Rock shearing criticality discrimination method based on joint surface three-dimensional curvature automatic identification

Abstract

The invention discloses a rock shearing criticality discrimination method based on joint surface three-dimensional curvature automatic identification, and relates to the technical field of deep rock mechanics and fracture mechanics. The method mainly solves the technical problems of the existing rock shearing critical discrimination method in the aspects of accuracy, quantificaiton and applicability. The method comprises the following steps: firstly, sampling and manufacturing a rock sample, and carrying out a shear test on the rock sample to obtain mechanical parameters and joint surface data of the sample; carrying out quantitative analysis on two-dimensional damage characteristics and three-dimensional curvature distribution of the rock joint surface; and finally, determining a shear critical condition for inducing rock shear fracture by combining an acoustic emission monitoring technology, and realizing accurate judgment and early warning of the rock shear critical condition. The method is particularly suitable for rock mass stability evaluation in high ground stress environments such as deep mines, tunnels and underground energy storage spaces, and has high practicability and popularization value.

Description

Technical Field

[0001] This invention relates to the fields of deep rock mechanics and fracture mechanics, specifically to a method for determining rock shear criticality based on automatic identification of three-dimensional curvature of joint surfaces. Background Technology

[0002] In the construction of modern deep underground engineering projects, with the continuous increase in development depth, the complex conditions of high ground stress, high ground temperature, and high-pressure water in the rock mass are increasingly becoming important factors restricting engineering safety. Deep rock masses are in a triaxial high-stress state. Excavation and unloading will cause stress redistribution, resulting in a significant increase in maximum shear stress. Once this stress exceeds the shear strength of the rock mass, it is highly likely to induce shear slip along primary fractures or potential slip surfaces. This is the fundamental reason for the frequent occurrence of shear failure under deep high-stress conditions. Such shear criticalities not only induce disasters such as rock slope slippage, fault or structural plane slip-type rockbursts, and fault activation slippage, but also have the characteristics of strong suddenness, concentrated energy release, and wide impact range, seriously threatening the safety of personnel and facilities. Compared with tensile (Type I) fracture, shear (Type II) fracture of rock is easier to initiate, propagates over longer distances, and is instantaneous. In deep, high-stress rock masses with well-developed structural planes, it often becomes the dominant failure mechanism. For example, in the Mponeng gold mine in South Africa, over 60% of rockbursts at depths exceeding 3500 meters were related to shear slip at structural surfaces, while the proportion of tensile failure was significantly lower. Therefore, researching the criteria for rock shear criticality is crucial for accurately predicting the location and timing of disasters, optimizing support design, and enhancing the disaster resistance of deep engineering projects. Current research on rock shear criticality largely focuses on simulating mechanical behavior at the rock scale. However, systematic simulations and analyses of the formation and propagation of rock shear cracks under multi-field coupling environments such as high ground stress, high ground temperature, and high-pressure water in deep geological conditions are still lacking. While existing shear failure discrimination methods can directly identify rock shear fracture behavior, they still have significant shortcomings in considering the influence of complex deep environments on rock shear fracture, making it difficult to accurately reflect the shear critical mechanism under complex geological conditions.

[0003] The relevant research in the existing technology mainly includes: CN116971733B discloses an integrated exploration and injection machine and process suitable for tunneling under complex geological conditions. It achieves drill bit disengagement by blocking water flow and shearing the shear pin using a pressure-blocking ball, and reinforces the tunnel with grouting screens and sealing devices. This solves the problem of tunneling equipment shutdown caused by critical abnormal damage to deep surrounding rock, improving construction efficiency and surrounding rock stability. However, this machine and process do not address adaptability to extreme environments such as high temperatures and high ground pressures.

[0004] CN119811014A discloses an early warning method, system, and terminal based on a fault critical assessment model. It uses multi-sensor data fusion and machine learning models to predict fault fracture propagation parameters, thereby achieving critical early warning and graded response. However, this method is aimed at fault critical early warning and does not involve the microscopic mechanism of shear cracks inside rocks. Furthermore, the model has insufficient applicability under force-thermal-fluid multi-field coupling conditions and cannot accurately predict the shear critical point.

[0005] CN116519477A discloses a method, device, and storage medium for determining rock criticality. This method calculates fractal information by combining stress-strain curves and acoustic emission characteristics, and uses joint mapping to accurately identify critical precursors in rocks, improving the accuracy of rock fracture monitoring and rockburst prevention. However, this method is mainly designed for testing rock samples in a laboratory environment and does not consider multi-field coupling conditions such as high temperature and high hydraulic pressure, making it unsuitable for direct application to criticality prediction in deep rock masses under complex force-thermal-fluid environments.

[0006] It is evident that the aforementioned existing technologies have technical problems in terms of accuracy, quantification, and applicability in rock shear discrimination, and these existing technologies need further improvement. Summary of the Invention

[0007] The purpose of this invention is to provide a rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces. By establishing a quantitative discrimination criterion for rock shear criticality, it achieves accurate discrimination of shear fracture critical conditions and solves the technical problems existing in the prior art.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: A method for determining rock shear criticality based on automatic identification of three-dimensional curvature of joint surfaces includes the following steps: a. Take samples and prepare rock specimens, and conduct direct shear tests on the rock specimens under different normal stress conditions to obtain the mechanical parameters and joint surface data of the rock specimens before and after shearing. b. The joint surface of the rock sample after shearing is image processed by the binarization method based on the Otsu algorithm. The microscopic features of the shear and tensile regions of the joint surface are observed by scanning electron microscopy to determine the shear damage area and tensile damage area of ​​the joint surface. The proportion of shear damage area of ​​the joint surface under different normal stress conditions is quantitatively calculated. c. Establish a method for quantifying the geometric features of joint surfaces based on automatic identification of three-dimensional curvature: three-dimensional scanning technology is used to obtain three-dimensional point cloud data of joint surfaces; based on Otsu threshold segmentation and iterative optimization strategy, the three-dimensional curvature intervals of joint surfaces are automatically identified, and the distribution ratio of different curvature intervals is calculated as a three-dimensional feature index for shear criticality judgment. d. Analyze the changing trends of the proportion of shear damage area and high curvature of joint surfaces under different normal stress conditions, establish the relationship between the ratio of effective normal stress and shear stress and the shear criticality, and further determine the shear critical condition; e. Collect on-site data through the acoustic emission monitoring system and compare it with the test data obtained in step d to verify the accuracy of the test data.

[0009] In the above-mentioned rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces, in step a, rock cross-section samples are taken at the engineering site, and point cloud data is obtained using a handheld three-dimensional scanner; the rock sample preparation method is as follows: through precise core taking, cutting and grinding processes, cylindrical samples with a diameter of 50mm and a height of 100mm and cubic samples of 50mm×50mm×50mm are prepared, and the parallelism of the cross-section of all rock samples is controlled within ±0.02mm.

[0010] In the above-mentioned rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surface, in step a, the different normal stress conditions are 20%, 40%, 60% and 80% of the average compressive strength of the rock sample.

[0011] In the above-mentioned rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surface, in step b, the binarization method uses Gaussian filtering with kernel size (5,5) for preprocessing, and adaptive histogram equalization is used to enhance image contrast. The parameters are set as clipLimit=0.02 and tileGridSize=(8,8). The Otsu algorithm is applied for binarization processing to divide the joint surface of the rock sample into black area and white area. The black area is the tensile damage area of ​​the joint surface, and the white area is the shear damage area of ​​the joint surface.

[0012] In the above-mentioned rock shear criticality discrimination method based on automatic recognition of three-dimensional curvature of joint surface, step b is defined as the proportion of shear damage area as the ratio of white pixels on joint surface to the total number of pixels.

[0013] The above-mentioned rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces, in step c, the specific steps based on Otsu threshold segmentation and iterative optimization strategy include: (1) Curvature data normalization: The curvature distribution of points is transformed into a probability distribution; (1); In formula (1): For having curvature value The number of points; This represents the total points. For having curvature value The proportion or frequency of a point in the total number of points; (2) Calculate the category probability: Calculate the probability at a given threshold. and In the case of points distributed within different curvature intervals; (2-1); (2-2); (2-3); In formula (2): ， The curvature threshold is used to classify points into three categories; For curvature values ​​less than or equal to The probability of a point; For curvature values ​​greater than and less than or equal to The probability of a point; For curvature values ​​greater than The probability of a point; (3) Calculate the average curvature of each category: Calculate the average curvature value of points within each category to reflect the curvature characteristics of each category; (3-1); (3-2); (3-3); In equation (3-1): For curvature values ​​greater than 0 and less than or equal to 0 The average curvature of the point; In equation (3-2): For curvature values ​​greater than and less than or equal to The average curvature of the point; In equation (3-3): For curvature values ​​greater than The average curvature of the point; (4) Calculate the overall average curvature: Calculate the average curvature of the entire point set as a global reference value; (4); In equation (4): The average curvature value across all points; This represents the total points. (5) Calculate the inter-class variance: given a threshold and Below, the curvature difference between different categories is used to judge the segmentation effect; (5); In formula (5): Represents inter-class variance, which measures the curvature difference between different classes; (6) Finding the optimal threshold pair: representing the inter-class variance by the difference between the class probability and the class mean curvature; (6); (7) Iterative optimization of threshold: By traversing all possible combinations of thresholds Find the optimal threshold pair that maximizes the inter-class variance. This optimal threshold is used to distinguish different curvature categories, achieving optimal segmentation of the point cloud. (7); In equation (7): The optimal curvature threshold range maximizes the inter-class variance.

[0014] In the above-mentioned rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces, in step d, when the rock sample is limestone, the determined rock shear criticality condition range is [1.4, 1.6]. When this ratio exceeds 1.6, the rock will experience shear criticality; when the rock sample is sandstone, the determined rock shear criticality condition range is [1.3, 1.5]. When this ratio exceeds 1.5, the rock will experience shear criticality; when the rock sample is granite, the determined rock shear criticality condition range is [1.5, 1.7]. When this ratio exceeds 1.7, the rock will experience shear instability.

[0015] In the above-mentioned rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surface, step e includes an acoustic emission monitoring system comprising acoustic emission sensors. Several acoustic emission sensors are arranged on the surface of the rock sample. The frequency range of each acoustic emission sensor is 100 to 1000 kHz, the sampling rate is set to 5 MHz, the preset gain is 40 dB, and the threshold is set to 45 dB.

[0016] In the above-mentioned rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces, if the field data and experimental data are inconsistent, steps a to d are repeated sequentially in step e.

[0017] Compared with the prior art, the present invention brings the following beneficial technical effects: (1) This invention proposes a rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces, which realizes the binary quantitative characterization of shear damage features of rock joint surfaces. Through high-precision image acquisition and adaptive binarization algorithm, combined with scanning electron microscopy observation, the tensile damage area and shear damage area on the joint surface are accurately distinguished, and the proportion of shear damage area of ​​joint surface under different stress conditions is quantified, providing an intuitive and reliable two-dimensional feature index for shear criticality discrimination. The preprocessing technology based on Gaussian filtering and adaptive histogram equalization, combined with the OTSU optimal threshold algorithm, ensures the accuracy and stability of image segmentation.

[0018] (2) This invention establishes a method for quantifying the geometric features of joint surfaces based on automatic three-dimensional curvature identification. It innovatively proposes an automatic joint surface curvature identification algorithm that integrates OTSU threshold segmentation and iterative optimization, achieving accurate characterization of the three-dimensional morphological features of joint surfaces. By calculating and analyzing the distribution ratio of different curvature intervals (low curvature, medium curvature, and high curvature), the evolution law of joint surface geometric features during rock shearing is revealed, providing a novel three-dimensional feature index for shear criticality discrimination. This algorithm is computationally efficient and highly adaptable, capable of handling complex and varied joint surface morphologies.

[0019] (3) This invention determines the critical threshold range for rock shear fracture and establishes a quantitative shear criticality discrimination criterion. By systematically analyzing the variation law of the proportion of shear damage area and high curvature of joint surfaces under different stress conditions, and combining acoustic emission monitoring technology, the effective normal stress to shear stress ratio range [1.4, 1.6] is determined as the critical threshold range for inducing rock shear fracture. This discrimination criterion has clear physical meaning and high applicability, providing a reliable theoretical basis for shear criticality risk assessment in deep underground engineering. The system utilizes multiple monitoring technologies for cross-validation, significantly improving the reliability and accuracy of the discrimination results.

[0020] (4) This invention proposes a multi-parameter fusion-based rock shear critical early warning mechanism. Through comprehensive analysis of two-dimensional damage characteristics of joint surfaces, three-dimensional curvature distribution, and acoustic emission monitoring data, a multi-parameter fusion-based rock shear critical early warning model was established. This model can monitor changes in rock state in real time and issue timely warnings when key parameters approach critical values, providing technical support for the safe construction and operation of deep underground engineering projects. The early warning system adopts a graded early warning mechanism, issuing different levels of warning signals according to the degree of risk, facilitating engineers to take corresponding protective measures.

[0021] (5) It has broad applicability and practical value. This method is not only applicable to rock mechanics research under laboratory conditions, but can also be applied to in-situ monitoring in actual engineering projects with appropriate improvements, especially for rock mass stability assessment in high geostress environments such as deep mines, tunnels, and underground energy storage spaces. The implementation process of the method of this invention is simple and feasible, and the required equipment is widely available, which has strong practicality and promotion value. Attached Figure Description

[0022] The present invention will be further described below with reference to the accompanying drawings: Figure 1 This is a flowchart of a rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces according to the present invention.

[0023] Figure 2 The images are binarized images of limestone joint surfaces under different normal stress conditions.

[0024] Figure 3 This is a SEM scan image of the shear damage area on the joint surface of limestone.

[0025] Figure 4 Flowchart for Otsu threshold segmentation of joint surface curvature intervals.

[0026] Figure 5 The percentage of low curvature surfaces on the upper and lower joint surfaces under different normal stress conditions.

[0027] Figure 6 This is a graph showing the relationship between the ratio of effective normal stress to shear stress, the proportion of shear damage area, and the proportion of high curvature.

[0028] Figure 7 The cumulative energy variation curves of acoustic emission under different effective normal stress to shear stress ratios are shown. Figure 8 The curves show the changes in damage variables under different ratios of effective normal stress to shear stress. Detailed Implementation

[0029] This invention proposes a rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces. In order to make the advantages and technical solutions of this invention clearer and more explicit, the invention will be further described below with reference to specific embodiments.

[0030] like Figure 1 As shown, the present invention provides a method for determining rock shear criticality based on automatic identification of three-dimensional curvature of joint surfaces, which specifically includes the following steps: Step 1: Take samples and prepare rock specimens. Conduct direct shear tests on the rock specimens under different normal stress conditions to obtain the mechanical parameters and joint surface data of the rock specimens before and after shearing.

[0031] First, rock samples were taken from the cross-section at the engineering site, and point cloud data was acquired using a handheld 3D scanner. Complete limestone samples were selected as test materials, and cylindrical specimens with a diameter of 50 mm and a height of 100 mm and cubic specimens of 50 mm × 50 mm × 50 mm were prepared according to standards. Uniaxial compression and direct shear tests were conducted using the ROCK TEST SYSTEM 816.01 shear testing system. First, the average compressive strength of the cylindrical specimens was measured; in this example, it was 112.5 MPa. Then, direct shear tests were conducted on the cubic specimens with different normal stresses, set as 20% (22.5 MPa), 40% (45 MPa), 60% (67.5 MPa), and 80% (90 MPa) of the average compressive strength of the cylindrical specimens. During the direct shear test, the lower shear box was kept stationary, while the upper shear box was loaded backward at a constant rate of 0.05 mm / min until the specimen failed.

[0032] Step Two, as follows Figure 2 As shown, the joint surface of the sheared rock sample is image-processed using a binarization method based on the Otsu algorithm. The microscopic features of the shear and tensile regions of the joint surface are observed by scanning electron microscopy to determine the shear damage region and tensile damage region of the joint surface. The proportion of shear damage area of ​​the joint surface under different normal stress conditions is then quantitatively calculated.

[0033] After shear failure, the sample was removed and the resulting upper and lower joint surfaces were photographed in high resolution. A digital camera with a resolution of 5184×3456 was used for photography, employing center-weighted metering mode, an aperture of f / 12, and maintaining a shooting distance of 50cm. A tripod was used to ensure stable shooting. The acquired images are RGB color images with a bit depth of 24 bits.

[0034] The acquired joint surface images were preprocessed. First, Gaussian filtering with a kernel size of (5,5) was used to remove noise. Then, adaptive histogram equalization (CLAHE) was used to enhance image contrast, with parameters set to clipLimit=0.02 and tileGridSize=(8,8). Subsequently, Otsu's automatic thresholding method was used for binarization to distinguish the joint surface into black and white regions.

[0035] like Figure 3 As shown, scanning electron microscopy (SEM) was used to observe the black and white areas on the joint surface. The SEM acceleration voltage was set to 15 kV, the working distance to 10 mm, and the magnification to 500x. The observation results showed that the black area had an uneven surface with high roughness and numerous microcracks and pores, indicating tensile damage. The white area had a relatively smooth and flat surface without obvious cracks, exhibiting typical shear mirror surface characteristics, indicating shear damage.

[0036] The proportion of white pixels to total pixels on the joint surface was calculated to obtain the shear damage area percentage. Analysis shows that the proportion of shear damage area on the joint surface gradually increases with increasing normal stress. Under different normal stress conditions (20%, 40%, 60%, 80%), the shear damage area percentages for the upper joint surface were 43.28%, 52.11%, 65.12%, and 70.43%, respectively, while those for the lower joint surface were 43.27%, 52.42%, 64.25%, and 69.61%, respectively.

[0037] Step 3, as follows Figure 4 As shown, a method for quantifying the geometric features of joint surfaces based on automatic three-dimensional curvature identification is established: three-dimensional scanning technology is used to obtain three-dimensional point cloud data of joint surfaces; based on Otsu threshold segmentation and iterative optimization strategy, the three-dimensional curvature intervals of joint surfaces are automatically identified, and the distribution ratio of different curvature intervals is calculated as a three-dimensional feature index for shear criticality judgment.

[0038] The sheared joint surface was scanned in three dimensions using a Geomagic Capture ultra-precision blue LED 3D scanner to obtain 3D point cloud data. The scanner has a resolution of 0.06 mm and an error range controlled within ±0.025 mm. During the scanning process, the sample was fixed on a rotating platform, and multi-angle scanning was performed to ensure no blind spots; each scan took 5 minutes. The acquired point cloud data was automatically stitched, optimized, and fused using Geomagic processing software to ultimately form a complete 3D point cloud model of the joint surface.

[0039] The curvature of the acquired 3D point cloud data is calculated using the automatic curvature recognition method proposed in this invention, which integrates Otsu threshold segmentation and iterative optimization. The specific steps are as follows: (1) Curvature data normalization: The curvature distribution of points is transformed into a probability distribution; (1); In formula (1): For having curvature value The number of points; This represents the total points. For having curvature value The proportion or frequency of a point in the total number of points; (2) Calculate the category probability: Calculate the probability at a given threshold. and In the case of points distributed within different curvature intervals; (2-1); (2-2); (2-3); In formula (2): ， The curvature threshold is used to classify points into three categories; For curvature values ​​less than or equal to The probability of a point; For curvature values ​​greater than and less than or equal to The probability of a point; For curvature values ​​greater than The probability of a point; (3) Calculate the average curvature of each category: Calculate the average curvature value of points within each category to reflect the curvature characteristics of each category; (3-1); (3-2); (3-3); In equation (3-1): For curvature values ​​greater than 0 and less than or equal to 0 The average curvature of the point; In equation (3-2): For curvature values ​​greater than and less than or equal to The average curvature of the point; In equation (3-3): For curvature values ​​greater than The average curvature of the point; (4) Calculate the overall average curvature: Calculate the average curvature of the entire point set as a global reference value; (4); In equation (4): The average curvature value across all points; This represents the total points. (5) Calculate the inter-class variance: given a threshold and Below, the curvature difference between different categories is used to judge the segmentation effect; (5); In formula (5): Represents inter-class variance, which measures the curvature difference between different classes; (6) Finding the optimal threshold pair: representing the inter-class variance by the difference between the class probability and the class mean curvature; (6); (7) Iterative optimization of threshold: By traversing all possible combinations of thresholds Find the optimal threshold pair that maximizes the inter-class variance. This optimal threshold is used to distinguish different curvature categories, achieving optimal segmentation of the point cloud. (7); In equation (7): The optimal curvature threshold range maximizes the inter-class variance.

[0040] Based on the above method, the curvature of the joint surface is divided into a low curvature interval (0-t1*), a medium curvature interval (t1*-t2*), and a high curvature interval (t2*-maximum curvature value), and the proportion of points in each interval to the total number of points is calculated. For example... Figure 5 The percentage of low curvature surfaces on the upper and lower joint surfaces is shown under different normal stress conditions. Analysis indicates that the percentage of high curvature surfaces on the joint surfaces gradually decreases with increasing normal stress. Under different normal stress conditions (20%, 40%, 60%, and 80%), the percentages of high curvature on the upper joint surface are 77.66%, 61.45%, 51.56%, and 55.59%, respectively, while those on the lower joint surface are 71.03%, 63.42%, 49.53%, and 47.49%, respectively.

[0041] Step 4: Analyze the changing trends of the proportion of shear damage area and high curvature of joint surfaces under different normal stress conditions, establish the relationship between the ratio of effective normal stress and shear stress and the shear criticality, and further determine the shear critical condition.

[0042] To determine the critical conditions for shear fracture in limestone, the ratio of effective normal stress to shear stress under different working conditions was calculated, and its relationship with the proportion of shear damage area and high curvature of joint surfaces was analyzed. Figure 6 As shown, the study found that: (1) As the ratio of effective normal stress to shear stress increases, the change in the proportion of shear damage area on the joint surface can be divided into three stages: slow growth stage (ratio 0.8-1.4), rapid growth stage (ratio 1.4-1.6), and high-level slow growth stage (ratio 1.6-2.0). (2) As the ratio of effective normal stress to shear stress increases, the change in the proportion of high curvature of joint surfaces can also be divided into three stages: the initial slow decline stage (ratio 0.8-1.4), the rapid decline stage (ratio 1.4-1.6), and the slow decay stage (ratio 1.6-2.0). (3) Linear fitting is performed on the proportion of shear damage area and the proportion of high curvature of joint surface. The x-coordinate of the intersection of the fitted lines falls within the interval [1.4, 1.6]. At this time, the proportion of shear damage area and the proportion of high curvature of joint surface are both about 60%.

[0043] Step 5: Collect on-site data through the acoustic emission monitoring system and compare it with the test data obtained in Step 4 to verify the accuracy of the test data.

[0044] During the direct shear test, an acoustic emission monitoring system (PCI-2 type acoustic emission instrument) was used to collect the acoustic emission signals released during the failure of the limestone. Four acoustic emission sensors were arranged on the sample surface, with a frequency range of 100-1000kHz, a sampling rate of 5MHz, a preset gain of 40dB, and a threshold of 45dB.

[0045] By comparing field and experimental data, the cumulative energy of peak-front acoustic emission and damage variables under different effective normal stress to shear stress ratios were analyzed, such as... Figure 7 , Figure 8 As shown. Figure 7 The curves show the cumulative acoustic emission energy variation under different effective normal stress to shear stress ratios. Figure 8 The figures show the damage variable variation curves under different effective normal stress to shear stress ratios. The results indicate that when the ratio is less than 1.4, the cumulative acoustic emission energy and damage variable increase slowly; within the ratio range of 1.4-1.6, both cumulative acoustic emission energy and damage variable begin to increase rapidly; and after the ratio exceeds 1.6, both exhibit explosive growth. This result further verifies that the effective normal stress to shear stress ratio range [1.4, 1.6] is the critical threshold range for inducing shear fracture in limestone. If the results are incorrect, repeat the above steps.

[0046] Based on the above analysis, this invention determines the effective normal stress to shear stress ratio range [1.4, 1.6] as the critical threshold range for limestone shear fracture. When this ratio exceeds 1.6, the rock will undergo shear instability.

[0047] Any parts not mentioned in this invention can be achieved by referring to existing technologies.

[0048] Those skilled in the art should recognize that the above embodiments are only used to illustrate this application and are not intended to limit this application. Any appropriate changes and variations made to the above embodiments within the essential spirit and scope of this application fall within the scope of protection claimed in this application.

Claims

1. A method for determining the critical shear strength of rocks based on automatic identification of the three-dimensional curvature of joint surfaces, characterized in that, Includes the following steps: a. Take samples and prepare rock specimens, and conduct direct shear tests on the rock specimens under different normal stress conditions to obtain the mechanical parameters and joint surface data of the rock specimens before and after shearing. b. The joint surface of the rock sample after shearing is image processed by the binarization method based on the Otsu algorithm. The microscopic features of the shear and tensile regions of the joint surface are observed by scanning electron microscopy to determine the shear damage area and tensile damage area of ​​the joint surface. The proportion of shear damage area of ​​the joint surface under different normal stress conditions is quantitatively calculated. c. Establish a method for quantifying the geometric features of joint surfaces based on automatic identification of three-dimensional curvature: three-dimensional scanning technology is used to obtain three-dimensional point cloud data of joint surfaces; based on Otsu threshold segmentation and iterative optimization strategy, the three-dimensional curvature intervals of joint surfaces are automatically identified, and the distribution ratio of different curvature intervals is calculated as a three-dimensional feature index for shear criticality judgment. d. Analyze the changing trends of the proportion of shear damage area and high curvature of joint surfaces under different normal stress conditions, establish the relationship between the ratio of effective normal stress and shear stress and the shear criticality, and further determine the shear critical condition; e. Collect on-site data through the acoustic emission monitoring system and compare it with the test data obtained in step d to verify the accuracy of the test data.

2. The rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces according to claim 1, characterized in that: In step a, rock cross-section samples are taken at the engineering site, and point cloud data is obtained using a handheld 3D scanner. The method for preparing rock samples is as follows: through precise core taking, cutting and grinding processes, cylindrical samples with a diameter of 50mm and a height of 100mm and cubic samples of 50mm×50mm×50mm are prepared. The parallelism of the cross-section of all rock samples is controlled within ±0.02mm.

3. The rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces according to claim 1, characterized in that: In step a, the different normal stress conditions are 20%, 40%, 60%, and 80% of the average compressive strength of the rock sample.

4. The rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces according to claim 1, characterized in that: In step b, the binarization method uses Gaussian filtering with a kernel size of (5,5) for preprocessing, and adaptive histogram equalization is used to enhance image contrast. The parameters are set to clipLimit=0.02 and tileGridSize=(8,8). The Otsu algorithm is applied for binarization to divide the joint surface of the rock sample into black and white regions. The black region is the tensile damage region of the joint surface, and the white region is the shear damage region of the joint surface.

5. The rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces according to claim 1, characterized in that: In step b, the proportion of shear damage area is defined as the ratio of the number of white pixels on the joint surface to the total number of pixels.

6. The rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces according to claim 1, characterized in that: In step c, the specific steps based on the Otsu threshold segmentation and iterative optimization strategy include: (1) Curvature data normalization: The curvature distribution of points is transformed into a probability distribution; （1）； In formula (1): For having curvature value The number of points; This represents the total points. For having curvature value The proportion or frequency of a point in the total number of points; (2) Calculate the category probability: Calculate the probability at a given threshold. and In the case of points distributed within different curvature intervals; （2-1）； （2-2）； （2-3）； In formula (2): ， The curvature threshold is used to classify points into three categories; For curvature values ​​less than or equal to The probability of a point; For curvature values ​​greater than and less than or equal to The probability of a point; For curvature values ​​greater than The probability of a point; (3) Calculate the average curvature of each category: Calculate the average curvature value of points within each category to reflect the curvature characteristics of each category; （3-1）； （3-2）； （3-3）； In equation (3-1): For curvature values ​​greater than 0 and less than or equal to 0 The average curvature of the point; In equation (3-2): For curvature values ​​greater than and less than or equal to The average curvature of the point; In equation (3-3): For curvature values ​​greater than The average curvature of the point; (4) Calculate the overall average curvature: Calculate the average curvature of the entire point set as a global reference value; （4）； In equation (4): The average curvature value across all points; This represents the total points. (5) Calculate the inter-class variance: given a threshold and Below, the curvature difference between different categories is used to judge the segmentation effect; （5）； In formula (5): Represents inter-class variance, which measures the curvature difference between different classes; (6) Finding the optimal threshold pair: representing the inter-class variance by the difference between the class probability and the class mean curvature; （6）； (7) Iterative optimization of threshold: By traversing all possible combinations of thresholds Find the optimal threshold pair that maximizes the inter-class variance. This optimal threshold is used to distinguish different curvature categories, achieving optimal segmentation of the point cloud. （7）； In equation (7): The optimal curvature threshold range maximizes the inter-class variance.

7. The rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces according to claim 1, characterized in that: In step d, when the rock sample is limestone, the determined critical shear condition range is [1.4, 1.6]. When this ratio exceeds 1.6, the rock will experience critical shear. When the rock sample is sandstone, the determined critical shear condition range is [1.3, 1.5]. When this ratio exceeds 1.5, the rock will experience critical shear. When the rock sample is granite, the determined critical shear condition range is [1.5, 1.7]. When this ratio exceeds 1.7, the rock will experience shear instability.

8. The rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces according to claim 1, characterized in that: In step e, the acoustic emission monitoring system includes acoustic emission sensors. Several acoustic emission sensors are arranged on the surface of the rock sample. The frequency range of each acoustic emission sensor is 100 to 1000 kHz, the sampling rate is set to 5 MHz, the preset gain is 40 dB, and the threshold is set to 45 dB.

9. The rock shear criticality discrimination method based on automatic identification of three-dimensional curvature of joint surfaces according to claim 1, characterized in that: In step e, if the field data and the experimental data are inconsistent, steps a to d are repeated sequentially.

Images