Tunnel surrounding rock grading method and device

Abstract

The invention discloses a tunnel surrounding rock grading method and device. The method comprises the following steps: shooting a first tunnel image of a first tunnel face at at least one angle; the first tunnel face is intervened into the second tunnel face through excitation; shooting a second tunnel image of the second tunnel face at the angle; obtaining tunnel face change according to the first tunnel image and the second tunnel image; and grading the tunnel face by adopting a neural network model according to the face change. The method comprises the following steps: firstly, intervening a tunnel face by adopting active excitation to obtain geological information change of the tunnel face before and after intervention; and evaluating the change of the tunnel face by using the neural network model so as to make different grades on the variable degree of the current tunnel face of the tunnel surrounding rock.

Description

Technical Field

[0001] The present invention relates to the field of geological exploration, and in particular to a tunnel surrounding rock classification method and device. Background Art

[0002] There are many types of soils that tunnels pass through, mainly including rock, silt, clay, sandy soil, pebble soil, gravel soil, etc. Among them, rock is the hardest soil with high density and high compressive strength, but it also has special structures such as faults, joints, and caves; silt and clay have small particles, low density, and are easily affected by moisture, so attention should be paid to them in tunnel design and construction. Tunnel rock and soil show inherent rheological properties, and the deformation of the surrounding rock of the tunnel face also has certain rules. The excavation of the tunnel causes the surrounding rock supporting the tunnel body to be dug out, and there is an empty space behind the face, causing the surrounding rock to deform in the direction of the tunnel clearance. At the same time, with the construction of the tunnel, the current face will show characteristics such as deformation and rupture. As the tunnel construction progresses, using the deformation and rupture that may be shown on the face to evaluate and grade the face will be one of the effective measures to analyze the current face during the construction process to improve construction safety.

[0003] During the tunnel construction process, accidents caused by unclear classification of surrounding rock of the excavation face and improper judgment of safety status still occur frequently. The current situation that engineers do not accurately identify the geological characteristics around the tunnel, rely heavily on delayed geological survey reports, and use experience to determine the surrounding rock grade to provide direct guidance for tunnel construction safety judgment has not changed. At the same time, in the practice of traditional rock tunnels, there is a serious disconnect between support design and construction, and it is difficult to integrate dynamic design and construction. The process of detection, design, and construction is highly dependent on manpower, and there are problems such as difficulty in data acquisition, slow information transmission, slow analysis speed, and low accuracy. It is still a huge challenge to fully extract, analyze, and use the characteristics of the excavation face to identify and classify the surrounding rock of the tunnel and design efficient dynamic support. In addition, the early geological survey is difficult to reflect the overall engineering and hydrogeological characteristics of the tunnel, and the rich experience of engineers and the scale of manpower are difficult to match the rapidly developing scale of tunnel construction, which often leads to misjudgment of surrounding rock grade and unreasonable support schemes, causing engineering disasters such as surrounding rock collapse and sudden water and mud. Therefore, it is urgent to accurately, quickly and objectively identify and evaluate the geological information of the tunnel face and the surrounding rock grade during tunnel construction, so as to improve the level of judgment on the tunnel safety status and ensure construction safety. Summary of the invention

[0004] Based on this, a first aspect of an embodiment of the present invention discloses a tunnel surrounding rock classification method.

[0005] Wherein, the method comprises:

[0006] capturing a first tunnel image of a first tunnel face at at least one angle;

[0007] The first tunnel face is transformed into the second tunnel face by using vibration intervention;

[0008] photographing a second tunnel image of the second tunnel face at the angle;

[0009] Obtaining a tunnel face change according to the first tunnel image and the second tunnel image;

[0010] A neural network model is used to classify tunnel faces according to the face changes.

[0011] Wherein, obtaining the tunnel face change includes:

[0012] Obtaining a grayscale ratio image of the first tunnel image and the second tunnel image;

[0013] Using a clustering algorithm to obtain a clustering result of the grayscale ratio image;

[0014] Obtaining a clustering matrix representing the first tunnel image and the second tunnel image according to the clustering result;

[0015] Extracting the number of pixels of the clustering matrix;

[0016] The coordinate relationship of each pixel point of the clustering matrix is ​​extracted, and the number of at least paired or grouped pixel combinations in the clustering matrix is ​​obtained according to the adjacency relationship. The number of pixel combinations and the number of pixel points represent the change of the palm face.

[0017] Wherein, a first gradient image of the first tunnel image and a second gradient image of the second tunnel image are obtained by wavelet transform;

[0018] The grayscale ratio image is obtained according to the first gradient image and the second gradient image.

[0019] Wherein, the first gradient image and the second gradient image at a certain resolution are obtained by using Haar wavelet transform;

[0020] The resolution is selected based on a judgment of the clustering result.

[0021] Wherein, when the number of pixels in the clustering matrix deviates too much from the intensity and / or frequency of the excitation intervention, the resolution is reduced;

[0022] Or when the number of pixels in the clustering matrix deviates too little from the intensity and / or frequency of the excitation intervention, the resolution is increased.

[0023] Wherein, the FCMA clustering algorithm is used to obtain the clustering result of the grayscale ratio image.

[0024] Among them, the FCMA clustering algorithm includes:

[0025]

[0026] Among them, v j is the cluster center of the jth category, N is the total number of pixels in the grayscale ratio image, μ is the NxC matrix, μ ij For x i The degree of membership to category j, C is the number of clusters, m is the weight, ||x i -v j || represents the Euclidean distance from the i-th data to the j-th cluster center, ‖x i -v k ‖ represents the Euclidean distance from the i-th data to the k-th cluster center,

[0027] The value of C is 2, which is used to represent the clustering of changed pixels and unchanged pixels.

[0028] The value of m is a natural number greater than 1.

[0029] The value of m is selected based on the judgment of the clustering result.

[0030] Wherein, obtaining the tunnel face change includes:

[0031] Aggregating a preset number of neighborhood relationships to obtain a pixel region composed of pixels in the clustering matrix;

[0032] Calculate the area ratio of the maximum pixel area to the minimum pixel area;

[0033] The number of pixel combinations, the number of pixel points and the area ratio represent the change of the tunnel face.

[0034] A second aspect of the present invention discloses a tunnel surrounding rock classification device.

[0035] Wherein, the device comprises:

[0036] A first photographing unit, used for photographing a first tunnel image of a first tunnel face at at least one angle;

[0037] A vibration intervention device, used for transforming the first tunnel face into the second tunnel face by using vibration intervention;

[0038] A second photographing unit, configured to photograph a second tunnel image of the second tunnel face at the angle;

[0039] A change analysis unit, configured to obtain a tunnel face change according to the first tunnel image and the second tunnel image;

[0040] A change evaluation unit is used to classify the tunnel face using a neural network model according to the face change.

[0041] In view of the above-mentioned solution, the present invention describes the disclosed exemplary embodiments in detail below with reference to the accompanying drawings, so that other features and advantages of the embodiments of the present invention are also made clear. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 A schematic diagram showing the flow of the tunnel surrounding rock classification method of this embodiment is shown;

[0044] Figure 2 A schematic diagram of the process of obtaining the change of the tunnel face in this embodiment is shown;

[0045] Figure 3 A schematic diagram of the process of obtaining the change of the tunnel face according to the clustering matrix is ​​shown;

[0046] Figure 4 A schematic structural diagram of a tunnel surrounding rock grading device disclosed in this embodiment is shown. DETAILED DESCRIPTION

[0047] In order to facilitate the understanding of the present application, the present application will be described more fully below with reference to the relevant drawings. The drawings provide embodiments of the present application. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0049] This embodiment discloses a tunnel surrounding rock classification method, which is used to make a classification evaluation on the changes of the tunnel surrounding rock face during construction, and provide support for tunnel construction safety, support design, etc.

[0050] Figure 1 A schematic flow chart of the tunnel surrounding rock classification method according to this embodiment is shown.

[0051] Figure 1The method shown includes the following steps.

[0052] 10. Capture a first tunnel image of a first tunnel face at at least one angle.

[0053] 20 The first tunnel face was transformed into the second tunnel face by using vibration intervention.

[0054] 30 A second tunnel image of the second tunnel face is captured at an angle.

[0055] 40. Obtaining tunnel face changes according to the first tunnel image and the second tunnel image.

[0056] 50 Tunnel faces are classified using a neural network model according to face changes.

[0057] Therefore, the method of this embodiment first uses active excitation to intervene in the tunnel face to obtain changes in deformation, cracks, etc. of the tunnel face before and after the intervention; then the neural network model is used to evaluate the changes in the face to achieve different classifications of the degree of changeability of the current face of the tunnel surrounding rock.

[0058] Wherein, in step 10,

[0059] Shooting angle refers to shooting the tunnel face at a fixed angle in the tunnel to ensure that the visible features of the tunnel face can be retained and identified during shooting.

[0060] The first tunnel image of the first tunnel face refers to the visible light image of the tunnel face captured by visual equipment such as a camera. The "first" designation of the tunnel face is used to distinguish the face that has not changed before the excitation. The "first" designation of the tunnel image is used to distinguish the visible light image captured of the face that has not changed before the excitation.

[0061] Wherein, in step 20,

[0062] Vibration intervention refers to the transmission of vibration components in one or more directions along the longitudinal, transverse or depth of the tunnel face by directly or indirectly acting on the tunnel face through one or more vibration devices, so as to affect the entire or partial surface of the tunnel face, or affect the tunnel face to a certain depth inward, causing changes in the tunnel face, such as deformation, cracking, etc.

[0063] The vibration device may be a vibration hammer, which strikes the tunnel face vertically or obliquely once or repeatedly; the vibration device may also be a drilling device, which drills holes in the tunnel face vertically or obliquely. In short, the vibration device is used to stimulate the tunnel face in one or more directions to simulate possible tunnel construction on the tunnel face as much as possible.

[0064] Wherein, in step 30,

[0065] The first tunnel image of the second tunnel face refers to the "second" designation of the tunnel face used to distinguish the face that has changed after the excitation. The "second" designation of the tunnel image is used to distinguish the visible light image taken of the face that has changed after the excitation.

[0066] Wherein, in step 40,

[0067] The tunnel face change refers to the changes in cracks, deformations, etc. that can be directly observed from the visible light image after the excitation. This embodiment implements different classifications of the current tunnel face of the surrounding rock of the tunnel by analyzing the tunnel face change.

[0068] Figure 2 A schematic diagram of the process of obtaining the change of the tunnel face in this embodiment is shown.

[0069] Figure 2 It is shown that obtaining the tunnel face change includes the following steps.

[0070] 41 obtains the grayscale ratio image of the first tunnel image and the second tunnel image.

[0071] 42 The clustering algorithm is used to obtain the clustering results of the grayscale ratio image.

[0072] 43 obtains a clustering matrix representing the first tunnel image and the second tunnel image according to the clustering result.

[0073] 44 extracts the number of pixel points of the clustering matrix, extracts the coordinate relationship of each pixel point of the clustering matrix, and obtains the number of at least paired or grouped pixel combinations in the clustering matrix according to the adjacency relationship. The number of pixel combinations and the number of pixel points represent the change of the palm face.

[0074] Further, in step 41,

[0075] For pixels that have changed between the first tunnel image and the second tunnel image, the corresponding pixels in the grayscale ratio image are not 1.

[0076] Preferably, the value of each pixel in the grayscale ratio image can be logarithmically calculated or divided by a weighted value to compress the grayscale value of each pixel in the grayscale ratio image.

[0077] The grayscale ratio image may be obtained by first using wavelet transform to obtain a first gradient image of the first tunnel image and a second gradient image of the second tunnel image, and then obtaining the grayscale ratio image according to the first gradient image and the second gradient image.

[0078] Preferably, Haar wavelet transform is used to obtain the first gradient image and the second gradient image at a certain resolution.

[0079] In some embodiments, the resolution of the Haar wavelet transform can be determined and selected based on the clustering result of step 42. For example, after the Haar wavelet transform selects the lowest resolution in step 41, in step 42, it is found through manual observation that a large number of cracks and deformation protrusions in the clustering results are classified in the same way, and it can be considered to increase the resolution, or increase the level of resolution.

[0080] Furthermore, when the number of pixels in the clustering matrix of step 43 deviates too much from the intensity and / or frequency of the excitation intervention, it is also possible to consider reducing the resolution of step 41 , or when it deviates too little, it is also possible to consider increasing the resolution of step 41 .

[0081] Further, in step 42,

[0082] The FCMA clustering algorithm is used to obtain the clustering results of the grayscale ratio image.

[0083] Specifically, the FCMA clustering algorithm in this embodiment includes:

[0084]

[0085] Among them, v j is the cluster center of the jth category, N is the total number of pixels in the grayscale ratio image, μ is the NxC matrix, μ ij For x i The degree of membership to category j, C is the number of clusters, m is the weight, ‖x i -v j || represents the Euclidean distance from the i-th data to the j-th cluster center, ‖x i -v k ‖ represents the Euclidean distance from the i-th data to the k-th cluster center,

[0086] The value of C is 2, which is used to represent the clustering of changed pixels and unchanged pixels.

[0087] The value of m is a natural number greater than 1.

[0088] Preferably, the value of m can be selected based on the judgment of the clustering result of step 42.

[0089] For example, m is initially a random number, and the value of m is adjusted by observing the clustering results.

[0090] Further, in step 44,

[0091] Figure 3 A schematic diagram of the process of obtaining the change of the tunnel face according to the clustering matrix is ​​shown.

[0092] Figure 3 It is shown that obtaining the change of the tunnel face according to the clustering matrix includes the following steps.

[0093] 441 extracts the number of pixels of the clustering matrix.

[0094] 442 extracts the coordinate relationship of each pixel point in the clustering matrix, and obtains the number of at least paired or grouped pixel combinations in the clustering matrix according to the adjacency relationship.

[0095] 443 obtains a pixel region composed of each pixel group in the clustering matrix according to a preset number of neighborhood relationships; and calculates the area ratio of the maximum pixel region to the minimum pixel region.

[0096] 444 Characterize the changes in the face according to the number of pixel combinations, the number of pixel points and the area ratio.

[0097] Furthermore, in step 441, the number of pixels of the clustering matrix is ​​extracted, and the pixel number ratio of the pixels of the clustering matrix and the total pixel number ratio of the grayscale ratio image is compared. In step 444, the change of the face is characterized according to the number of pixel combinations, the pixel number ratio and the area ratio.

[0098] Wherein, in step 50,

[0099] The neural network model refers to the BP neural network model. The BP neural network model is a multi-layer feedforward neural network based on error back propagation training. The BP neural network model includes an input layer, a hidden layer and an output layer. The input layer of the BP neural network model contains multiple input neurons, and the multiple input neurons correspond to the input of the neural network model respectively. The BP neural network model has several hidden layers, and the hidden layer includes multiple hidden neurons. The output layer of the BP neural network model has 1 output neuron. The BP neural network model can use an S-type transfer function,

[0100] and the back propagation error function, Where Ti is the expected output and Oi is the calculated output of the network.

[0101] Furthermore, the BP neural network model is a fully connected BP neural network model, and its input layer may include 3 input neurons, corresponding to the number of pixel combinations, the ratio of the number of pixel points and the area of ​​the region respectively; the output layer may include 1 output neuron, and the output neuron corresponds to different levels of variability of the current face of the tunnel surrounding rock.

[0102] Of course, the neural network model of this embodiment can select other supervised neural network models and be obtained through training with a large number of samples.

[0103] Figure 4 A schematic structural diagram of a tunnel surrounding rock grading device disclosed in this embodiment is shown.

[0104] Figure 4The device shown includes a first shooting unit, a vibration intervention device, a second shooting unit, a change analysis unit, and a change evaluation unit. The first shooting unit is used to shoot a first tunnel image of a first tunnel face at at least one angle. The vibration intervention device is used to use vibration intervention to change the first tunnel face into a second tunnel face. The second shooting unit is used to shoot a second tunnel image of the second tunnel face at an angle. The change analysis unit is used to obtain a face change based on the first tunnel image and the second tunnel image. The change evaluation unit is used to classify tunnel faces by face changes through a neural network model.

[0105] Through the above description of the implementation method, technical personnel in the relevant field can clearly understand that the present invention can be implemented with the help of software and necessary general-purpose hardware, and of course can also be implemented through hardware, but in many cases the former is a better implementation method.

[0106] Based on such an understanding, the technical solution of the present invention can essentially and / or contribute to the prior art in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk and / or optical disk, etc., including a number of instructions for enabling a computer device (which can be a personal computer, server, and / or network device, etc.) to execute the methods of various embodiments of the present invention.

[0107] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention is described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A tunnel surrounding rock classification method, characterized in that: The method comprises, capturing a first tunnel image of a first tunnel face at at least one angle; The first tunnel face is transformed into the second tunnel face by using vibration intervention; photographing a second tunnel image of the second tunnel face at the angle; Obtaining a tunnel face change according to the first tunnel image and the second tunnel image; A neural network model is used to classify tunnel faces according to the face changes.

2. The tunnel surrounding rock classification method according to claim 1, characterized in that: Acquiring the tunnel face change includes: Obtaining a grayscale ratio image of the first tunnel image and the second tunnel image; Using a clustering algorithm to obtain a clustering result of the grayscale ratio image; Obtaining a clustering matrix representing the first tunnel image and the second tunnel image according to the clustering result; Extracting the number of pixels of the clustering matrix; The coordinate relationship of each pixel point of the clustering matrix is ​​extracted, and the number of at least paired or grouped pixel combinations in the clustering matrix is ​​obtained according to the adjacency relationship. The number of pixel combinations and the number of pixel points represent the change of the palm face.

3. The tunnel surrounding rock classification method according to claim 2, characterized in that: Using wavelet transform to obtain a first gradient image of the first tunnel image and a second gradient image of the second tunnel image; The grayscale ratio image is obtained according to the first gradient image and the second gradient image.

4. The tunnel surrounding rock classification method according to claim 3, characterized in that: Using Haar wavelet transform to obtain the first gradient image and the second gradient image at a certain resolution; The resolution is selected based on a judgment of the clustering result.

5. The tunnel surrounding rock classification method according to claim 4, characterized in that: When the number of pixels in the clustering matrix deviates too much from the intensity and / or frequency of the excitation intervention, the resolution is reduced, or When the number of pixels in the clustering matrix deviates too little from the intensity and / or frequency of the excitation intervention, the resolution is increased.

6. The tunnel surrounding rock classification method according to claim 2, characterized in that: The clustering result of the grayscale ratio image is obtained by using the FCMA clustering algorithm.

7. The tunnel surrounding rock classification method according to claim 6, characterized in that: The FCMA clustering algorithm includes: Among them, v j is the cluster center of the jth category, N is the total number of pixels in the grayscale ratio image, μ is the NxC matrix, μ ij For x i The degree of membership to category j, C is the number of clusters, m is the weight, ||x i -v j ‖ represents the Euclidean distance from the i-th data to the j-th cluster center, ‖x i -v k ‖ represents the Euclidean distance from the i-th data to the k-th cluster center, The value of C is 2, which is used to represent the clustering of changed pixels and unchanged pixels. The value of m is a natural number greater than 1.

8. The tunnel surrounding rock classification method according to claim 7, characterized in that: The value of m is selected based on the judgment of the clustering result.

9. The tunnel surrounding rock classification method according to claim 2, characterized in that: Acquiring the tunnel face change includes: Aggregating a preset number of neighborhood relationships to obtain a pixel region composed of pixels in the clustering matrix; Calculate the area ratio of the maximum pixel area to the minimum pixel area; The number of pixel combinations, the number of pixel points and the area ratio represent the change of the tunnel face.

10. A tunnel surrounding rock classification device, characterized in that: The device comprises, A first photographing unit, used for photographing a first tunnel image of a first tunnel face at at least one angle; A vibration intervention device, used for transforming the first tunnel face into the second tunnel face by using vibration intervention; A second photographing unit, configured to photograph a second tunnel image of the second tunnel face at the angle; A change analysis unit, configured to obtain a tunnel face change according to the first tunnel image and the second tunnel image; A change evaluation unit is used to classify the tunnel face using a neural network model according to the face change.

Images