Method and device for estimating the strength of the surrounding rock of a roadway supported by rock bolts

Abstract

The application provides an anchor rod supporting roadway surrounding rock strength estimation method and device, relates to the technical field of surrounding rock strength estimation method, and comprises the following steps: multiple monitoring sections are arranged at equal intervals along the axis of the roadway, two-dimensional data of wave velocity at the surrounding rock of each monitoring section is obtained, linear mapping is carried out according to the wave velocity value, an anchor rod monitoring area is set based on each monitoring section, and anchor rod parameters of the surrounding rock crack are obtained; a grayscale image is processed, a crack and anchor rod coupling algorithm is used to construct the surrounding rock strength coefficient of a single monitoring section; a time-dependent coupling is used to construct an anchor rod attenuation factor, the anchor rod attenuation factor is used as a correction term of the surrounding rock strength of a single monitoring section; and a spatial weighted average method is used to obtain the overall surrounding rock strength value. The crack features are automatically extracted by using a convolutional neural network, the problems of low efficiency and large error of the traditional method are solved, the anchor rod attenuation factor is dynamically corrected, the timeliness of early warning is significantly improved, and the roadway collapse risk is reduced.

Description

Technical Field

[0001] This invention relates to the technical field of methods for estimating the strength of surrounding rock, specifically to a method and apparatus for estimating the strength of surrounding rock in rock-supported roadways. Background Technology

[0002] Traditional measurement methods rely on manual experience or single physical parameters such as displacement and stress to assess the state of the surrounding rock, making it difficult to capture dynamic characteristics such as the density and direction of crack propagation in real time, resulting in delayed early warning. Existing technologies often analyze anchor bolt parameters such as diameter and spacing independently with the strength of the surrounding rock, ignoring the inhibitory effect of anchor bolt support on crack evolution and the cumulative impact of anchor bolt fatigue damage, resulting in biased strength assessment. The lack of an effective fusion mechanism for multi-dimensional data such as image features and anchor bolt parameters makes it difficult to construct a comprehensive strength model, which restricts the intelligence level of the monitoring system.

[0003] In the prior art, CN118395548A discloses a method that considers the process of rock bolt support and surrounding rock deformation and failure, making the collective effect of rock bolt support adapt to the deformation and failure of the surrounding rock in the roadway. It provides a method for determining the strength of the surrounding rock after rock bolt support during roadway excavation, using the rock bolt working load non-uniformity coefficient to determine the rock bolt support strength. The surrounding rock is divided into two cases: those with fracture zones and those without. If a fracture zone exists, the compressive strength of the rock bolt-supported surrounding rock is directly estimated; otherwise, the force-deformation relationship of the roadway surrounding rock support is predicted first, followed by the estimation of the compressive strength of the rock bolt-supported surrounding rock. However, this method lacks real-time analysis of fractures, does not consider the impact of rock bolt attenuation on the strength of the surrounding rock, and lacks a step from local monitoring to global safety early warning, failing to consider the correlation between the overall strength of the surrounding rock and fractures and rock bolts.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method and apparatus for estimating the strength of the surrounding rock in rock bolt-supported roadways, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The method for estimating the strength of the surrounding rock in rock bolt-supported roadways includes the following steps:

[0008] S1: Along the tunnel axis, multiple monitoring sections are set at equal intervals. Two-dimensional wave velocity data at the surrounding rock of each monitoring section are obtained. Linear mapping is performed based on the wave velocity value to generate a grayscale image of each monitoring section. Anchor bolt monitoring areas are set based on each monitoring section, and anchor bolt parameters within each monitoring area are obtained. The anchor bolt parameters include anchor bolt diameter, spacing between adjacent anchor bolts, axial stress, and shear strain.

[0009] S2: For the grayscale image corresponding to each monitoring section, feature extraction is performed using a convolutional neural network model. The features include a binary mask image. The fracture density is obtained by labeling the connected components of the binary mask image. The fracture mask in the binary mask image is skeletonized. The orientation angle of each fracture pixel is generated by local gradient calculation to obtain the orientation field vector image. Based on the orientation field vector image, the orientation angle of the main fracture is calculated using the vector averaging method.

[0010] S3: Construct the fracture influence components of fracture density and main fracture direction angle for each monitoring section, set the influence range of anchor bolts on surrounding rock fractures, construct the effective anchor bolt support distance, construct the anchor bolt influence component for each monitoring section based on the anchor bolt diameter and the effective anchor bolt support distance, and use linear coupling to integrate the fracture influence component and the anchor bolt influence component to obtain the surrounding rock strength coefficient of a single monitoring section.

[0011] S4: Based on the shear strain in the anchor bolt parameters, the fatigue coefficient of the anchor bolt material is set, and the fatigue coefficient is mapped to the anchor bolt attenuation factor through a nonlinear exponential function. The anchor bolt attenuation factor is used as a correction term for the surrounding rock strength of a single monitoring section to obtain the corrected surrounding rock strength of the monitoring section.

[0012] S5: The weighted average method is used to obtain the overall surrounding rock strength value. A safety threshold for the surrounding rock strength is set. The overall surrounding rock strength value is compared with the safety threshold. When it is lower than the safety threshold, the system issues an alarm.

[0013] Furthermore, two-dimensional wave velocity data at the surrounding rock of each monitoring section are acquired, and a grayscale image of each monitoring section is generated by linear mapping based on the wave velocity values. This process includes the following steps:

[0014] Each monitoring section is divided into equally spaced grid points, and the wave velocity value at each grid point is obtained to form a two-dimensional matrix:

[0015]

[0016] Indicates the first Two-dimensional distribution matrix of wave velocity at each monitoring section; Indicates the number of monitoring section indices; Indicates in Wave velocity value at that location; Indicates the first The horizontal coordinates of each horizontal grid point; Indicates the first Vertical coordinates of each vertical grid point;

[0017] The first Two-dimensional distribution matrix of wave velocity at each monitoring section Normalize:

[0018]

[0019] in,

[0020]

[0021]

[0022] in, Normalized Wave velocity value at that location; Indicates the first The minimum value of all wave velocities in a monitoring section; Indicates the first The maximum value of all wave velocities in each monitoring section;

[0023] Map the normalized wave velocity matrix to grayscale range:

[0024]

[0025] in, express The grayscale value at that location; This represents the rounding function;

[0026] Generate a grayscale matrix of the same size as the wave velocity matrix:

[0027]

[0028] in, Indicates the first The grayscale matrix of each monitoring section is obtained; the grayscale matrix is ​​then converted into an image through image recognition.

[0029] Furthermore, feature extraction is performed on the grayscale image corresponding to each monitoring section using a convolutional neural network model, specifically including the following steps:

[0030] Grayscale images were obtained as an image dataset. Cracks in the monitored sections of the image dataset were manually labeled. 80% of the images were used as the training set and 20% of the images were used as the test set. The training set was used as the input of the model and the binary mask image was used as the output label to train the convolutional neural network model.

[0031] A trained convolutional neural network model is obtained, the test set is used as the input of the model, and the binary mask is used as the output of the model. The crack density of the monitored section is obtained by calculating the connected components of the binary mask.

[0032] Furthermore, the crack density of the monitored section is obtained by performing connected component calculations on the binary mask image, specifically including the following steps:

[0033] Based on a binary mask image, connected components are labeled on the mask. The number of crack voxels and the total number of image voxels in the monitored section are statistically analyzed through image recognition, and the crack density of the monitored section is calculated.

[0034]

[0035] in,

[0036]

[0037]

[0038] Indicates the first The fracture density of the surrounding rock at each monitoring section; Indicates the first The number of fracture voxels in the surrounding rock monitoring section of each monitoring section; Indicates the first The total area of ​​the monitoring area of ​​each monitoring section; Indicates the first The crack area of ​​each monitoring section; Indicates the first The number of all voxels within the monitoring area of ​​each monitoring section; This represents the vertical direction of the monitoring section on the binary mask image, and the actual length corresponding to each pixel. This represents the horizontal direction of the monitoring section on the binary mask image, and the actual length corresponding to each pixel.

[0039] Furthermore, the crack mask of the monitoring section in the binary mask image is skeletonized, and the orientation angle of the crack pixel of each monitoring section is generated by local gradient calculation. The specific steps are as follows:

[0040] Repeated erosion operations are used to gradually remove the crack edge pixels of the monitored section until only the center line that can no longer be eroded is retained. After each erosion, it is checked whether the pixel belongs to the endpoint or intersection of the skeleton. If it does, it is not removed.

[0041] For each skeleton pixel, select In the neighborhood of a pixel, the horizontal and vertical gradients of each pixel are calculated using the Sobel operator. The formula for the orientation angle of a pixel can be expressed as:

[0042]

[0043] in, Represents the horizontal gradient; Represents the vertical gradient; Indicates the orientation angle of a pixel;

[0044] For the orientation angle of a pixel Calculate the horizontal and vertical components of the unit vector:

[0045]

[0046] in, This represents the projection of the direction vector onto the horizontal axis; This represents the projection of the direction vector onto the vertical axis.

[0047] Furthermore, the fracture orientation angle of the main monitoring section is calculated using the vector averaging method. The specific steps are as follows:

[0048] Based on the orientation field vector map, only the vector values ​​of the crack region in the monitored section are retained to calculate the principal orientation angle:

[0049]

[0050] in,

[0051]

[0052]

[0053] Indicates the fracture region of the monitoring section. The mean; Indicates the fracture region of the monitoring section. The mean; This represents the total number of fracture region vectors at the monitored section; Indicates the vector index of the fracture region at the monitoring section; Indicates the first The main direction angle of each monitoring section.

[0054] Furthermore, linear coupling is used to synthesize the fracture influence component and the anchor bolt influence component of the monitoring section to obtain the surrounding rock strength coefficient of a single monitoring section. The specific steps are as follows:

[0055] The influence range of the anchor bolt on the cracks in the surrounding rock monitoring section is set to the area on both sides of the crack line at a distance from the main monitoring section. A strip-shaped area of ​​rice:

[0056]

[0057] in,

[0058]

[0059] This indicates the effective influence distance of the anchor bolt on the crack in the main monitoring section; Indicates the original anchor bolt design spacing; This represents the floor function; Indicates the uniaxial compressive strength of the surrounding rock; Indicates the effective anchor bolt support distance; This represents the anchor bolt support efficiency coefficient; Indicates the diameter of the anchor bolt; Indicates the first [number] before the correction The surrounding rock strength coefficient of each monitoring section; This indicates axial stress.

[0060] Furthermore, the anchor bolt attenuation factor is used as a correction term for the surrounding rock strength of a single monitoring section to obtain the corrected surrounding rock strength of the monitoring section. This process specifically includes the following steps:

[0061] Define the fatigue coefficient of the anchor bolt material, and map the fatigue coefficient to the anchor bolt attenuation factor through a nonlinear exponential function:

[0062]

[0063] in,

[0064]

[0065] Indicates the anchor bolt attenuation factor; Indicates the support time of the anchor bolt; Indicates the fatigue coefficient of the anchor bolt material; Indicates the first Shear strain of anchor bolts at each monitoring section; Indicates the first The shear strain of the anchor bolt at each monitoring section is 1.5 times the power of the shear strain.

[0066] The corrected surrounding rock strength of the monitoring section can be expressed as:

[0067]

[0068] This indicates the strength of the surrounding rock at the corrected monitoring section; Indicates the weighting coefficient for surrounding rock strength; This represents the strain correction factor.

[0069] Furthermore, the overall surrounding rock strength value is obtained using a weighted average method, a safe threshold for the surrounding rock strength is set, and the overall surrounding rock strength value is compared with the safe threshold. This process specifically includes the following steps:

[0070] The weighted average method is used to obtain the surrounding rock strength value, which is then compared with a set safety threshold. If the surrounding rock strength is less than the safety threshold, an alarm is issued; otherwise, it indicates normal operation. The specific expression is as follows:

[0071]

[0072] in,

[0073]

[0074] in, Indicates an alarm signal; Indicates the safety threshold; This indicates the total number of monitoring sections.

[0075] The present invention also provides an apparatus for estimating the strength of the surrounding rock in a rock bolt-supported roadway, the apparatus being used in the above-described estimation method, comprising:

[0076] The data acquisition module is used to set up multiple monitoring sections at equal intervals along the tunnel axis, acquire two-dimensional wave velocity data at the surrounding rock of each monitoring section, perform linear mapping based on the wave velocity value to generate a grayscale image of each monitoring section, set an anchor monitoring area based on each monitoring section, and acquire the anchor parameters in each monitoring area. The anchor parameters include the anchor diameter, the spacing between adjacent anchors, axial stress, and shear strain.

[0077] The preprocessing module extracts features from the grayscale image corresponding to each monitoring section using a convolutional neural network model. Features include a binary mask image. The module obtains the fracture density by labeling connected components in the binary mask image, skeletonizes the fracture mask in the binary mask image, and generates the orientation angle of each fracture pixel using local gradient calculation, resulting in an orientation field vector map. Based on the orientation field vector map, the orientation angle of the main fracture is calculated using the vector averaging method. The surrounding rock strength analysis module constructs the surrounding rock strength coefficient of a single monitoring section based on the fracture density, the fracture orientation angle of the main monitoring section, and the anchor bolt parameters, using a fracture-anchor bolt coupling algorithm.

[0078] The surrounding rock strength module is used to construct the fracture influence components of fracture density and main fracture direction angle for each monitoring section, set the influence range of anchor bolts on surrounding rock fractures, construct the effective anchor bolt support distance, construct the anchor bolt influence component for each monitoring section based on the anchor bolt diameter and the effective anchor bolt support distance, and use linear coupling to integrate the fracture influence component and the anchor bolt influence component to obtain the surrounding rock strength coefficient of a single monitoring section.

[0079] The strength correction module is used to set the fatigue coefficient of the anchor material based on the shear strain in the anchor parameters, map the fatigue coefficient to the anchor attenuation factor through a nonlinear exponential function, and use the anchor attenuation factor as a correction term for the surrounding rock strength of a single monitoring section to obtain the corrected surrounding rock strength of the monitoring section.

[0080] The threshold comparison module is used to obtain the overall surrounding rock strength value using a weighted average method, set a safe threshold for the surrounding rock strength, compare the overall surrounding rock strength value with the safe threshold, and issue an alarm when the value is lower than the safe threshold.

[0081] Compared with the prior art, the beneficial effects of the present invention are:

[0082] Based on convolutional neural networks and directional field vector maps, the system automatically extracts fracture density and principal orientation angles, overcoming the limitations of low efficiency and large errors in traditional manual interpretation, and improving the real-time performance and accuracy of fracture evolution analysis. By linearly coupling the fracture influence component and the anchor bolt influence component, the system quantifies the enhancing effect of anchor bolt support on the surrounding rock strength. Combined with the dynamic correction strength model of the anchor bolt attenuation factor, the system solves the problem of coupled analysis of anchor bolt fatigue damage and surrounding rock stability. By integrating multi-dimensional data such as wave velocity field grayscale images, anchor bolt parameters, and fatigue coefficients, a weighted average surrounding rock strength assessment system is constructed, achieving a leap from local monitoring to global safety early warning, and providing a reliable basis for roadway support optimization. Attached Figure Description

[0083] Figure 1 This is a schematic diagram of the overall method flow of the present invention.

[0084] Figure 2 This is a graph showing the relationship between the surrounding rock strength coefficient and the corrected surrounding rock strength coefficient.

[0085] Figure 3 This is a graph showing the relationship between shear strain and the corrected surrounding rock strength coefficient.

[0086] Figure 4 A graph showing the relationship between the anchor bolt damage factor and the corrected surrounding rock strength coefficient.

[0087] Figure 5 This is a schematic diagram of the overall device flow of the present invention. Detailed Implementation

[0088] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0089] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0090] Example:

[0091] Please see Figures 1-4 The present invention provides a technical solution:

[0092] The method for estimating the strength of the surrounding rock in rock bolt-supported roadways includes the following steps:

[0093] S1: Along the tunnel axis, multiple monitoring sections are set at equal intervals. Two-dimensional wave velocity data at the surrounding rock of each monitoring section are obtained. Linear mapping is performed based on the wave velocity value to generate a grayscale image of each monitoring section. Anchor bolt monitoring areas are set based on each monitoring section, and anchor bolt parameters within each monitoring area are obtained. The anchor bolt parameters include anchor bolt diameter, spacing between adjacent anchor bolts, axial stress, and shear strain.

[0094] The process of acquiring two-dimensional wave velocity data at the surrounding rock of each monitoring section, and generating a grayscale image of each monitoring section by linear mapping based on the wave velocity values, specifically includes the following steps:

[0095] Each monitoring section is divided into equally spaced grid points, and the wave velocity value at each grid point is obtained to form a two-dimensional matrix:

[0096]

[0097] Indicates the first Two-dimensional distribution matrix of wave velocity at each monitoring section; Indicates the number of monitoring section indices; Indicates in Wave velocity value at that location; Indicates the first The horizontal coordinates of each horizontal grid point; Indicates the first Vertical coordinates of each vertical grid point;

[0098] The first Two-dimensional distribution matrix of wave velocity at each monitoring section Normalize:

[0099]

[0100] in,

[0101]

[0102]

[0103] in, Normalized Wave velocity value at that location; Indicates the first The minimum value of all wave velocities in a monitoring section; Indicates the first The maximum value of all wave velocities in each monitoring section;

[0104] Map the normalized wave velocity matrix to grayscale range:

[0105]

[0106] in, express The grayscale value at that location; This represents the rounding function;

[0107] Directly decided brightness, The larger the value, the closer the grayscale value is to 255, which reflects better integrity of the surrounding rock;

[0108] Generate a grayscale matrix of the same size as the wave velocity matrix:

[0109]

[0110] in, Indicates the first The grayscale matrix of each monitoring section is obtained; the grayscale matrix is ​​then converted into an image through image recognition.

[0111] Will Each element in The image is mapped to a two-dimensional image grid based on row and column positions, where the matrix rows correspond to the vertical direction of the image and the columns correspond to the horizontal direction. Subsequently, an 8-bit grayscale image is generated based on the grayscale value range, where 0 represents pure black, 255 represents pure white, and intermediate values ​​correspond to different gray levels, ultimately forming a digital image consistent with the wave velocity distribution space.

[0112] S2: For the grayscale image corresponding to each monitoring section, feature extraction is performed using a convolutional neural network model. The features include a binary mask image. The fracture density is obtained by labeling the connected components of the binary mask image. The fracture mask in the binary mask image is skeletonized. The orientation angle of each fracture pixel is generated by local gradient calculation to obtain the orientation field vector image. Based on the orientation field vector image, the orientation angle of the main fracture is calculated using the vector averaging method.

[0113] The grayscale image corresponding to each monitoring section is used to extract features through a convolutional neural network model, specifically including the following steps:

[0114] Grayscale images were obtained as an image dataset. Cracks in the monitored sections of the image dataset were manually labeled. 80% of the images were used as the training set and 20% of the images were used as the test set. The training set was used as the input of the model and the binary mask image was used as the output label to train the convolutional neural network model.

[0115] A trained convolutional neural network model is obtained, the test set is used as the input of the model, and the binary mask is used as the output of the model. The crack density of the monitored section is obtained by calculating the connected components of the binary mask.

[0116] The method of obtaining the crack density of the monitored section by performing connected component calculation on the binary mask image specifically includes the following steps:

[0117] Based on a binary mask image, connected components are labeled on the mask. The number of crack voxels and the total number of image voxels in the monitored section are statistically analyzed through image recognition, and the crack density of the monitored section is calculated.

[0118]

[0119] in,

[0120]

[0121]

[0122] Indicates the first The fracture density of the surrounding rock at each monitoring section; Indicates the first The number of fracture voxels in the surrounding rock monitoring section of each monitoring section; Indicates the first The total area of ​​the monitoring area of ​​each monitoring section; Indicates the first The crack area of ​​each monitoring section; Indicates the first The number of all voxels within the monitoring area of ​​each monitoring section; This represents the vertical direction of the monitoring section on the binary mask image, and the actual length corresponding to each pixel. This represents the horizontal direction of the monitoring section on the binary mask image, and the actual length corresponding to each pixel.

[0123] The process of skeletonizing the crack mask of the monitoring section in the binary mask image and generating the orientation angle of the crack pixel of each monitoring section using local gradient calculation is as follows:

[0124] Repeated erosion operations are used to gradually remove the crack edge pixels of the monitored section until only the center line that can no longer be eroded is retained. After each erosion, it is checked whether the pixel belongs to the endpoint or intersection of the skeleton. If it does, it is not removed.

[0125] For each skeleton pixel, select In the neighborhood of a pixel, the horizontal and vertical gradients of each pixel are calculated using the Sobel operator. The formula for the orientation angle of a pixel can be expressed as:

[0126]

[0127] in, Represents the horizontal gradient; Represents the vertical gradient; Indicates the orientation angle of a pixel;

[0128] For the orientation angle of a pixel Calculate the horizontal and vertical components of the unit vector:

[0129]

[0130] in, This represents the projection of the direction vector onto the horizontal axis; This represents the projection of the direction vector onto the vertical axis.

[0131] The method of calculating the fracture orientation angle of the main monitoring section using the vector averaging method involves the following steps:

[0132] Based on the orientation field vector map, only the vector values ​​of the crack region in the monitored section are retained to calculate the principal orientation angle:

[0133]

[0134] in,

[0135]

[0136]

[0137] Indicates the fracture region of the monitoring section. The mean; Indicates the fracture region of the monitoring section. The mean; This represents the total number of fracture region vectors at the monitored section; Indicates the vector index of the fracture region at the monitoring section; Indicates the first The main direction angle of each monitoring section.

[0138] S3: Construct the fracture influence components of fracture density and main fracture direction angle for each monitoring section, set the influence range of anchor bolts on surrounding rock fractures, construct the effective anchor bolt support distance, construct the anchor bolt influence component for each monitoring section based on the anchor bolt diameter and the effective anchor bolt support distance, and use linear coupling to integrate the fracture influence component and the anchor bolt influence component to obtain the surrounding rock strength coefficient of a single monitoring section.

[0139] The method of using linear coupling to synthesize the fracture influence component and the anchor bolt influence component of the monitoring section to obtain the surrounding rock strength coefficient of a single monitoring section is as follows:

[0140] The influence range of the anchor bolt on the cracks in the surrounding rock monitoring section is set to the area on both sides of the crack line at a distance from the main monitoring section. A strip-shaped area of ​​rice:

[0141]

[0142] in,

[0143]

[0144] This indicates the effective influence distance of the anchor bolt on the crack in the main monitoring section; Indicates the original anchor bolt design spacing; This represents the floor function; Indicates the uniaxial compressive strength of the surrounding rock; Indicates the effective anchor bolt support distance; This represents the anchor bolt support efficiency coefficient; Indicates the diameter of the anchor bolt; Indicates the first [number] before the correction The surrounding rock strength coefficient of each monitoring section; This indicates axial stress.

[0145] The above formula comprehensively reflects the synergistic enhancement effect of the surrounding rock's own compressive strength and the anchor bolt support effect on the stability of the surrounding rock. By quantifying the surrounding rock-anchor bolt coupling strength, it provides a direct quantitative index for the optimization of support parameters, avoiding the subjective errors of traditional empirical methods. and Characterizes the positive contribution of the surrounding rock's own strength and the fracture orientation angle to stability; The strength of the surrounding rock is weakened by the denominator term; the higher the fracture density, the lower the strength. and Positively correlated with the tensile strength of the anchor bolt This reflects the decrease in support effect caused by the increase in anchor bolt spacing; along with , , , , It increases with the increase of, and with and The increase leads to a decrease;

[0146] Based on the effective influence distance of the anchor bolts on the cracks and the original design spacing, a reasonable spacing that meets the requirements of uniform support of the surrounding rock is generated by dynamic adjustment. along with The larger the effective influence distance, the more the actual spacing needs to be appropriately widened to cover a larger area. along with Increasing the spacing in the original design means that the actual spacing needs to be more compact after adjustment to compensate for the support strength; this is affected by the rounding function. and , It changes in a segmented, stepped manner, with the final spacing value strictly meeting the requirements. Ensure that the support strength is not lower than the design requirements.

[0147] S4: Based on the shear strain in the anchor bolt parameters, the fatigue coefficient of the anchor bolt material is set, and the fatigue coefficient is mapped to the anchor bolt attenuation factor through a nonlinear exponential function. The anchor bolt attenuation factor is used as a correction term for the surrounding rock strength of a single monitoring section to obtain the corrected surrounding rock strength of the monitoring section.

[0148] The step of using the anchor bolt attenuation factor as a correction term for the surrounding rock strength of a single monitoring section to obtain the corrected surrounding rock strength of the monitoring section specifically includes the following steps:

[0149] Define the fatigue coefficient of the anchor bolt material, and map the fatigue coefficient to the anchor bolt attenuation factor through a nonlinear exponential function:

[0150]

[0151] in,

[0152]

[0153] Indicates the anchor bolt attenuation factor; Indicates the support time of the anchor bolt; Indicates the fatigue coefficient of the anchor bolt material; Indicates the first Shear strain of anchor bolts at each monitoring section; Indicates the first The shear strain of the anchor bolt at each monitoring section is 1.5 times the power of the shear strain.

[0154] The above formula quantifies the performance of anchor bolt support over time. and shear strain The degree of degradation reflects the contribution of the anchor bolt to the long-term stability of the surrounding rock; Through fatigue coefficient The greater the shear strain, the more significant the fatigue damage accumulation, which affects the attenuation rate. Through the index term The attenuation rate can be directly controlled; the longer the support time, the more severe the performance degradation of the anchor bolt. It reflects the nonlinear amplification effect of shear strain on fatigue damage, which is consistent with the law of plastic deformation of materials; Follow As the growth rate increases, the exponential decreases, and the extended time leads to accelerated performance degradation. Follow Increased shear strain leads to decreased shear strain, which exacerbates fatigue damage.

[0155] The corrected surrounding rock strength of the monitoring section can be expressed as:

[0156]

[0157] This indicates the strength of the surrounding rock at the corrected monitoring section; Indicates the weighting coefficient for surrounding rock strength; This represents the strain correction factor.

[0158] Through linear terms Provides the basic value of the surrounding rock strength, weighting coefficient. Control its contribution ratio; pass Nonlinear correction of surrounding rock strength, strain correction factor Adjusting the sensitivity to strain effects; As a global attenuation factor, it reflects the overall weakening effect of anchor performance degradation on the surrounding rock strength; logarithmic function It is used to mitigate the abrupt effects of shear strain, which is consistent with the characteristics of progressive failure of surrounding rock; along with , , , The increase in strength leads to an increase in the original strength, weighting coefficient, and strain correction term, all of which enhance the surrounding rock strength assessment value. along with As the anchor bolt attenuation factor decreases, the surrounding rock strength assessment value increases.

[0159] In this embodiment, the following parameters are fixed: , Only change Observe the value. The changes in , and the experimentally labeled data are shown in Table 1:

[0160] Table 1: and Linear positive correlation data table

[0161]

[0162] In the above set of data, when , When it remains unchanged, only change The size can be seen The growth of is linear, reflecting that the corrected surrounding rock strength coefficient increases as the uncorrected surrounding rock strength coefficient increases;

[0163] Fixed parameters: , Only change Observe the value. The changes in , and the experimentally labeled data are shown in Table 2:

[0164] Table 2: and Nonlinear data table

[0165]

[0166] In the above set of data, when , When it remains unchanged, only change The value can be seen from the value. Follow The increase in shear strain reflects the influence of shear strain on the strength of the surrounding rock.

[0167] Fixed parameters: , Only change Observe the value. The changes in , and the experimentally labeled data are shown in Table 3:

[0168] Table 3: and Linear positive correlation data table

[0169]

[0170] In the above set of data, , Only change The value can be seen from the value. Follow The increase and decrease of the anchor bolt damage factor reflect the influence of the anchor bolt damage factor on the strength of the surrounding rock.

[0171] S5: The weighted average method is used to obtain the overall surrounding rock strength value. A safety threshold for the surrounding rock strength is set. The overall surrounding rock strength value is compared with the safety threshold. When it is lower than the safety threshold, the system issues an alarm.

[0172] The process of obtaining the overall surrounding rock strength value using a weighted average method, setting a safe threshold for the surrounding rock strength, and comparing the overall surrounding rock strength value with the safe threshold specifically includes the following steps:

[0173] The weighted average method is used to obtain the surrounding rock strength value, which is then compared with a set safety threshold. If the surrounding rock strength is less than the safety threshold, an alarm is issued; otherwise, it indicates normal operation. The specific expression is as follows:

[0174]

[0175] in,

[0176]

[0177] in, Indicates an alarm signal; Indicates the safety threshold; This indicates the total number of monitoring sections.

[0178] The present invention also provides an apparatus for estimating the strength of surrounding rock in rock-supported roadways, the apparatus being used to perform the above-described estimation method, comprising:

[0179] The data acquisition module is used to set up multiple monitoring sections at equal intervals along the tunnel axis, acquire two-dimensional wave velocity data at the surrounding rock of each monitoring section, perform linear mapping based on the wave velocity value to generate a grayscale image of each monitoring section, set an anchor monitoring area based on each monitoring section, and acquire the anchor parameters in each monitoring area. The anchor parameters include the anchor diameter, the spacing between adjacent anchors, axial stress, and shear strain.

[0180] The preprocessing module extracts features from the grayscale image corresponding to each monitoring section using a convolutional neural network model. Features include a binary mask image. The module obtains the fracture density by labeling connected components in the binary mask image, skeletonizes the fracture mask in the binary mask image, and generates the orientation angle of each fracture pixel using local gradient calculation, resulting in an orientation field vector map. Based on the orientation field vector map, the orientation angle of the main fracture is calculated using the vector averaging method. The surrounding rock strength analysis module constructs the surrounding rock strength coefficient of a single monitoring section based on the fracture density, the fracture orientation angle of the main monitoring section, and the anchor bolt parameters, using a fracture-anchor bolt coupling algorithm.

[0181] The surrounding rock strength module is used to construct the fracture influence components of fracture density and main fracture direction angle for each monitoring section, set the influence range of anchor bolts on surrounding rock fractures, construct the effective anchor bolt support distance, construct the anchor bolt influence component for each monitoring section based on the anchor bolt diameter and the effective anchor bolt support distance, and use linear coupling to integrate the fracture influence component and the anchor bolt influence component to obtain the surrounding rock strength coefficient of a single monitoring section.

[0182] The strength correction module is used to set the fatigue coefficient of the anchor material based on the shear strain in the anchor parameters, map the fatigue coefficient to the anchor attenuation factor through a nonlinear exponential function, and use the anchor attenuation factor as a correction term for the surrounding rock strength of a single monitoring section to obtain the corrected surrounding rock strength of the monitoring section.

[0183] The threshold comparison module is used to obtain the overall surrounding rock strength value using a weighted average method, set a safe threshold for the surrounding rock strength, compare the overall surrounding rock strength value with the safe threshold, and issue an alarm when the value is lower than the safe threshold.

[0184] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0185] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0186] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0187] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for estimating the strength of surrounding rock in rock bolt-supported roadways, characterized by the following steps: include: S1: Along the tunnel axis, multiple monitoring sections are set at equal intervals. Two-dimensional wave velocity data at the surrounding rock of each monitoring section are obtained. Linear mapping is performed based on the wave velocity value to generate a grayscale image of each monitoring section. Anchor bolt monitoring areas are set based on each monitoring section, and anchor bolt parameters within each monitoring area are obtained. The anchor bolt parameters include anchor bolt diameter, spacing between adjacent anchor bolts, axial stress, and shear strain. S2: For the grayscale image corresponding to each monitoring section, feature extraction is performed using a convolutional neural network model. The features include a binary mask image. The fracture density is obtained by labeling the connected components of the binary mask image. The fracture mask in the binary mask image is skeletonized. The orientation angle of each fracture pixel is generated by local gradient calculation to obtain the orientation field vector image. Based on the orientation field vector image, the orientation angle of the main fracture is calculated using the vector averaging method. S3: Construct the fracture influence components of fracture density and main fracture direction angle for each monitoring section, set the influence range of anchor bolts on surrounding rock fractures, construct the effective anchor bolt support distance, construct the anchor bolt influence component for each monitoring section based on the anchor bolt diameter and the effective anchor bolt support distance, and use linear coupling to integrate the fracture influence component and the anchor bolt influence component to obtain the surrounding rock strength coefficient of a single monitoring section. S4: Based on the shear strain in the anchor bolt parameters, the fatigue coefficient of the anchor bolt material is set, and the fatigue coefficient is mapped to the anchor bolt attenuation factor through a nonlinear exponential function. The anchor bolt attenuation factor is used as a correction term for the surrounding rock strength of a single monitoring section to obtain the corrected surrounding rock strength of the monitoring section. S5: The weighted average method is used to obtain the overall surrounding rock strength value. A safety threshold for the surrounding rock strength is set. The overall surrounding rock strength value is compared with the safety threshold. When it is lower than the safety threshold, the system issues an alarm.

2. The method for estimating the strength of surrounding rock in a rock bolt-supported roadway according to claim 1, characterized in that, The process of acquiring two-dimensional wave velocity data at the surrounding rock of each monitoring section, and generating a grayscale image of each monitoring section by linear mapping based on the wave velocity values, specifically includes the following steps: Each monitoring section is divided into equally spaced grid points, and the wave velocity value at each grid point is obtained to form a two-dimensional matrix: Indicates the first Two-dimensional distribution matrix of wave velocity at each monitoring section; Indicates the number of monitoring section indices; Indicates in Wave velocity value at that location; Indicates the first The horizontal coordinates of each horizontal grid point; Indicates the first The vertical coordinates of each vertical grid point; The first Two-dimensional distribution matrix of wave velocity at each monitoring section Normalize: in, in, Normalized Wave velocity value at that location; Indicates the first The minimum value of all wave velocities in a monitoring section; Indicates the first The maximum value of all wave velocities in each monitoring section; Map the normalized wave velocity matrix to grayscale range: in, express The grayscale value at that location; This represents the rounding function; Generate a grayscale matrix of the same size as the wave velocity matrix: in, Indicates the first The grayscale matrix of each monitoring section is obtained; the grayscale matrix is ​​then converted into an image through image recognition.

3. The method for estimating the strength of surrounding rock in a roadway supported by anchor bolts according to claim 1, characterized in that, The grayscale image corresponding to each monitoring section is used to extract features through a convolutional neural network model, specifically including the following steps: Grayscale images were obtained as an image dataset. Cracks in the monitored sections of the image dataset were manually labeled. 80% of the images were used as the training set and 20% of the images were used as the test set. The training set was used as the input of the model and the binary mask image was used as the output label to train the convolutional neural network model. A trained convolutional neural network model is obtained, the test set is used as the input of the model, and the binary mask is used as the output of the model. The crack density of the monitored section is obtained by calculating the connected components of the binary mask.

4. The method for estimating the strength of surrounding rock in a rock bolt-supported roadway according to claim 3, characterized in that, The method of obtaining the crack density of the monitored section by performing connected component calculation on the binary mask image specifically includes the following steps: Based on a binary mask image, connected components are labeled on the mask. The number of crack voxels and the total number of image voxels in the monitored section are statistically analyzed through image recognition, and the crack density of the monitored section is calculated. in, Indicates the first The fracture density of the surrounding rock at each monitoring section; Indicates the first The number of fracture voxels in the surrounding rock monitoring section of each monitoring section; Indicates the first The total area of ​​the monitoring area of ​​each monitoring section; Indicates the first The crack area of ​​each monitoring section; Indicates the first The number of all voxels within the monitoring area of ​​each monitoring section; This represents the vertical direction of the monitoring section on the binary mask image, and the actual length corresponding to each pixel. This represents the horizontal direction of the monitoring section on the binary mask image, and the actual length corresponding to each pixel.

5. The method for estimating the strength of surrounding rock in a roadway supported by anchor bolts according to claim 1, characterized in that, The process of skeletonizing the crack mask of the monitoring section in the binary mask image and generating the orientation angle of the crack pixel of each monitoring section using local gradient calculation is as follows: Repeated erosion operations are used to gradually remove the crack edge pixels of the monitored section until only the center line that can no longer be eroded is retained. After each erosion, it is checked whether the pixel belongs to the endpoint or intersection of the skeleton. If it does, it is not removed. For each skeleton pixel, select In the neighborhood of a pixel, the horizontal and vertical gradients of each pixel are calculated using the Sobel operator. The formula for the orientation angle of a pixel can be expressed as: in, Represents the horizontal gradient; Represents the vertical gradient; Indicates the orientation angle of a pixel; For the orientation angle of a pixel Calculate the horizontal and vertical components of the unit vector: in, This represents the projection of the direction vector onto the horizontal axis; This represents the projection of the direction vector onto the vertical axis.

6. The method for estimating the strength of surrounding rock in a roadway supported by anchor bolts according to claim 1, characterized in that, The method of calculating the fracture orientation angle of the main monitoring section using the vector averaging method involves the following steps: Based on the orientation field vector map, only the vector values ​​of the crack region in the monitored section are retained to calculate the principal orientation angle: in, Indicates the fracture region of the monitoring section. The mean; Indicates the fracture region of the monitoring section. The mean; This represents the total number of fracture region vectors at the monitored section; Indicates the vector index of the fracture region at the monitoring section; Indicates the first The main direction angle of each monitoring section.

7. The method for estimating the strength of surrounding rock in a roadway supported by anchor bolts according to claim 1, characterized in that, The method of using linear coupling to synthesize the fracture influence component and the anchor bolt influence component of the monitoring section to obtain the surrounding rock strength coefficient of a single monitoring section is as follows: The influence range of the anchor bolt on the cracks in the surrounding rock monitoring section is set to the area on both sides of the crack line at a distance from the main monitoring section. A strip-shaped area of ​​rice: in, This indicates the effective influence distance of the anchor bolt on the crack in the main monitoring section; Indicates the original anchor bolt design spacing; This represents the floor function; Indicates the uniaxial compressive strength of the surrounding rock; Indicates the effective anchor bolt support distance; This represents the anchor bolt support efficiency coefficient; Indicates the diameter of the anchor bolt; Indicates the first [number] before the correction The surrounding rock strength coefficient of each monitoring section; ; This indicates axial stress.

8. The method for estimating the strength of surrounding rock in a roadway supported by anchor bolts according to claim 1, characterized in that, The step of using the anchor bolt attenuation factor as a correction term for the surrounding rock strength of a single monitoring section to obtain the corrected surrounding rock strength of the monitoring section specifically includes the following steps: Define the fatigue coefficient of the anchor bolt material, and map the fatigue coefficient to the anchor bolt attenuation factor through a nonlinear exponential function: in, Indicates the anchor bolt attenuation factor; Indicates the support time of the anchor bolt; Indicates the fatigue coefficient of the anchor bolt material; Indicates the first Shear strain of anchor bolts at each monitoring section; Indicates the first The shear strain of the anchor bolt at each monitoring section is 1.5 times the power of the shear strain. The corrected surrounding rock strength of the monitoring section can be expressed as: This indicates the strength of the surrounding rock at the corrected monitoring section; Indicates the weighting coefficient for surrounding rock strength; This represents the strain correction factor.

9. The method for estimating the strength of surrounding rock in a roadway supported by anchor bolts according to claim 1, characterized in that, The process of obtaining the overall surrounding rock strength value using a weighted average method, setting a safe threshold for the surrounding rock strength, and comparing the overall surrounding rock strength value with the safe threshold specifically includes the following steps: The weighted average method is used to obtain the surrounding rock strength value, which is then compared with a set safety threshold. If the surrounding rock strength is less than the safety threshold, an alarm is issued; otherwise, it indicates normal operation. The specific expression is as follows: in, in, Indicates an alarm signal; Indicates the safety threshold; This indicates the total number of monitoring sections.

10. A device for estimating the strength of surrounding rock in rock bolt-supported roadways, characterized in that: The estimation device is used to perform the estimation method according to any one of claims 1-9, including: The data acquisition module is used to set up multiple monitoring sections at equal intervals along the tunnel axis, acquire two-dimensional wave velocity data at the surrounding rock of each monitoring section, perform linear mapping based on the wave velocity value to generate a grayscale image of each monitoring section, set an anchor monitoring area based on each monitoring section, and acquire the anchor parameters in each monitoring area. The anchor parameters include the anchor diameter, the spacing between adjacent anchors, axial stress, and shear strain. The preprocessing module extracts features from the grayscale image corresponding to each monitoring section using a convolutional neural network model. Features include a binary mask image. The module obtains the fracture density by labeling connected components in the binary mask image, skeletonizes the fracture mask in the binary mask image, and generates the orientation angle of each fracture pixel using local gradient calculation, resulting in an orientation field vector map. Based on the orientation field vector map, the orientation angle of the main fracture is calculated using the vector averaging method. The surrounding rock strength analysis module constructs the surrounding rock strength coefficient of a single monitoring section based on the fracture density, the fracture orientation angle of the main monitoring section, and the anchor bolt parameters, using a fracture-anchor bolt coupling algorithm. The surrounding rock strength module is used to construct the fracture influence components of fracture density and main fracture direction angle for each monitoring section, set the influence range of anchor bolts on surrounding rock fractures, construct the effective anchor bolt support distance, construct the anchor bolt influence component for each monitoring section based on the anchor bolt diameter and the effective anchor bolt support distance, and use linear coupling to integrate the fracture influence component and the anchor bolt influence component to obtain the surrounding rock strength coefficient of a single monitoring section. The strength correction module is used to set the fatigue coefficient of the anchor material based on the shear strain in the anchor parameters, map the fatigue coefficient to the anchor attenuation factor through a nonlinear exponential function, and use the anchor attenuation factor as a correction term for the surrounding rock strength of a single monitoring section to obtain the corrected surrounding rock strength of the monitoring section. The threshold comparison module is used to obtain the overall surrounding rock strength value using a weighted average method, set a safe threshold for the surrounding rock strength, compare the overall surrounding rock strength value with the safe threshold, and issue an alarm when the value is lower than the safe threshold.

Images