding rock were enhanced by histogram equalization to improve image contrast, and then the images were denoised by median filtering.

[0012] Furthermore, when constructing a three-dimensional model of the surrounding rock fracture structure, the fractures are represented as geometric entities in three-dimensional space based on their spatial distribution and shape characteristics, and then modeled accordingly.

[0013] Furthermore, during the on-site digital drilling test of the surrounding rock, images of cracks on the surface and inside of the surrounding rock from different orientations were collected.

[0014] Furthermore, the extraction of crack feature points and crack contour lines from the crack image includes: Image processing models are used to extract fracture feature points and fracture contour lines from fracture images; The image processing model is trained by a neural network. The fracture feature parameters in the set of fracture feature parameters corresponding to the input layer nodes are set as input variables, and the surrounding rock fracture feature points and contour lines corresponding to the nodes of the intermediate hidden layer and the output layer nodes are set as output variables.

[0015] Compared with the prior art, the advantages and positive effects of this invention are: To address the current difficulty in obtaining surface-to-interior structural characteristic parameters of surrounding rock, on-site digital drilling tests were conducted. By monitoring drilling parameters and combining them with images of fractures on the surface and inside the surrounding rock, a set of fracture characteristic parameters was established after processing, and a three-dimensional model of the surrounding rock fracture structure was constructed. This enabled the identification of the surface and interior structures of the surrounding rock, overcoming the obstacle of obtaining characteristic structural parameters of internal fractures in the surrounding rock. It also solved the problem of difficulty in combining large-scale identification of surface-to-interior fracture structural characteristics of surrounding rock, avoiding the operational steps of core sampling and manual identification in traditional drilling and sampling methods, improving identification efficiency and reducing identification costs.

[0016] By acquiring images of fractures on the surface and inside of the surrounding rock from different orientations, and utilizing parametric discrimination analysis and image processing models, the structural characteristic parameters of the fractures on the surface and inside the surrounding rock were identified, forming a set of "surface-interior" fracture structural characteristic parameters. A neural network was used to preprocess the images, solving for the coordinates of three-dimensional key points of the fractures inside and outside the surrounding rock, constructing a three-dimensional model of the surrounding rock fracture structure, and performing comprehensive structural identification of the surrounding rock in the field. This solved the problem of difficulty in obtaining characteristic parameters of the internal fracture structure of the surrounding rock and the difficulty in combining methods for identifying the surface and internal structure of the surrounding rock in the field. Attached Figure Description

[0017] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0018] Figure 1 This is a schematic diagram of the method for identifying the surface and interior structure of the surrounding rock in underground engineering according to Embodiment 1 of the present invention.

[0019] Figure 2 This is a schematic diagram of the process of identifying the surface and internal structure of the surrounding rock in underground engineering in Embodiment 1 of the present invention.

[0020] Figure 3 This is a schematic diagram of the three-dimensional key points of the crack structure in Embodiment 1 of the present invention.

[0021] Figure 4 This is a basic framework diagram of the MATLAB neural network regarding crack feature parameters in Embodiment 1 of the present invention.

[0022] In the diagram, 1. Borehole in surrounding rock; 2. Borehole inspection instrument; 21. Borehole inspection instrument probe; 22. Borehole inspection instrument connecting rod; 23. Borehole inspection instrument connecting cable reel; 24. Borehole inspection instrument image processing equipment; 3. Image acquisition equipment; 4. Digital drilling system. Detailed Implementation

[0023] Example 1 In a typical embodiment of the present invention, such as Figures 1-4 As shown, a method for identifying the surface and internal structure of surrounding rock in underground engineering is presented.

[0024] Methods for identifying the surface and internal structure of surrounding rock in underground engineering include: Conduct on-site digital drilling tests of the surrounding rock to obtain drilling parameters of the rock mass and collect images of fractures on the surface and inside the surrounding rock. Structural feature parameters of fractures are extracted by analyzing drilling parameters of rock mass, and geometric feature parameters of fractures are obtained by analyzing fracture images. The structural feature parameters and geometric feature parameters are combined to form a set of fracture feature parameters. After preprocessing the fracture images, the fracture feature points and fracture contour lines are extracted from the fracture images, the coordinates of the three-dimensional key points of the fracture structure are calculated, and a three-dimensional model of the surrounding rock fracture structure is constructed by combining the fracture feature parameter set. Identify the surface and internal structure of the surrounding rock based on a three-dimensional model of the surrounding rock fracture structure.

[0025] Specifically, by conducting on-site digital drilling tests of the surrounding rock, the inversion curve of the rock mass compressive strength with drilling depth during the drilling process was analyzed; Images of fractures on the surface and inside of the surrounding rock from different orientations were collected. A CFS parameter discrimination analysis model and a LABELME image processing model were established to obtain the structural feature parameters of the surrounding rock surface and inside, forming a set of "surface-interior" fracture feature parameters. The image was preprocessed using a MATLAB neural network to extract the fracture feature points and fracture contour lines from the fracture image; The coordinates of three-dimensional key points on the surface and internal fractures of the surrounding rock are solved, and a three-dimensional model of the fracture structure of the surrounding rock is formed by using the SURF function calculation method.

[0026] In this embodiment, identifying the surface and internal structure of the surrounding rock based on a three-dimensional model of the surrounding rock fracture structure refers to constructing a three-dimensional model that includes the spatial distribution and morphological characteristics of fractures on the surface and inside of the surrounding rock, thereby achieving integrated analysis and identification of the surface and internal structural features of the surrounding rock. In "surface-internal", "surface" refers to the surface structure and "internal" refers to the internal structure.

[0027] Surface structure identification includes: In a 3D model, the spatial distribution and morphology of surface cracks (such as crack length, width, and orientation) can be directly mapped using image label data processed by LABELME, intuitively presenting the degree of development of surface cracks. For example, the density and connectivity of surface cracks represented in the 3D model can be used to determine the stability of the surrounding rock surface.

[0028] Internal structure identification includes: A 3D model, through the fusion of drilling parameters and borehole inspection data, transforms parameters such as the location (depth), density, and orientation of internal fractures into coordinate points and geometric contours in 3D space. The 3D model can mark internal fractures at corresponding depths and quantify the development of internal fractures by using CFS parameters to discriminate and analyze parameters such as fracture density calculated by the model.

[0029] Combination Figures 1-4 The method for identifying the surface and internal structure of surrounding rock in underground engineering specifically includes the following steps: S101: Digital drilling tests were conducted on the surrounding rock of underground engineering using Digital Drilling System 4 to obtain rock mass parameters during drilling. The inverse variation curve of the rock mass compressive strength with drilling depth was obtained through the formula for the evolution of the surrounding rock compressive strength during drilling. The formula for the evolution of the surrounding rock compressive strength during drilling is as follows: (1); in, For rock mass compressive strength, A and B These are the fitting coefficients. N The drill bit rotation speed, M This is the drilling torque. F For drilling pressure, V For drilling speed, r Where is the drill bit radius. μ The friction coefficient between the drill bit cutting edge and the rock mass is denoted as . l i The length of the cutting edge of each drill bit. n The number of cutting edge columns. α The cutting angle, β This refers to the drill bit inclination angle.

[0030] S102: Collect images of fractures on the surface and inside of the surrounding rock from different orientations, and establish a CFS parameter discrimination analysis model: (2); in, D For crack density, d This represents the amount by which the rock mass compressive strength decreases with drilling depth (when the difference in the drilling curve is greater than or equal to 35%, it is considered that there are cracks inside the borehole). W The width of the crack. H This refers to the drilling depth. SThe direction of the fracture. p Number the boreholes. q Number the fractures in each borehole peep image. K The distance is the direction of the fracture. T This is the vertical distance between the borehole and the top surface of the surrounding rock.

[0031] S103: As Figure 2 As shown, a digital drilling test of the surrounding rock of underground engineering was carried out using the digital drilling system 4. The borehole inspection method and external high-definition camera technology were combined to obtain the "surface-inside" fracture images of the borehole, that is, the fracture images of the surface and inside of the surrounding rock.

[0032] Fracture images are acquired through borehole inspection and external camera capture. The digital drilling system 4 includes a borehole inspection instrument 2 and an image acquisition device 3. The borehole inspection instrument 2 includes a borehole inspection probe 21, a borehole inspection connecting rod 22, a borehole inspection connecting cable reel 23, and a borehole inspection image processing device 24. The borehole inspection probe 21 is connected to the connecting rod 22. The wiring of the borehole inspection probe 21 is connected to the borehole inspection image processing device 24 through the borehole inspection connecting cable reel 23 to send the acquired data to the borehole inspection image processing device 24. The image acquisition device 3 can acquire fracture information on the surrounding rock surface and the distribution information of the surrounding rock boreholes 1.

[0033] Using the CFS parameter analysis model and the LABELME image processing model in the MATLAB environment, structural feature parameters such as crack density, width and orientation are extracted. The location, area and length of cracks in the acquired crack images are marked, and crack image information is converted into crack image label data. The crack feature parameter set is formed by combining structural feature parameters and geometric feature parameters.

[0034] Specifically, based on the drilling parameters of the rock mass, the inversion curve of the rock mass compressive strength with drilling depth is obtained; and the inversion curve is used to construct a discriminant analysis model for the characteristic parameters of the surrounding rock fracture. The drilling parameters of the rock mass are input into the discriminant analysis model to extract the structural characteristic parameters of the fractures. The structural characteristic parameters include fracture density, fracture width, and fracture orientation.

[0035] Using an image processing model, geometric feature parameters in the fracture image are labeled, and the fracture image information is converted into fracture image label data. The geometric feature parameters include fracture location, fracture area, and fracture length.

[0036] Structural characteristic parameters and geometric characteristic parameters constitute the set of crack characteristic parameters.

[0037] S104: As Figure 4As shown, a neural network model is constructed using MATLAB to enhance the acquired images. Histogram equalization is used to improve image contrast, and median filtering is used to denoise the optimized images, thereby improving image quality.

[0038] like Figure 4 As shown, the CFS parameter analysis model and the LABELME image processing model are trained using MATLAB neural networks. The fracture feature parameters in the fracture feature parameter set corresponding to the input layer nodes are set as input variables, and the surrounding rock fracture feature points and contour lines corresponding to the output layer nodes are set as output variables through the nodes of the intermediate hidden layer.

[0039] like Figure 3 As shown, the method for calculating the coordinates of the three-dimensional key points of the fracture structure is as follows: (3); in, X For the thickness of the surrounding rock, This is the ratio coefficient between the image and the actual distance. e Number the images. f Number the pixels in the crack outline of the image.

[0040] S105: Using the SURF function in the MATLAB language environment, the fractures are represented as geometric entities in three-dimensional space according to their spatial distribution and shape characteristics, and a three-dimensional model of the surrounding rock fracture structure is constructed.

[0041] The SURF function uses the Hessian matrix to calculate the feature points of the fracture structure. Assume the feature point function is... F(y,z) The specific matrix calculation formula is as follows: (4); The matrix discriminant is: (5); H The eigenvalues ​​of the matrix discriminant can be classified according to the sign of the calculated result to determine whether the coordinate point is an extreme point. This can be achieved using image points. L(y,z) Equivalent substitution F(y,z) Calculate the H matrix using the following formula: (6); in, Represents the second-order differential of Gauss. With image pixels (y,z) Convolution at the point, Represents the second-order differential of Gauss. With image pixels (y,z) Convolution at the point, Represents the second-order differential of Gauss. With image pixels (y,z) Convolution at the point.

[0042] As a further implementation, the determinant of the Hessian matrix is ​​calculated, and the result is: (7); This embodiment can realize the identification of the "surface-interior" structure of the surrounding rock in underground engineering, and achieve complete identification of the surface and internal structural characteristic parameters of the surrounding rock on site.

[0043] By conducting on-site digital drilling tests of the surrounding rock, a formula for the evolution of the compressive strength of the surrounding rock during drilling was established using the monitored parameters, and the inversion curve of the compressive strength of the rock mass with the drilling depth was analyzed.

[0044] By collecting images of fractures on the surface and inside of the surrounding rock from different orientations, a CFS parameter discrimination analysis model and a LABELME image processing model were established. These models can identify the structural feature parameters of fractures on the surface and inside the surrounding rock, and form a set of structural feature parameters of fractures on the surface and inside the surrounding rock.

[0045] By preprocessing the image using MATLAB neural network, the coordinates of the three-dimensional key points of the internal and external fissures in the surrounding rock are obtained. The SURF function calculation method is used to form a three-dimensional model of the surrounding rock fissure structure, and the structure of the surrounding rock in the field is identified in all aspects. This solves the problems of difficulty in obtaining the characteristic parameters of the internal fissure structure of the surrounding rock and difficulty in combining the identification methods of the surface and internal structure of the surrounding rock.

[0046] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for identifying the structure of surrounding rock of underground engineering, characterized in that, The method comprises the following steps: Carrying out field digital drilling test of surrounding rock to obtain rock mass while drilling parameters, and collecting fracture images on the surface and inside of the surrounding rock; Analyzing the rock mass while drilling parameters to extract structural characteristic parameters of the fractures, analyzing the fracture images to obtain geometric characteristic parameters of the fractures, and combining the structural characteristic parameters and the geometric characteristic parameters to form a fracture characteristic parameter set; After pre-processing the fracture images, extracting fracture feature points and fracture contour lines of the fracture images, calculating three-dimensional key point coordinates of the fracture structure, and combining the fracture characteristic parameter set to construct a three-dimensional model of the fracture structure of the surrounding rock; Identifying the surface and internal structure of the surrounding rock based on the three-dimensional model of the fracture structure of the surrounding rock.

2. The method according to claim 1, wherein The step of analyzing the rock mass while drilling parameters to extract structural characteristic parameters of the fractures comprises the following steps: Based on the rock mass while drilling parameters, obtaining an inversion variation curve of the compressive strength of the rock mass with the drilling depth; And using the inversion variation curve to construct a fracture characteristic parameter discriminant analysis model, inputting the rock mass while drilling parameters into the fracture characteristic parameter discriminant analysis model, and extracting the structural characteristic parameters of the fractures.

3. The method for identifying the surface and internal structure of surrounding rock in underground engineering as described in claim 2, characterized in that, The structural characteristic parameters include fracture density, fracture width and fracture strike.

4. The method according to claim 1, wherein The step of analyzing the fracture images to obtain geometric characteristic parameters of the fractures comprises the following steps:

5. The method for identifying the surface and internal structure of surrounding rock in underground engineering as described in claim 4, characterized in that, Using an image processing model to mark the geometric characteristic parameters in the fracture images, and converting the fracture image information into fracture image label data.

6. The method according to claim 1, wherein The geometric characteristic parameters include fracture position, fracture area and fracture length.

7. The method according to claim 1 or 6, wherein The fracture images are obtained through borehole peeping and external camera shooting. The step of pre-processing the fracture images comprises the following steps:

8. The method for identifying the surface and internal structure of surrounding rock in underground engineering as described in claim 7, characterized in that, Enhancing the fracture images collected on the surface and inside of the surrounding rock, using a histogram equalization method to improve the contrast of the images, and then using a median filter algorithm to denoise the images.

9. The method according to claim 1, wherein the method is characterized by, When constructing the three-dimensional model of the fracture structure of the surrounding rock, the fractures are represented as geometric entities in three-dimensional space according to the spatial distribution and shape characteristics of the fractures, and modeling is performed.

10. The method according to claim 1, wherein the method is characterized by, In the field digital drilling test of the surrounding rock, fracture images on the surface and inside of the surrounding rock in different directions are collected. The step of extracting fracture feature points and fracture contour lines of the fracture images comprises the following steps: Using an image processing model to extract the fracture feature points and the fracture contour lines of the fracture images; Training the image processing model through a neural network, setting the fracture characteristic parameters in the fracture characteristic parameter set corresponding to the input layer nodes as input variables, passing through the nodes of the intermediate hidden layer, and setting the fracture feature points and the contour lines of the surrounding rock corresponding to the output layer nodes as output variables.

Images