Method for visualizing evolution process of loose circle of surrounding rock of deep underground engineering and related equipment

Abstract

The present application relates to the technical field of deep underground engineering, and more particularly to a deep underground engineering surrounding rock loose circle evolution process visualization method and related equipment. The technical scheme comprises: scanning the rock samples collected on site at different loading stages by a CT scanner to obtain CT scan images, arranging monitoring sections and monitoring points on site of the deep underground engineering, installing displacement, stress and strain sensors, and obtaining surrounding rock information using borehole camera technology; using a finite element software to establish a numerical model of the deep underground engineering, selecting a viscoelastic plastic constitutive model considering rock damage and rheological properties, assigning model parameters according to the CT scan images and rock mechanics test results, and correcting the numerical simulation data by parameter inversion to obtain corrected numerical simulation data. The present application obtains reliable surrounding rock information by comprehensively using various data acquisition means, and uses the viscoelastic plastic constitutive model considering rock damage and rheological properties to improve the accuracy of the numerical simulation results.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of deep underground engineering, and particularly relates to a deep underground engineering surrounding rock loose circle evolution process visualization method and related equipment. BACKGROUND

[0002] In deep underground engineering, the evolution of the surrounding rock loose circle is crucial to the stability and safety of the engineering. Currently, the research on the surrounding rock loose circle mainly has the following deficiencies:

[0003] The existing technology cannot comprehensively and accurately obtain the information of the surrounding rock at different loading stages, and cannot completely reflect the evolution process of the surrounding rock loose circle.

[0004] The traditional numerical simulation method is not accurate enough when considering the complex mechanical properties of rock, resulting in deviations between the simulation results and the actual situation.

[0005] The processing and analysis method of the CT scan image is not perfect, and cannot effectively extract the structural characteristics of the rock and accurately identify the loose circle.

[0006] There is a lack of intuitive and dynamic visualization display means, which is not conducive to the understanding and analysis of the evolution of the surrounding rock loose circle by engineering personnel, and further affects the engineering decision.

[0007] In view of the above deficiencies, the present application provides a deep underground engineering surrounding rock loose circle evolution process visualization method to solve the above problems in the prior art, and to realize accurate and intuitive display of the evolution process of the surrounding rock loose circle in deep underground engineering, and to provide reliable basis for engineering design and construction. SUMMARY

[0008] The purpose of the present application is to solve the problems in the background art, and to provide a deep underground engineering surrounding rock loose circle evolution process visualization method which can realize accurate and intuitive display of the evolution process of the surrounding rock loose circle, and provide reliable basis for engineering design and construction.

[0009] In a first aspect, the present application provides a deep underground engineering surrounding rock loose circle evolution process visualization method, comprising the following steps:

[0010] Data acquisition: scanning the rock samples collected on site at different loading stages by a CT scanner to obtain CT scan images, and arranging monitoring sections and monitoring points on site in deep underground engineering, installing displacement, stress, and strain sensors, and obtaining surrounding rock information by using borehole camera technology;

[0011] Numerical simulation: a numerical model of deep underground engineering is established by using finite element software, a viscoelastic plastic constitutive model considering rock damage and rheological properties is selected, model parameters are assigned according to CT scan images and rock mechanics test results, and corrected numerical simulation data are obtained through parameter inversion correction;

[0012] Image processing and analysis: pre-processing of CT scan images, extraction of structural features, and identification of loose annulus using specific algorithms;

[0013] Visualization: a three-dimensional model of surrounding rock is constructed using three-dimensional modeling software, the corrected numerical simulation data are mapped onto the three-dimensional model, and interactive functions are designed to view the evolution of the loose annulus of surrounding rock at different times and the data at specific locations.

[0014] Optionally, the parameter settings of the laboratory CT scan are as follows: voltage 160kV-220kV, current 10mA-20mA, slice thickness 0.1mm-0.5mm, and loading rate control 0.001mm / s-0.01mm / s.

[0015] Optionally, in the field monitoring, the displacement sensor accuracy reaches 0.01mm, the stress sensor accuracy reaches 0.1MPa, and the strain gauge accuracy reaches 10 -6 .

[0016] Optionally, the viscoelastic plastic constitutive model considering rock damage and rheological properties is a model coupled with the generalized Kelvin model and the Mohr-Coulomb model, and the stress-strain relationship is as follows:

[0017]

[0018] wherein σ is stress, ε is strain, E1 and E2 are elastic modulus, used to measure the ability of rock to resist elastic deformation, η is viscosity coefficient, reflecting the viscous properties of rock, c is cohesive force, which is the cementing force between rock particles, φ is internal friction angle, reflecting the friction characteristics between rock particles, t is time, and τ is integral variable; the values of each parameter are determined according to rock mechanics test, and the stress-strain state of rock under different stress conditions is calculated by substituting the formula, thereby providing an accurate viscoelastic plastic constitutive model for obtaining corrected numerical simulation data.

[0019] Optionally, in the image pre-processing, median filtering algorithm is used to remove salt and pepper noise in the image, and the filter window size is 3x3-5x5, and the median filtering algorithm formula is as follows:

[0020] g(x,y)=median{f(x-i,y-i),(x,y)∈W}

[0021] Wherein, f(x, y) is the pixel value of the original image at (x, y), g(x, y) is the pixel value of the filtered image at (x, y), W is the filter window, the formula represents that the pixel values in the window are sorted, and the middle value is taken as the filtered pixel value, thereby removing the noise; the histogram equalization algorithm is used to improve the image contrast.

[0022] Optionally, in the structural feature extraction, the Canny algorithm is used to identify the rock crack edges in the image, and the Gaussian filter formula in the Canny algorithm steps is:

[0023]

[0024] Wherein, σ is the standard deviation of the Gaussian kernel, used to control the smoothing degree of the Gaussian filter, (x, y) is the image coordinate, the formula filters the image by calculating the two-dimensional Gaussian function value, smooths the image and reduces the noise influence; the amplitude and direction of the image gradient are calculated; the non-maximum suppression is performed on the gradient amplitude; the edges are detected and connected through the double-threshold algorithm; the cracks are refined and filled through the morphological dilation and corrosion operations, and the geometric parameters of the crack length, width and azimuth angle are extracted; the threshold segmentation algorithm is used to segment the pores in the rock, and the area and perimeter parameters of the pores are calculated.

[0025] Optionally, the threshold-based region growing algorithm is used to identify the loose circle, an initial threshold T0 is determined according to the physical properties and mechanical parameters of the rock and in combination with the gray value distribution of the CT image; one of the pixel points (x0, y0) in the image is taken as a seed point, if the gray value I(x0, y0) is less than T0 and adjacent to the pixel points marked as the loose region, the pixel point is marked as the loose region, and then the loose region is continuously expanded, and the expansion condition is that for the pixel point (x, y) in the currently marked loose region, the adjacent pixel points (x ′ ,y ′ ) satisfy |I(x, y)-I(x ′ ,y ′ )|<T1 and I(x ′ ,y ′ )<T0, wherein T1 is a set gray difference threshold, used to control the difference degree of the adjacent pixel gray values; I(x, y) and I(x ′ ,y ′ ) are the gray values of the pixel points (x, y) and (x i ,y i ) respectively, and the formula represents that when the difference between the gray values of the adjacent pixel points is less than the gray difference threshold T1 and the gray value of the adjacent pixel point is less than the initial threshold T0, the adjacent pixel point is marked as the loose region, until there is no pixel point satisfying the condition.

[0026] Optionally, a three-dimensional model of the deep underground engineering surrounding rock is constructed by using three-dimensional modeling software Maya, 3ds Max, according to CT scan images and the corrected numerical simulation data results, the internal structure information of the rock in the CT scan images is mapped to the surface of the three-dimensional model, and the model surface is made to present the real texture of the rock through the texture mapping technology.

[0027] Optionally, in the data fusion and mapping, the stress, strain and loose circle data obtained by numerical simulation are mapped to the corresponding positions of the three-dimensional model through an interpolation algorithm, and an inverse distance weighted interpolation algorithm is adopted, and the formula is as follows:

[0028]

[0029] Wherein, Z(x0,y0) is an interpolation result at a point (x0,y0) to be interpolated, Z i is the value of a known data point i, d i is the distance from the point to be interpolated to the known data point i, and p is a weight coefficient, and the value is 2, the formula determines the value of the point to be interpolated by calculating the weighted average of the distance between the point to be interpolated and the known data point, and realizes the mapping of data from numerical simulation to the three-dimensional model; different colors and transparencies are used to represent different physical quantities and states, red color represents a high stress area, blue color represents a low stress area, yellow color represents a loose area, and the transparent part represents a pore and a cavity; in the interactive function design, a time axis and an interactive button are arranged in the visualization interface, a user can view the evolution of the surrounding rock loose circle at different times by dragging the time axis, and can also view detailed data of stress, strain and displacement at a specific position by clicking the interactive button, and the three-dimensional model is also supported to be rotated, scaled and translated by the user.

[0030] On the other hand, the application provides a computer device, comprising a processor, a memory, a communication interface, and one or more programs, the one or more programs are stored in the memory and configured to be executed by the processor, and the program comprises instructions for executing steps in the method of the first aspect.

[0031] Compared with the prior art, the application has at least one of the following beneficial technical effects:

[0032] The application comprehensively uses various data acquisition means, can comprehensively and accurately obtain surrounding rock information, and provides reliable data for subsequent analysis.

[0033] Further using image processing and analysis method can effectively extract rock structure characteristics and identify loose circle, and provide accurate data support for visual display. The visual display method enables engineering personnel to intuitively and dynamically view the evolution of the surrounding rock loose circle, facilitates understanding and analysis, and significantly improves the accuracy of engineering decision. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 A flowchart of the deep underground engineering surrounding rock loose circle evolution process visualization method. DETAILED DESCRIPTION

[0035] The technical solutions of the present application will be further described below in combination with the drawings and specific embodiments.

[0036] As shown in Figure 1 , the deep underground engineering surrounding rock loose circle evolution process visualization method proposed by the present application includes data acquisition: scanning the rock samples collected on site at different loading stages by a CT scanner to obtain CT scan images, and arranging monitoring sections and monitoring points on the deep underground engineering site, installing displacement, stress, and strain sensors, and obtaining surrounding rock information by using borehole camera technology.

[0037] In this embodiment, when the CT scanner scans the rock sample, the CT scanner is pre-set, the voltage is set to 160kV-220kV, and the current is set to 10mA-20mA. When slicing the rock sample, the slice thickness is controlled within the range of 0.1mm-0.5mm, and the loading rate of the CT scanner is controlled so that the loading rate is within the range of 0.001mm / s-0.01mm / s. Finally, the rock sample is scanned at different loading stages. During the on-site monitoring process, first, the monitoring sections and monitoring points are arranged on the deep underground engineering site, then the displacement sensor (with an accuracy of 0.01mm), the stress sensor (with an accuracy of 0.1MPa), and the strain gauge (with an accuracy of 10 -6 ) are installed; finally, a suitable position on the monitoring section is selected for drilling, the diameter and depth of the drilling should meet the requirements of the borehole camera equipment, the drilling diameter is 50-100mm, and the depth is determined according to the needs. During the drilling process, attention should be paid to maintaining the perpendicularity and stability of the drilling to avoid bending or collapse of the drilling. After the drilling operation is completed, the image information of the surrounding rock can be obtained by using the borehole camera technology.

[0038] As shown in Figure 1 , the collected data is numerically simulated, a finite element software is used to establish a numerical model of the deep underground engineering, a viscoelastic plastic constitutive model considering rock damage and rheological properties is selected, the model parameters are assigned according to the CT scan images and rock mechanics test results, and the corrected numerical simulation data is obtained through parameter inversion correction.

[0039] In this embodiment, when selecting a viscoelastic plastic constitutive model considering rock damage and rheological properties, a model coupling the generalized Kelvin model and the Mohr-Coulomb model is adopted, and the stress-strain relationship is as follows:

[0040]

[0041] wherein σ is stress, ε is strain, E1 and E2 are elastic modulus and are used to measure the ability of rock to resist elastic deformation, η is viscosity coefficient and can reflect the viscous properties of rock, c is cohesion and is the cementing force between rock particles, φ is internal friction angle and can reflect the friction characteristics between rock particles, t is time, and τ is integral variable; the parameter values are determined according to rock mechanics tests, and the stress-strain state of rock under different stress conditions is calculated by substituting the formula, so as to provide accurate viscoelastic plastic constitutive model data for the modified numerical simulation data;

[0042] Elastic modulus determination stage: a uniaxial compression test of rock is designed and performed, a plurality of regular rock test pieces are prepared, and the rock test pieces are placed on a pressure testing machine, an axial pressure is applied at a constant loading rate, and a strain gauge device is used to accurately measure the axial strain of the test piece during the loading process. According to Hooke's law, the elastic modulus E1 and E2 are calculated by the slope of the initial linear segment of the stress-strain curve. For rock test pieces of different lithology and structure, multiple repeated tests need to be performed to obtain representative elastic modulus values;

[0043] Viscosity coefficient measurement stage: a creep test of rock is performed. The rock test piece is placed under a constant axial stress, and a high-precision displacement measuring device is used to monitor the deformation of the test piece over time for a long time. Through analysis of the creep curve and combination of the constitutive relationship of viscous materials, the viscosity coefficient η is back calculated. During the test, attention should be paid to keeping the environmental conditions stable and reducing the interference of external factors on the test results, so as to ensure the accuracy of the measurement data;

[0044] When the cohesion and internal friction angle are determined, a direct shear test of rock is performed. A plurality of rock test pieces are prepared, a shear force is applied to the test pieces under different normal stresses, and the shear stress when the test piece is sheared is recorded. According to the Mohr-Coulomb strength theory, the Mohr circle is drawn and the envelope line is determined, so as to calculate the cohesion c and internal friction angle φ of the rock. The normal stress values of each test should have a certain gradient to ensure the accuracy of the results.

[0045] In this embodiment, the CT scan image is preprocessed and structural features are extracted, and a specific algorithm is used to identify the loose circle.

[0046] In the image preprocessing stage of the embodiment, a median filtering algorithm is used to remove salt and pepper noise in the image, the filtering window size is 3x3-5x5, and the median filtering algorithm formula is: g(x, y) = median {f(x-i, y-i), (x, y) e W}, where f(x, y) is the pixel value of the original image at (x, y), g(x, y) is the pixel value of the filtered image at (x, y), W is the filtering window, and the formula indicates that the pixel values in the window are sorted, and the middle value is taken as the filtered pixel value, so as to remove the noise; a histogram equalization algorithm is used to improve the image contrast;

[0047] The above median filtering algorithm includes the following steps:

[0048] First, determine the filtering window size: according to the noise condition and detail requirement of the image, select a suitable filtering window size. If the image noise is dense, select a 3x3 window, which can not only reduce the calculation amount but also effectively remove the noise; if the image noise is dispersed, select a 5x5 window to better preserve the edge and texture information of the image.

[0049] Second, traverse the image pixels: start from the top-left corner pixel of the image, take the selected filtering window size as the step, and process each pixel in the image row by row and column by column. For the pixels at the image boundary, since the window may exceed the image range, a boundary expansion method is used to copy the boundary pixel value to fill the exceeding part.

[0050] Third, extract the pixel values in the window: take the current processing pixel as the center, and extract the gray values of all pixels in the window according to the selected window size. For a 3x3 window, the gray values of the current pixel and its surrounding 8 adjacent pixels are included; for a 5x5 window, the gray values of 25 pixels are included.

[0051] Fourth, sort the pixel values: sort the extracted gray values of all pixels in the window from small to large (or from large to small). Common sorting algorithms are used, including quicksort and bubble sort, and the purpose of sorting is to find the pixel value at the middle position.

[0052] Fifth, determine the filtered pixel value: according to the sorting result, take the pixel value at the middle position as the filtered gray value of the current processing pixel. If the number of pixels in the window is odd, directly take the middle value; if it is even, take the average of the middle two values as the filtered pixel value. For a 3x3 window, the number of pixels is 9, and the 5th sorted pixel value is taken; for a 4x4 window, the number of pixels is 16, and the average of the 8th and 9th sorted pixel values is taken.

[0053] Step 6, update the image: assign the calculated filtered pixel value to the corresponding pixel position in the original image, complete the filtering of the pixel. Continue to traverse the next pixel of the image, repeat steps 3 to 6, until all pixels in the image are processed, get the image after removing the salt and pepper noise.

[0054] The specific use method of the histogram equalization algorithm described above is as follows:

[0055] Step 1, calculate the image histogram: count the number of pixels of each gray level in the image. The gray level of the image is usually between 0-255 (8-bit gray image), create an array with a length of 256 to record the number of pixels of each gray level. Traverse all pixels of the image, and when a pixel is encountered, add 1 to the count of its corresponding gray level in the array.

[0056] Step 2, calculate the cumulative distribution function (CDF): calculate the cumulative distribution function of each gray level, that is, the proportion of the total number of pixels from gray level 0 to the current gray level in the total number of pixels in the image. Let the total number of pixels in the image be N, and the number of pixels of gray level i be ni, then the cumulative distribution function of gray level i is:

[0057]

[0058] Step 3, map the gray value: according to the cumulative distribution function, map each gray level in the original image to a new gray level to achieve histogram equalization. The new gray level g(i) = round{(L-1)×CDF(i)}, where L is the total number of gray levels of the image (for 8-bit gray image, L = 256), and the round() function is the rounding function. Through this formula, the gray value of the original image is redistributed according to the cumulative distribution function, so that the gray distribution of the image is more uniform, thereby improving the contrast of the image.

[0059] Step 4, generate the equalized image: traverse all pixels of the original image, replace the gray value of each pixel with the new gray value calculated in step 3 according to the mapping relationship, thereby generating the histogram equalized image.

[0060] In the structural feature extraction stage, Canny algorithm is used to identify the rock crack edge in the image. The Gaussian filter formula in the Canny algorithm steps is:

[0061]

[0062] The σ in the above formula is the standard deviation of the Gaussian kernel function, used to control the smoothing degree of the Gaussian filter, and (x, y) is the image coordinate. The formula filters the image by calculating the value of the two-dimensional Gaussian function, smoothes the image and reduces the noise effect; calculates the amplitude and direction of the image gradient; performs non-maximum suppression on the gradient amplitude; detects and connects the edges through a double-threshold algorithm; refines and fills the cracks through morphological dilation and erosion operations, extracts the geometric parameters of the length, width and azimuth angle of the cracks; and uses a threshold segmentation algorithm to segment the pores in the rock and calculate the area and perimeter parameters of the pores.

[0063] In identifying the loose circle, a threshold-based region growing algorithm is used, an initial threshold T0 is determined according to the physical properties and mechanical parameters of the rock and the gray value distribution of the CT image; starting from the edge of the image, one of the pixel points (x0, y0) is taken as a seed point, if the gray value I(x0, y0) is less than T0 and adjacent to the pixel point marked as the loose region, the pixel point is marked as the loose region, then the loose region is continuously expanded, and the expansion condition is that for the pixel point (x, y) in the currently marked loose region, the adjacent pixel points (x ′ ,y ′ ) satisfy |I(x, y)-I(x ′ ,y ′ )|<T1 and I(x ′ ,y ′ )<T0, wherein T1 is a set gray difference threshold, used to control the difference degree of the adjacent pixel gray values; I(x, y) and I(x ′ ,y ′ ) are respectively the pixel points (x, y) and (x ′ ,y ′) is less than the initial threshold value T0, the adjacent pixel point is marked as a loose area until there is no pixel point meeting the condition; the above algorithm determines the initial threshold value according to the physical properties and mechanical parameters of the rock and the CT image gray value distribution, is designed closely around the characteristics of the rock loose circle, can accurately locate the area related to rock loosening, and improves the identification accuracy; starting from the pixel points at the image edge as seed points, the potential loose area is gradually excavated by fully utilizing the image edge features and through reasonable growth expansion conditions, the rock structure information contained in the CT image gray value is effectively integrated, the expansion of the loose area is stable and consistent with the actual situation of rock loosening by setting strict expansion conditions, that is, the gray value difference between adjacent pixels is controlled by the gray difference threshold value, only when the gray value difference between adjacent pixels is less than the gray difference threshold value and the gray value is less than the initial threshold value, the pixel point is marked as a loose area, the reliability of the identification result is improved; and once the initial threshold value and the seed point are determined, the algorithm can automatically expand the loose area according to the set conditions until there is no pixel point meeting the condition, the artificial intervention is reduced, the identification efficiency is improved, and the algorithm is suitable for processing of large-scale CT image data; the gray difference threshold value parameter can be adjusted according to different rock characteristics and actual needs, the adaptability of the algorithm to different geological conditions and rock types is enhanced, and the loose circle can be accurately identified in various complex situations.

[0064] As shown in Figure 1 , a three-dimensional modeling software is further used to construct a surrounding rock three-dimensional model, the corrected numerical simulation data is mapped to the three-dimensional model, and an interactive function is designed to realize viewing of surrounding rock loose circle evolution at different times and data at specific positions, so as to achieve the purpose of visual display.

[0065] In this embodiment, three-dimensional model construction is first performed, a three-dimensional modeling software Maya and 3ds Max are used, the three-dimensional model of the surrounding rock of the deep underground engineering is constructed according to the CT scan image and the corrected numerical simulation data result, the internal structure information of the rock in the CT scan image is mapped to the surface of the three-dimensional model, and the model surface is made to present the real texture of the rock through the texture mapping technology.

[0066] In data fusion and mapping, the stress, strain and loose circle data obtained by numerical simulation are mapped to the corresponding positions of the three-dimensional model through an interpolation algorithm, and an inverse distance weighted interpolation algorithm is adopted, and the formula is as follows:

[0067]

[0068] Z (x0, y0) in the above formula is the interpolation result at the interpolation point (x0, y0), Z iis the value of the known data point i, d i is the distance between the to-be-interpolated point and the known data point i, p is a weight coefficient with a value of 2, and the formula determines the value of the to-be-interpolated point by calculating a weighted average of the distances between the to-be-interpolated point and the known data points, thereby realizing the mapping of data from numerical simulation to a three-dimensional model; different colors and transparencies are used to represent different physical quantities and states, for example, red is used to represent a high stress area, blue is used to represent a low stress area, yellow is used to represent a loose area, and transparent parts represent pores and cavities; texture mapping technology is used to make the model surface exhibit the real texture of the rock; the stress, strain, and loose circle data obtained through numerical simulation are mapped to the corresponding positions on the surrounding rock three-dimensional model through the inverse distance weighted interpolation algorithm, and different colors and transparencies are used to represent different physical quantities and states, with clear color differentiation, which facilitates the observation and recording of staff.

[0069] In the interactive function design, a time axis and interactive buttons are set in the visualization interface. Users can view the evolution of the surrounding rock loose circle at different times by dragging the time axis. Users can intuitively see the evolution of the surrounding rock loose circle over time, which helps users understand the development trend of the surrounding rock loose circle when the underground engineering is in different construction stages or is affected by different external factors, thereby predicting potential risks in advance and providing a strong basis for engineering decision-making. During tunnel construction, it can be clearly observed how the range of the loose circle expands as the excavation depth increases, so as to timely adjust the support scheme; by clicking the interactive buttons, detailed data of stress, strain, and displacement at specific positions can be viewed. The design of the interactive buttons enables users to quickly obtain detailed data of stress, strain, and displacement at specific positions, which is crucial for engineering and technical personnel to analyze the stress and deformation conditions at key positions. When analyzing the stability of underground caverns, the positions prone to problems at the cavern vault and corners can be accurately located, the stress and strain states can be viewed, and it can be judged whether the safety threshold is exceeded, thereby providing data support for engineering reinforcement or adjustment of construction technology; at the same time, the three-dimensional model can be rotated, scaled, and translated by the user, greatly improving the user's observation experience of the model. The user can observe the overall situation of the surrounding rock and the distribution characteristics of the loose circle from different angles and at different scales. This helps to more deeply analyze the spatial structure of the surrounding rock and the relationship between the loose circle and the surrounding rock mass, discover possible overlooked details, and more accurately assess the stability of underground engineering; this interactive function design integrates multiple information acquisition methods, enabling users to quickly understand the dynamic changes of the surrounding rock loose circle and detailed data at key positions, and also enabling users to observe the surrounding rock three-dimensional model from all angles. These rich and intuitive information can help engineers and decision-makers make scientific and reasonable decisions quickly, avoid decision-making errors caused by incomplete or untimely information, and improve the safety and efficiency of engineering.

[0070] The above specific embodiments are only several optional embodiments of the present application, and based on the technical scheme of the present application and the related inspiration of the above embodiments, the person skilled in the art can make various alternative improvements and combinations on the above specific embodiments.

Claims

1. A method for visualizing the evolution process of the loose circle of surrounding rock of deep underground engineering, characterized in that, The method comprises the following steps: Data acquisition: scanning the rock samples collected on site by a CT scanner at different loading stages to obtain CT scan images, and arranging monitoring sections and monitoring points on the deep underground engineering site, installing displacement, stress, and strain sensors, and obtaining surrounding rock information using borehole camera technology; Numerical simulation: establishing a numerical model of the deep underground engineering using finite element software, selecting a viscoelastic plastic constitutive model considering rock damage and rheological properties, assigning model parameters according to CT scan images and rock mechanics test results, and obtaining corrected numerical simulation data through parameter inversion correction; wherein the viscoelastic plastic constitutive model considering rock damage and rheological properties is a model coupling the generalized Kelvin model and the Mohr-Coulomb model, and the stress-strain relationship is: ； wherein, is the stress, is the strain, , is the elastic modulus, which measures the ability of the rock to resist elastic deformation, is the viscosity coefficient, which reflects the viscous properties of the rock, is the cohesion, which is the cementing force between the rock particles, is the internal friction angle, which embodies the friction characteristics between the rock particles, is the time, is the integral variable; Image processing and analysis: pre-processing of CT scan images, structural feature extraction, and identification of loose annulus using specific algorithms; the identification of loose annulus uses a threshold-based region growing algorithm, according to the physical properties and mechanical parameters of the rock, combined with the gray value distribution of the CT image, to determine an initial threshold ; starting from the edge of the image, taking one of the pixel points as the seed point, if its gray value is less than , and adjacent to the pixel points marked as loose area, then mark this pixel point as loose area, and then continuously expand the loose area, the expansion condition is that for the pixel points of the current marked loose area, the adjacent pixel points satisfy and , wherein is a set gray difference threshold value, used to control the difference degree of the gray values of adjacent pixels; are the gray values of pixel points , , the formula indicates that when the difference between the gray values of adjacent pixel points is less than the gray difference threshold value , and the gray value of the adjacent pixel point is less than the initial threshold value , the adjacent pixel point is marked as a loose area, until there is no pixel point that meets the condition; Visualization display: constructing a three-dimensional model of the surrounding rock using three-dimensional modeling software, mapping the corrected numerical simulation data onto the three-dimensional model, and designing interactive functions to realize the viewing of surrounding rock loose circle evolution at different times and specific location data.

2. The method according to claim 1, wherein, The parameter settings for the laboratory CT scanning are: voltage 160kV - 220kV, current 10mA - 20mA, slice thickness 0.1mm - 0.5mm, and loading rate control 0.001mm / s - 0.01mm / s.

3. The method according to claim 1, wherein, In the field monitoring, the displacement sensor accuracy reaches 0.01mm, the stress sensor accuracy reaches 0.1MPa, and the strain gauge accuracy reaches .

4. The method according to claim 1, wherein, In the image preprocessing, the median filter algorithm is used to remove salt and pepper noise in the image, the filter window size is 3x3 - 5x5, and the median filter algorithm formula is: ； wherein, is a pixel value of the original image at is a pixel value of the filtered image at is a pixel value of the original image at is a pixel value of the filtered image at W is a filter window.

5. The method of visualizing the evolution of the disturbed zone of the surrounding rock mass of a deep underground engineering according to claim 1, characterized in that, In the structure feature extraction, the Canny algorithm is used to identify the rock fracture edges in the image, and the Gaussian filter formula in the Canny algorithm steps is: ； wherein, is a standard deviation of a Gaussian kernel for controlling a smoothing degree of the Gaussian filter, is an image coordinate.

6. The method of visualizing the evolution of the disturbed zone of the surrounding rock mass of a deep underground engineering according to claim 1, characterized in that, In the three-dimensional modeling software constructing a three-dimensional model of the surrounding rock, three-dimensional modeling software Maya and 3ds Max are used to construct a three-dimensional model of the surrounding rock of the deep underground engineering according to the CT scan images and the corrected numerical simulation data results, and the internal structure information of the rock in the CT scan images is mapped onto the surface of the three-dimensional model.

7. The method of visualizing the evolution of the disturbed zone of the surrounding rock mass of a deep underground engineering according to claim 1, characterized in that, In the data fusion and mapping, the stress, strain, and loose circle data obtained by numerical simulation are mapped to the corresponding positions of the three-dimensional model through an interpolation algorithm, and the inverse distance weighted interpolation algorithm is used, and the formula is: ； in, The point to be interpolated The interpolation result at that point, Known data points The value, It is the distance from the point to be interpolated to the known data point. distance, It is a weighting coefficient with a value of 2. The formula determines the value of the interpolation point by calculating the weighted average of the distances between the interpolation point and the known data points, thus realizing the mapping of data from numerical simulation to three-dimensional model.

8. A computer device, comprising: The user equipment includes a processor, a memory, a communication interface, and one or more programs stored in the memory and configured to be executed by the processor, and the programs include instructions for performing the steps in the method of any one of claims 1-7.

Images