Inversion method and inversion system for uneven stress distribution in heterogeneous rock

Abstract

The invention relates to an inversion method and an inversion system for uneven stress distribution in heterogeneous rocks, and belongs to the technical field of deformation and damage of heterogeneous rocks. Aiming at the problem that the existing numerical simulation method is difficult to accurately reflect the deformation and failure characteristics of the heterogeneous rock, the method comprises the following steps: selecting and pretreating a heterogeneous rock sample, and carrying out nanoindentation test on the heterogeneous rock sample to obtain the elasticity modulus of composition minerals of the sample; the method comprises the following steps: extracting composition mineral types and distribution ranges of a heterogeneous rock sample to obtain an elastic modulus distribution matrix, monitoring a test deformation process of the heterogeneous rock sample, calculating a strain field of the heterogeneous rock sample to obtain a strain field matrix, and calculating a non-uniform stress distribution field through the elastic modulus distribution matrix and the strain field matrix. And comparing the calculated stress value with the applied stress value, optimizing a threshold interval of composition minerals of the sample, and outputting a non-uniform stress distribution field. Compared with a numerical simulation method, the method can evaluate the stress field more visually and accurately, and has important physical significance.

Description

Technical Field

[0001] The invention relates to the technical field of deformation and failure of heterogeneous rocks, and in particular to an inversion method and an inversion system for uneven stress distribution inside heterogeneous rocks. Background Art

[0002] The study of deformation and failure of heterogeneous rocks is one of the basic issues in the field of geotechnical engineering. Exploring the deformation and failure mechanism of heterogeneous rocks is an important research direction in geotechnical engineering. The deformation and failure characteristics of heterogeneous rocks are strongly affected by stress concentration and stress redistribution caused by the differential deformation of the constituent minerals.

[0003] So far, the evaluation of stress concentration and inversion of differential stress fields in heterogeneous rocks still mainly rely on numerical simulation methods. However, the existing numerical simulation methods are limited by assumptions, boundary conditions, and computing resources, and it is difficult to accurately reflect the deformation and failure characteristics of heterogeneous rocks. In addition, most of the parameters in the numerical simulation methods still need to be obtained from indoor experiments.

[0004] Based on the above situation, how to monitor stress concentration and stress redistribution in indoor experiments and intuitively display the high stress concentration areas of heterogeneous rocks are the key points and difficulties that need to be solved in studying the deformation and fracture mechanism of heterogeneous rocks. Summary of the invention

[0005] In view of the deficiencies in the above-mentioned prior art, the purpose of the present invention is to provide an inversion method and inversion system for the uneven stress distribution inside heterogeneous rocks. Based on the idea that the deformation of the rock before crack initiation is elastic deformation, and the stress field in the elastic stage is obtained by inverting the strain field and the elastic modulus of the mineral, nanoindentation testing is combined with DIC technology and rock microscopic image analysis. The elastic modulus of various minerals in the rock is obtained by nanoindentation testing, and the deformation field of the mineral is obtained by DIC, so as to calculate the uneven stress distribution of the heterogeneous rock, and realize the inversion of the uneven stress field in the process of rock deformation and destruction in indoor experiments, thereby quantitatively revealing the deformation and destruction mechanism of heterogeneous rocks.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention discloses an inversion method for uneven stress distribution inside heterogeneous rocks, comprising the following steps:

[0008] S1: Select heterogeneous rock samples and pre-treat them;

[0009] S2: Perform nanoindentation test on the pretreated sample to obtain the elastic modulus of the constituent minerals of the sample;

[0010] S3: extracting the constituent mineral types and distribution range of the sample, and calculating the elastic modulus distribution matrix based on the constituent mineral types and distribution range and the elastic modulus of the constituent minerals of the sample;

[0011] S4: Conduct mechanical property test on the sample, load it with static load, monitor the stress loading deformation process of the sample, calculate the strain field of the sample, and obtain the strain field matrix;

[0012] S5: Calculate the non-uniform stress distribution field through the elastic modulus distribution matrix and the strain field matrix;

[0013] S6: Compare the stress value calculated in step S5 with the stress value loaded in step S4, and output the non-uniform stress distribution field of the sample when the calculated stress value corresponds to the loaded stress value by optimizing the distribution range threshold interval of the constituent minerals of the sample.

[0014] A further preferred embodiment of the above technical solution is: in step S2, the step of obtaining the elastic modulus of the constituent minerals of the sample comprises:

[0015] S21: Fix the sample with hot melt adhesive and magnetic patch;

[0016] S22: Nanoindentation test was performed on the sample using quasi-static dot-marking mode to obtain the indentation depth, load, correction factor and Poisson's ratio;

[0017] S23: Calculate the elastic modulus of the constituent minerals of the sample based on the indentation depth, load, correction factor and Poisson's ratio.

[0018] As a further preferred embodiment: in step S23, the elastic modulus of the constituent minerals of the sample is calculated as follows:

[0019]

[0020] Where: E r is the indentation modulus of the sample, E i is the elastic modulus of the indenter, υ i is the Poisson’s ratio of the pressure head, and υ is the Poisson’s ratio of the rock sample.

[0021] Based on the above scheme, further, in step S3, the specific process of extracting the constituent mineral types and distribution range of the sample is as follows:

[0022] S31: taking photos of the surface of the sample after pretreatment;

[0023] S32: converting the obtained rock photo into a grayscale image to obtain a grayscale distribution histogram;

[0024] S33: Determine the segmentation threshold according to the grayscale distribution histogram and the empirical range of mineral grayscale distribution, perform threshold segmentation in MATLAB, and extract the constituent mineral types and distribution ranges.

[0025] As a further preferred embodiment of the above scheme: in step S4, the specific steps of monitoring the test deformation process of the sample include:

[0026] S41: Connect the loader and height camera to the same system, and fix the specimen on the loader by a wedge-shaped fixture;

[0027] S42: starting the tensile test and image acquisition of the sample at the same time, stopping the test and image acquisition at the same time after the sample is destroyed, and saving the experimental data and images;

[0028] S43: Extract the image data of the sample before the crack initiation, select the ROI area in the center of the image, and use the forward accumulation FA-GN method in MATLAB to calculate the strain field of the ROI area to obtain the strain field matrix.

[0029] A further preferred solution is: in step S5, the specific process of calculating the non-uniform stress distribution field through the elastic modulus distribution matrix and the strain field matrix is ​​as follows:

[0030] According to the ratio of the ROI area to the heterogeneous rock sample image, the matrix of the corresponding part of the ROI in the elastic modulus distribution matrix is ​​extracted;

[0031] Multiply the strain field matrix with the matrix corresponding coordinates of the corresponding part of the ROI to obtain the matrix of stress field distribution;

[0032] According to the matrix of stress field distribution, the uneven stress field distribution diagram is drawn in MATLAB and the uneven stress value is output.

[0033] A further preferred solution is: in step S6, when comparing the calculated stress value with the loaded stress value, if the difference between the two exceeds ±15% of the actual stress value, the distribution range threshold interval of the constituent minerals of the sample is re-modified, and after the mineral distribution is redivided, it is imported into MATLAB again for matrix calculation until the output stress value corresponds to the stress value loaded in the mechanical property test experiment, and the uneven stress distribution field of the heterogeneous rock sample at this time is output.

[0034] In a second aspect, the present invention discloses an inversion system for uneven stress distribution inside heterogeneous rocks, comprising:

[0035] Input module, inputting nanoindentation test data of the sample;

[0036] The threshold segmentation and extraction module takes a photo of the pre-processed sample surface and converts it into a grayscale image, performs threshold segmentation on the grayscale image, and extracts the distribution of different mineral components;

[0037] The test monitoring module is used to monitor the specimen during the loading and deformation process and extract the ROI area of ​​the monitoring image;

[0038] A calculation module, which is called to realize the calculation of elastic modulus distribution matrix, strain field matrix and non-uniform stress distribution field;

[0039] The comparison and calibration module compares the calculated stress value with the loaded stress value and calibrates the deviation, and outputs the non-uniform stress distribution field of the sample when the calculated stress value corresponds to the loaded stress value.

[0040] In a third aspect, the present invention discloses an electronic device, including a processor and a memory, wherein the memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory to implement the steps of any of the above methods.

[0041] In a fourth aspect, the present invention discloses a computer-readable storage medium storing computer instructions, where the computer instructions are used to enable a computer to execute the steps of any of the above methods.

[0042] Compared with the prior art, the technical solution of the present invention has the following effects:

[0043] 1. The present invention combines digital image correlation technology, nanoindentation test method and mineral threshold segmentation method to monitor the distribution of the inhomogeneous stress field of heterogeneous rocks during crack initiation in indoor tests. The deformation field of heterogeneous rocks can be obtained by digital image correlation technology, the elastic parameters of the constituent minerals of the sample can be obtained by nanoindentation test, and the distribution characteristics of minerals can be obtained by threshold segmentation method, thus realizing accurate inversion of the inhomogeneous stress field inside heterogeneous rocks. It is more intuitive than conventional numerical simulation methods, has clearer physical meaning, and is not limited by various simplified assumptions in numerical simulation methods, making the assessment of the stress field of heterogeneous rocks more accurate.

[0044] 2. The method of the present invention can accurately obtain the stress distribution of heterogeneous rocks, provide a powerful means for analyzing the stress redistribution and regulation inside the rock under the action of external loads, quantitatively reveal the mechanism of deformation and failure of heterogeneous rocks, and provide important references for the stability assessment of deformation and failure of heterogeneous rock materials and the selection of protective facilities in the study of rock mechanics and engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required in the embodiments are briefly introduced below.

[0046] Figure 1 The present invention is a flowchart of the overall process of an inversion method for uneven stress distribution inside heterogeneous rocks.

[0047] Figure 2 This is the shape of the heterogeneous rock sample of the present invention.

[0048] Figure 3a This is the load-displacement curve of the nanoindentation test of the quartz sample in the present invention.

[0049] Figure 3b This is the load-displacement curve of the nanoindentation test of the andesine sample of the present invention.

[0050] Figure 3c This is the load-displacement curve of the nanoindentation test of the mica sample in the present invention.

[0051] Figure 4 It is a grayscale image of the heterogeneous rock sample in the present invention.

[0052] Figure 5 It is the grayscale distribution histogram of the heterogeneous rock sample in the present invention.

[0053] Figure 6 This is the n-value diagram of the heterogeneous rock sample in the present invention.

[0054] Figure 7 This is the load-time curve of the experimental loading of the heterogeneous rock sample in the present invention.

[0055] Figure 8 It is the rock strain field obtained by DIC calculation in the present invention.

[0056] Fig. 9 The non-uniform stress distribution of a heterogeneous rock sample is obtained by inversion using the method of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme in the embodiments will be described clearly and completely in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments, not all of the embodiments. For ordinary technicians in this field, other embodiments can be obtained without creative work.

[0058] Reference Figure 1-9The present invention proposes an inversion method for the uneven stress distribution inside heterogeneous rocks, which combines nanoindentation testing with digital image correlation technology. The purpose is to overcome the defect that the numerical simulation method in the prior art is difficult to accurately evaluate the uneven stress distribution of heterogeneous rocks, and to achieve the inversion of the uneven stress field in the rock deformation and failure process in indoor experiments, thereby revealing the deformation and failure mechanism of heterogeneous rocks.

[0059] The nanoindentation test used in the present invention is a powerful means to test the nanoscale mechanical properties of mineral units. Its basic principle is to press a sharp rigid tip into the surface of the sample under a set load, and the force and displacement (indentation depth) of the indenter will be monitored in real time to obtain a load-displacement curve, and calculate the elastic modulus, hardness, fracture toughness and creep coefficient of the material. Nanoindentation testing has the advantages of simple operation, high resolution, non-destructiveness and flexibility; in addition, it can be combined with electron microscopes SEM, atomic force microscopes AFM, etc., and is widely used in micromechanical testing of rocks, and can obtain accurate elastic characteristics of minerals.

[0060] The digital image correlation (DIC) technology used in the present invention is a non-contact optical measurement technology. Due to its many advantages such as simple principle, easy implementation, wide measurement range and high degree of automation, it has been widely used in strain measurement in rock compression / stretching process. The basic principle of DIC is to obtain the displacement vector of the pixel by tracking (or matching) the position of the same pixel in two images before and after the deformation of the object surface, so as to obtain the displacement field of the specimen surface. First, it is necessary to make the imaging surface of the specimen have a marker point that can reflect the deformation information, and the minerals in the rock are the marker points with good recognition; then, during the test process, the images of the specimen surface before and after loading are collected, and finally, the displacement information of the specimen surface is obtained by using the software program to adopt the relevant mathematical algorithm. In the study of rock mechanics, DIC is used to measure the real-time strain characteristics of the sample during the test process, and the DIC technology can give the strain distribution of the whole field in the sample, which provides a good way to analyze and study the deformation behavior and failure fracture mechanism of the rock.

[0061] Based on nanoindentation testing and DIC technology, the main implementation steps of the method of the present invention include the following parts:

[0062] (1) Determination of rock mineral elastic modulus: Select heterogeneous rock samples, polish the sample surface, conduct nanoindentation test on the constituent minerals of the polished samples, obtain the indentation depth, load, correction factor and Poisson's ratio, and calculate the elastic modulus of the constituent minerals of the sample based on the indentation depth, load, correction factor and Poisson's ratio;

[0063] (2) Rock mineral threshold segmentation extraction: Take a photo of the polished sample surface, convert the rock photo into a grayscale image, and obtain a grayscale distribution histogram. Determine the segmentation threshold based on the visual method, grayscale histogram, and mineral empirical value. Use the threshold segmentation method to extract the constituent mineral types and distribution ranges of the sample. Based on the constituent mineral types and distribution ranges and the elastic modulus of the constituent minerals of the sample, calculate the constituent mineral types and distribution ranges obtained by the threshold segmentation method with the corresponding mineral elastic modulus obtained by nanoindentation testing to obtain the elastic modulus distribution matrix.

[0064] (3) Calculation of sample strain field: A high-speed camera is used to monitor the loading and deformation process of the sample during the indoor test. A non-stroboscopic fill light is used to ensure sufficient light during the monitoring process. The image of the continuous deformation and failure process of the sample is obtained. The image is imported into MATLAB, and the strain field of the sample is calculated using the Digital Image Correlation (DIC) technology to obtain the strain field matrix of the sample.

[0065] (4) Computer verification of the non-uniform stress field of the specimen: Based on the theory of elastic mechanics and rock mechanics, the deformation of the rock before crack initiation is elastic deformation. The stress-strain relationship at this stage follows Hooke's theorem. Therefore, the non-uniform stress distribution field is calculated by the elastic modulus distribution matrix and the strain field matrix. The calculated stress value is compared with the stress value of the test load to see if they are close. If the calculated stress value differs too much from the actual stress value applied in the test, the mineral distribution is re-divided by adjusting the threshold interval of the constituent minerals of the specimen and the new round of mineral distribution matrix is ​​substituted for recalculation until the difference between the calculated stress value and the stress value of the test load is less than ±15% of the actual stress value. At this time, the non-uniform stress distribution field of the specimen is output.

[0066] The specific implementation modes of the present invention are further described in detail with reference to the following examples.

[0067] Example:

[0068] In this embodiment, a common heterogeneous rock (gneiss) in rock engineering is selected for tensile testing, and the method of the present invention is applied to calculate its non-uniform stress field.

[0069] like Figure 2As shown in Figure 1, the gneiss material was processed into a dog-bone specimen with a size of 50 mm × 10 mm × 3 mm. The surface of the gneiss specimen was polished smooth. The Ti980 (SLAB-SPM-0004) nanoindenter of Hysitron Corporation of the United States was used to carry out nanoindentation test of the constituent minerals of the gneiss specimen in accordance with the GBT 22458-2008 standard. The test adopted a quasi-static dot mode, and the gneiss specimen was fixed with hot melt adhesive + magnetic patch. The indenter information was a standard Berkovich (diamond) probe, the halfangle was 65.27°, and the included angle was 142.30°. The load-displacement curve obtained by the nanoindentation test is shown in Figure 11. Figure 3a-3c shown.

[0070] Since the penetration depth of the nanoindentation test is much smaller than the size and thickness of the gneiss sample being tested, the relationship between load and penetration depth is calculated by the following formula:

[0071] P = a(hh f ) m (1)

[0072] Where: P is the indenter load, a and m are the parameters obtained by the least squares fitting method, and h is the indentation depth; f is the contact depth after unloading;

[0073] Put equation (1) in h max Differentiate the elastic contact stiffness of the gneiss sample:

[0074]

[0075] Where: S is the contact stiffness;

[0076] Indentation contact depth h c It can be derived from the following expression:

[0077]

[0078] Where: h c is the indentation contact depth; h max is the maximum indentation depth; ε is a parameter related to the shape of the indenter, and ε of the Berkovich indenter is 0.7268; P max is the peak load of the pressure head;

[0079] The contact area between the indenter and the gneiss sample can be expressed as:

[0080]

[0081] Where: A c is the projected area of ​​the contact area;

[0082] Calculate the indentation modulus E of the tested sample r The calculation formulas for the indentation hardness H are as follows:

[0083]

[0084] Where: E r is the indentation modulus; β is the correction coefficient related to the shape of the indenter, β of the Berkovich indenter is 1.034; H is the indentation hardness;.

[0085] The actual elastic modulus E of the gneiss sample under test is calculated as follows:

[0086]

[0087] Where: E i and i and are the elastic modulus and Poisson’s ratio of the indenter, E i =1141GPa,υ i =0.07, υ is the Poisson's ratio of the gneiss sample being tested. In this embodiment, the Poisson's ratios of quartz, feldspar and mica taken from the gneiss sample are 0.15, 0.2 and 0.3 respectively.

[0088] Furthermore, the surface of the polished gneiss sample was photographed. To ensure the correct registration of the image, the VEO4K990 high-speed camera produced by Phantom used in DIC monitoring was used for the photography. The pixel size was adjusted to 1920×1080 pixels. Figure 4 As shown, the rock photos in RGB space are then converted into grayscale images (0-255) in MATLAB to obtain a grayscale distribution histogram (as shown in Figure 5 ); According to the grayscale distribution histogram and the empirical range of mineral grayscale distribution, the grayscale threshold range of the minerals in the gneiss used in this embodiment is: mica 0-90, quartz 90-150, feldspar 180-255, and then threshold segmentation is performed in MATLAB to extract the distribution range of different minerals;

[0089] like Figure 6 As shown, the grayscale image is converted into n-values, where n is the number of main mineral types. The gneiss used in this embodiment has n of 3. The grayscale value is converted into the elastic modulus of the mineral in GPa in MATLAB. According to the results of the nanoindentation test, the elastic modulus of quartz is 116.0 GPa, feldspar is 73.8 GPa, and mica is 46.9 GPa, and a matrix T1 (1920×1080) containing coordinates and elastic modulus is obtained.

[0090] Next, the DIC technique was used to monitor the deformation process of the stress loading test of the gneiss sample. The specific test process is as follows:

[0091] The loader used in this embodiment is a FullamSEM2000 tensile tester. The gneiss sample is fixed on the loader by a specific wedge-shaped fixture. The loading method is displacement loading, and the loading rate is 0.02 mm / min. Before the test starts, the gneiss sample is preloaded with a tensile force of 5 N, and the displacement is adjusted to zero. The VEO4K990 high-speed camera produced by Phantom is used to collect images using the DIC method. The image collection rate is 500 frames / second, and the image resolution is 1920×1080 pixels. The Newell CB200C non-stroboscopic fill light is used to ensure a clear image of the sample surface. The loader and the high-speed camera are connected to the same system to ensure the consistency of the start time and the end time. When all equipment is ready, the tensile test and image collection of the gneiss sample are started at the same time. After the gneiss is damaged, the test and image collection are stopped at the same time, and the test data and images are saved. The load-time curve of the test loading is as shown in the figure. Figure 7 shown.

[0092] Then, the image data before crack initiation was extracted, and the 20×6 mm range in the center of the gneiss sample was selected as the region of interest (ROI). The pixel size of the ROI was 1800×550 pixels. The forward accumulation Gauss-Newton method (FA-GN) was used in MATLAB to calculate the strain field in the ROI of the gneiss sample. The results were referred to Figure 8 , and read the strain field matrix T2 in the ROI calculated by the DIC method. The size of the matrix is ​​1800×550. According to the ratio of the ROI and the gneiss sample image, the corresponding part of the ROI in the matrix T1 of the elastic modulus is extracted as the matrix T3. The strain field matrix T2 is multiplied by the corresponding coordinates in the elastic modulus matrix T3 to obtain the matrix T4 of the stress field distribution. According to the matrix T4 of the stress field distribution, 在 Draw the uneven stress field distribution diagram in MATLAB and output the uneven stress data.

[0093] Compare the calculated non-uniform stress value with the corresponding stress value in the load-time curve. If the difference between the two exceeds ±15% of the actual stress value, the threshold interval of the constituent minerals of the sample is revised, the mineral distribution is redivided, and the matrix calculation is performed again in MATLAB until the output stress value corresponds to the loaded stress value. The non-uniform stress distribution field of the gneiss sample at this time is output, such as Fig. 9 shown.

[0094] The inversion method for the inhomogeneous stress field inside heterogeneous rocks proposed in the present invention can capture the uneven stress redistribution inside the rocks caused by the difference in mineral elastic modulus in indoor experiments, and is closer to the real mechanical properties of heterogeneous rocks. The method makes up for the shortcomings of the numerical simulation method, such as many assumptions and difficult parameter acquisition, and provides a new method and idea for the study of the failure mechanism of heterogeneous rocks. Moreover, as the main rock and soil body of underground tunnel engineering and high and steep slope engineering, the accurate evaluation of the mechanical properties and the revelation of the mechanism of heterogeneous rocks are very important for the stability and safety of rock engineering. The method of the present invention effectively solves the problem that the inhomogeneous stress and stress redistribution inside heterogeneous rocks are difficult to intuitively characterize and evaluate, and has important practical significance.

[0095] It should be clear that the above detailed description of the embodiments of the present invention provided in the accompanying drawings is only used to further illustrate the present invention and is not intended to limit the scope of the invention claimed for protection, but only represents selected embodiments of the present invention. Based on the embodiments of the present invention, some non-essential improvements and adjustments made by ordinary technicians in this field according to the above content of the present invention belong to the protection scope of the present invention.

Claims

1. An inversion method for uneven stress distribution inside heterogeneous rocks, characterized in that: The following steps are involved: S1: Select heterogeneous rock samples and pre-treat them; S2: Perform nanoindentation test on the pretreated sample to obtain the elastic modulus of the constituent minerals of the sample; S3: extracting the constituent mineral types and distribution range of the sample, and calculating the elastic modulus distribution matrix based on the constituent mineral types and distribution range and the elastic modulus of the constituent minerals of the sample; S4: Conduct mechanical property test on the sample, load it with static load, monitor the stress loading deformation process of the sample, calculate the strain field of the sample, and obtain the strain field matrix; S5: Calculate the non-uniform stress distribution field through the elastic modulus distribution matrix and the strain field matrix; S6: Compare the stress value calculated in step S5 with the stress value loaded in step S4, and output the non-uniform stress distribution field of the sample when the calculated stress value corresponds to the loaded stress value by optimizing the distribution range threshold interval of the constituent minerals of the sample.

2. The inversion method for inhomogeneous stress distribution inside heterogeneous rocks according to claim 1 is characterized in that: In step S2, the step of obtaining the elastic modulus of the constituent minerals of the sample includes: S21: Fix the sample with hot melt adhesive and magnetic patch; S22: Nanoindentation test was performed on the sample using quasi-static dot-marking mode to obtain the indentation depth, load, correction factor and Poisson's ratio; S23: Calculate the elastic modulus of the constituent minerals of the sample based on the indentation depth, load, correction factor and Poisson's ratio.

3. The inversion method for inhomogeneous stress distribution inside heterogeneous rocks according to claim 2 is characterized in that: In step S23, the elastic modulus of the constituent minerals of the sample is calculated using the following formula: Where: E r is the indentation modulus of the sample, E i is the elastic modulus of the indenter, υ i is the Poisson's ratio of the indenter, and υ is the Poisson's ratio of the sample.

4. The inversion method for inhomogeneous stress distribution inside heterogeneous rocks according to claim 3 is characterized in that: In step S3, the specific process of extracting the constituent mineral types and distribution range of the sample is as follows: S31: taking photos of the surface of the sample after pretreatment; S32: converting the obtained sample photo into a grayscale image to obtain a grayscale distribution histogram; S33: Determine the segmentation threshold according to the grayscale distribution histogram and the empirical range of mineral grayscale distribution, perform threshold segmentation in MATLAB, and extract the constituent mineral types and distribution ranges.

5. The inversion method for inhomogeneous stress distribution inside heterogeneous rocks according to claim 4, characterized in that: In step S4, the specific steps of monitoring the test deformation process of the sample include: S41: Connect the loader and height camera to the same system, and fix the specimen on the loader by a wedge-shaped fixture; S42: starting the tensile test and image acquisition of the sample at the same time, stopping the test and image acquisition at the same time after the sample is destroyed, and saving the experimental data and images; S43: Extract the image data of the sample before the crack initiation, select the ROI area in the center of the image, and use the forward accumulation FA-GN method in MATLAB to calculate the strain field of the ROI area to obtain the strain field matrix.

6. The inversion method for inhomogeneous stress distribution inside heterogeneous rocks according to claim 5, characterized in that: In step S5, the specific process of calculating the non-uniform stress distribution field through the elastic modulus distribution matrix and the strain field matrix is ​​as follows: S51: extracting the matrix of the corresponding part of the ROI in the elastic modulus distribution matrix according to the ratio of the ROI area to the sample image; S52: multiplying the strain field matrix with the matrix corresponding coordinates of the corresponding part of the ROI to obtain a matrix of stress field distribution; S53: According to the matrix of stress field distribution, a non-uniform stress field distribution diagram is drawn in MATLAB, and the non-uniform stress value is output.

7. The inversion method for inhomogeneous stress distribution inside heterogeneous rocks according to claim 5, characterized in that: In step S6, when comparing the calculated stress value with the loaded stress value, if the difference between the two exceeds ±15% of the actual stress value, the distribution range threshold interval of the constituent minerals of the sample is re-modified, and after the mineral distribution is redivided, it is imported into MATLAB again for matrix calculation until the output stress value corresponds to the stress value loaded in the mechanical property test experiment, and the non-uniform stress distribution field of the heterogeneous rock sample at this time is output.

8. An inversion system for uneven stress distribution inside heterogeneous rocks, characterized in that: The system is used to perform the steps of the method according to any one of claims 1 to 7, comprising: Input module, inputting nanoindentation test data of the sample; The threshold segmentation and extraction module takes a picture of the sample surface and converts it into a grayscale image, performs threshold segmentation on the grayscale image, and extracts the distribution range of different mineral compositions; The test monitoring module is used to monitor the specimen during stress loading and deformation, and extract the ROI area of ​​the monitoring image; A calculation module, which is called to respectively realize the calculation of the elastic modulus distribution matrix, the strain field matrix, and the non-uniform stress distribution field in claims 3-5; The comparison and calibration module compares the calculated stress value with the loaded stress value and calibrates the deviation, and outputs the uneven stress distribution field when the calculated stress value corresponds to the loaded stress value.

9. An electronic device comprising a processor and a memory, characterized in that: The memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory to implement the steps of the inversion method for uneven stress distribution inside heterogeneous rocks as described in any one of claims 1-7.

10. A computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to enable a computer to execute the steps of an inversion method for uneven stress distribution inside heterogeneous rocks as described in any one of claims 1-7.

Images