Visual characterization method, device and equipment for deformation rule of rock microstructure

Abstract

The invention provides a visual characterization method, device and equipment for a deformation rule of a rock microstructure, and relates to the technical field of oil-gas exploration. The method comprises the following steps: acquiring scanning grey-scale map data of an internal structure of a rock under different load conditions; establishing a rock three-dimensional digital core model under different load conditions according to the scanned grey-scale map data and the pre-trained neural network; performing pore structure feature extraction processing according to the rock three-dimensional digital core model to obtain rock pore structure feature parameters, and representing a dynamic deformation rule of the rock micro-pore structure along with effective stress loading according to the rock pore structure feature parameters; performing fracture morphological feature extraction processing according to the rock three-dimensional digital core model to obtain a rock fracture morphological feature parameter system, and representing a rock internal fracture morphological deformation evolution rule according to the rock fracture morphological feature parameter system. According to the method provided by the invention, the real rock micro-pore structure is accurately reflected.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil and gas exploration, and particularly relates to a method and device for visualizing rock microstructure deformation rules. BACKGROUND

[0002] The rock microstructure deformation rules are of great significance for revealing the changes of fluid flow in the reservoir, analyzing the stability of the reservoir, and guiding the development of oil and gas resources.

[0003] In the prior art, a three-dimensional digital core model of rock is reconstructed by using a simulated annealing method, a multi-point geostatistics method, or a Markov chain Monte Carlo numerical reconstruction method, so as to characterize the micro-pore structure of rock, usually through a small amount of two-dimensional rock thin section images.

[0004] However, the reconstruction effect of the numerical reconstruction method depends on the resolution and image size of the two-dimensional rock thin section images, and it is difficult to accurately reflect the real rock micro-pore structure. SUMMARY

[0005] The present application provides a method and device for visualizing rock microstructure deformation rules, which establishes a three-dimensional digital core model of rock under different load conditions by using scanning gray scale data of the internal structure of rock under different load conditions through a pre-trained neural network. Then, the rock microstructure deformation rules are characterized according to the three-dimensional digital core model of rock, so as to accurately reflect the real rock micro-pore structure.

[0006] In a first aspect, the present application provides a method for visualizing rock microstructure deformation rules, which comprises:

[0007] obtaining scanning gray scale data of the internal structure of rock under different load conditions;

[0008] establishing a three-dimensional digital core model of rock under different load conditions according to the scanning gray scale data and a pre-trained neural network;

[0009] performing pore structure feature extraction processing according to the three-dimensional digital core model of rock, obtaining rock pore structure feature parameters, and characterizing the dynamic deformation rules of rock micro-pore structure with effective stress loading according to the rock pore structure feature parameters;

[0010] performing fracture morphology feature extraction processing according to the three-dimensional digital core model of rock, obtaining a rock fracture morphology feature parameter system, and characterizing the deformation evolution rules of the internal fracture morphology of rock according to the rock fracture morphology feature parameter system.

[0011] In a possible design, the scanning gray scale data of the internal structure of rock under different load conditions is obtained, comprising:

[0012] acquire real-time stress value when the stress loading device loads the rock at a preset constant displacement speed;

[0013] When the real-time stress value reaches a fixed stress value, the stress loading device is paused, and the scanning grayscale image data of the internal structure of the rock under the load condition of the fixed stress value is collected by the rock scanning device;

[0014] Return to the step of acquiring the real-time stress value when the stress loading device loads the rock at a preset constant displacement speed, until the core of the rock is broken, and the scanning grayscale image data of the internal structure of the rock under different load conditions, stress change data and core axial displacement change data are obtained by the rock scanning device;

[0015] According to the stress change data and the core axial displacement change data, data processing is performed to obtain the stress-strain relationship of the rock.

[0016] In one possible design, according to the scanning grayscale image data and the pre-trained neural network, a three-dimensional digital core model of the rock under different load conditions is established, including:

[0017] According to the scanning grayscale image data and the pre-trained neural network, threshold segmentation processing is performed to obtain a two-dimensional distribution image of the skeleton and cracks-vacancies of the rock;

[0018] The two-dimensional distribution image is superimposed by a superimposition algorithm to establish a three-dimensional digital core model of the rock under different load conditions.

[0019] In one possible design, the superimposition algorithm is a moving stereoscopic algorithm, and the two-dimensional distribution image is an X-axis direction-Y-axis direction slice image;

[0020] Correspondingly, the three-dimensional digital core model of the rock under different load conditions is established by superimposing the two-dimensional distribution image by a superimposition algorithm, including:

[0021] The X-axis direction-Y-axis direction slice image is superimposed along the Z-axis direction by the moving stereoscopic algorithm to establish a three-dimensional digital core model of the rock under different load conditions.

[0022] In one possible design, the pre-trained neural network is obtained after model training using historical sample data, wherein the historical sample data is scanning grayscale sample image data of training sample rocks under different load conditions;

[0023] The model training process of the pre-trained neural network includes:

[0024] In response to the selection operation of the user, part of the image in the scanning grayscale sample image data is extracted from the historical sample data as a training sample set;

[0025] The rock pore, fracture and skeleton are trained by using a full convolutional neural network and a training sample set, and the accuracy and error rate of the neural network obtained through the last model training meet the preset training end condition, and the neural network obtained through the last model training is used as the pre-trained neural network.

[0026] In a possible design, the method further includes:

[0027] Obtaining stress change data and core axial displacement change data of the rock in the conversion process under different load conditions;

[0028] Generating a core stress-strain relationship curve according to the stress change data and the core axial displacement change data;

[0029] According to the preset strain equivalent assumption, a deep rock damage deformation constitutive relationship suitable for strong stress influence is established;

[0030] According to the deep rock damage deformation constitutive relationship and the core stress-strain relationship curve, fitting calculation and processing are performed to obtain the initial elastic modulus, peak stress and peak strain of the rock in the conversion process under different load conditions, so as to analyze the mechanical action mechanism.

[0031] In a possible design, the preset constant displacement speed is 0.03 millimeters per minute to 3.0 millimeters per minute.

[0032] In a second aspect, the application provides a device for visualizing the deformation rule of the microstructure of rock, which comprises:

[0033] The obtaining module is configured to obtain scanning grayscale image data of the internal structure of the rock under different load conditions;

[0034] The core model establishing module is configured to establish a three-dimensional digital core model of the rock under different load conditions according to the scanning grayscale image data and the pre-trained neural network;

[0035] The visualizing module is configured to perform pore structure feature extraction processing according to the three-dimensional digital core model of the rock, to obtain rock pore structure feature parameters, and to visualize the dynamic deformation rule of the micro-pore structure of the rock with effective stress loading according to the rock pore structure feature parameters;

[0036] The visualizing module is configured to perform fracture morphology feature extraction processing according to the three-dimensional digital core model of the rock, to obtain a rock fracture morphology feature parameter system, and to visualize the deformation evolution rule of the internal fracture morphology of the rock according to the rock fracture morphology feature parameter system.

[0037] In a possible design, the obtaining module comprises a stress value obtaining module, a data collecting module, a cyclic pressurizing module and a data processing module.

[0038] The stress value acquisition module is configured to acquire a real-time stress value when the stress loading device loads the rock at a preset constant displacement speed;

[0039] The data acquisition module is configured to pause the loading of the stress loading device when it is detected that the real-time stress value reaches a fixed stress value, and then acquire scanning grayscale image data of the internal structure of the rock under the load condition of the fixed stress value through the rock scanning device;

[0040] The cyclic pressurization module is configured to return to the step of acquiring the real-time stress value when the stress loading device loads the rock at the preset constant displacement speed, until the core of the rock is broken, and scanning grayscale image data of the internal structure of the rock under different load conditions, stress change data, and core axial displacement change data are obtained through the rock scanning device;

[0041] The data processing module is configured to perform data processing according to the stress change data and the core axial displacement change data to obtain a stress-strain correspondence of the rock.

[0042] In a possible design, the core model establishment module includes a segmentation module and a superposition module.

[0043] The segmentation module is configured to perform threshold segmentation processing according to the scanning grayscale image data and a pre-trained neural network to obtain a two-dimensional distribution image of the skeleton and the crack-void of the rock.

[0044] The superposition module is configured to perform superposition processing on the two-dimensional distribution image through a superposition algorithm to establish a three-dimensional digital core model of the rock under different load conditions.

[0045] In a possible design, the superposition algorithm is a moving stereoscopic algorithm, and the two-dimensional distribution image is an X-axis direction-Y-axis direction slice image.

[0046] Correspondingly, the superposition module is further configured to perform superposition on the X-axis direction-Y-axis direction slice image along the Z-axis direction through the moving stereoscopic algorithm to establish the three-dimensional digital core model of the rock under different load conditions.

[0047] In a possible design, the pre-trained neural network is obtained after model training using historical sample data, where the historical sample data is scanning grayscale sample image data of training sample rocks under different load conditions.

[0048] The core model establishment module further includes a model training module, and the model training module includes an extraction module and a cyclic training module.

[0049] The extraction module is configured to extract part of the image in the scanning grayscale sample image data as a training sample set in response to a selection operation of a user.

[0050] The cycle training module is configured to perform cycle model training on the rock pore, fracture and skeleton based on the full convolutional neural network and the training sample set until the accuracy and error rate of the neural network obtained through the last model training meet the preset training end condition, and then the neural network obtained through the last model training is taken as the pre-training neural network.

[0051] In a possible design, the apparatus further includes an acquisition change data module, a generation module, a constitutive relationship establishment module and a fitting calculation module.

[0052] The stress change data and the core axial displacement change data of the rock in the conversion process under different load conditions are acquired.

[0053] According to the stress change data and the core axial displacement change data, a core stress-strain relationship curve is generated.

[0054] According to the preset strain equivalent assumption, a deep rock damage deformation constitutive relationship suitable for strong stress influence is established.

[0055] According to the deep rock damage deformation constitutive relationship and the core stress-strain relationship curve, fitting calculation processing is performed to obtain the initial elastic modulus, the peak stress and the peak strain of the rock in the conversion process under different load conditions, so as to analyze the mechanics action mechanism.

[0056] In a possible design, in the stress value acquisition module, the preset constant displacement speed is 0.03 millimeters per minute to 3.0 millimeters per minute.

[0057] In a third aspect, the present application provides an electronic device, comprising a processor and a memory connected with the processor.

[0058] The memory stores computer execution instructions.

[0059] The processor executes the computer execution instructions stored in the memory, to implement the visualized characterization method of the rock microstructure deformation rule according to the first aspect.

[0060] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by the processor to implement the visualized characterization method of the rock microstructure deformation rule according to the first aspect.

[0061] In a fifth aspect, the present application provides a computer program product, comprising a computer program, and the computer program is executed by the processor to implement the visualized characterization method of the rock microstructure deformation rule according to the first aspect.

[0062] On the basis of the implementation manners of the above aspects, the application can be further combined to provide more implementation manners.

[0063] The application provides a visual characterization method, device and equipment for a rock microstructure deformation rule, which comprises the following steps: firstly, scanning gray map data of the internal structure of the rock under different load conditions is obtained; then, a three-dimensional digital core model of the rock under different load conditions is established according to the scanning gray map data and a pre-trained neural network; then, pore structure feature extraction processing is performed according to the three-dimensional digital core model of the rock, so that rock pore structure feature parameters are obtained, and the dynamic deformation rule of the rock micro-pore structure with effective stress loading is characterized according to the rock pore structure feature parameters; finally, crack morphology feature extraction processing is performed according to the three-dimensional digital core model of the rock, so that a rock crack morphology feature parameter system is obtained, and the rock internal crack morphology deformation evolution rule is characterized according to the rock crack morphology feature parameter system. The following technical effects are achieved: the three-dimensional digital core model of the rock under different load conditions is established by using the scanning gray map data of the internal structure of the rock under different load conditions through the pre-trained neural network. Then, the rock microstructure deformation rule is characterized according to the three-dimensional digital core model of the rock, so that the real rock micro-pore structure is accurately reflected, and the technical problems that the deep rock internal cracks, matrix and skeleton cannot be accurately segmented and the microstructure dynamic deformation cannot be accurately characterized are solved. BRIEF DESCRIPTION OF DRAWINGS

[0064] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0065] The drawings herein are incorporated into the specification and form part of the specification, show embodiments consistent with the present application, and together with the specification serve to explain the principles of the present application.

[0066] Figure 1 A flowchart of a visual characterization method for a rock microstructure deformation rule provided by the embodiments of the present application;

[0067] Figure 2 A device schematic diagram of an in-situ stress loading micro-focus CT scanning experiment provided by the embodiments of the present application;

[0068] Figure 3 A schematic diagram of a core stress and strain relationship curve provided by the embodiments of the present application;

[0069] Figure 4A three-dimensional image data body diagram of a core under different loading pressures provided by an embodiment of the present application;

[0070] Figure 5 A two-dimensional image data body diagram of a core under different loading pressures provided by an embodiment of the present application;

[0071] Figure 6 A fracture digital core model diagram under different loading pressures provided by an embodiment of the present application;

[0072] Figure 7 A curve diagram of a core internal fracture volume ratio change with stress provided by an embodiment of the present application;

[0073] Figure 8 A diagram of object topological connectivity under different Euler coefficients provided by an embodiment of the present application;

[0074] Figure 9 A curve diagram of a rock fracture morphology characteristic parameter change with stress provided by an embodiment of the present application;

[0075] Figure 10 A fracture spatial distribution diagram in a core provided by an embodiment of the present application;

[0076] Figure 11 A constitutive model fitting curve diagram of a core damage mechanics provided by an embodiment of the present application;

[0077] Figure 12 A structure diagram of a rock microstructure deformation law visualization device provided by an embodiment of the present application;

[0078] Figure 13 A structure diagram of an electronic device provided by an embodiment of the present application.

[0079] Reference signs:

[0080] 210 - a ray source; 220 - a stress loading device; 230 - a scintillator; 240 - a detector;

[0081] L1 - a curve connected by experimental data; L2 - a fitting curve;

[0082] 1210 - an acquisition module; 1220 - a core model establishing module; 1230 - a visualization representation module;

[0083] 1310 - a processor; 1320 - a memory; 1330 - a communication component; 1340 - a bus. DETAILED DESCRIPTION

[0084] The exemplary embodiments will be described in detail herein with reference to the attached drawings. In the following description, like reference numerals refer to like elements, unless the context clearly dictates otherwise. The following description is not meant to limit the application to all of the embodiments described herein. Rather, the following description is meant to provide examples of apparatus and methods consistent with some aspects of the application as detailed in the appended claims.

[0085] In the embodiments of the present application, the terms "first", "second", and the like are used to distinguish between similar or identical items or elements having substantially the same function and purpose. Those skilled in the art can understand that the terms "first", "second", and the like do not limit the quantity and execution order, and the terms "first", "second", and the like do not necessarily mean different. It should be noted that in the embodiments of the present application, the words "exemplary" or "for example" are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "for example" should not be construed as being preferred or superior over other embodiments or designs. Rather, the use of "exemplary" or "for example" is intended to present concepts in a concrete manner. In the embodiments of the present application, "at least one" means one or more, and "multiple" means two or more.

[0086] It should be noted that "at" in the embodiments of the present application can be at the moment when a certain condition occurs, or within a certain period of time after the occurrence of a certain condition, which is not specifically limited in the embodiments of the present application. In addition, the method for visualizing the deformation rule of rock microstructure provided in the embodiments of the present application is only an example, and the method for visualizing the deformation rule of rock microstructure can also include more or less content.

[0087] In order to clearly describe the technical solutions of the embodiments of the present application, the following briefly introduces some terms and technologies involved in the embodiments of the present application:

[0088] Gas permeability (gas permeability): refers to the ability of rock to allow fluid to pass through.

[0089] Dynamic load: refers to the load that changes over time.

[0090] Fracture spatial distribution: refers to the process of describing and analyzing the fracture system in rock or other geological materials in geology, especially in petroleum geology and rock mass mechanics.

[0091] Constitutive relation: refers to the relationship between stress and strain of a material under the action of external load or environmental conditions.

[0092] X-ray microfocus computed tomography (CT): a high-resolution three-dimensional imaging technique that uses an X-ray microfocus source for scanning to obtain detailed structural information inside the sample. Compared with traditional CT technology, X-ray microfocus CT, as an efficient non-destructive detection technology, has a resolution of sub-micron level. It can be used to obtain the internal fracture and matrix microstructure distribution of rocks, and can also be combined with an in-situ stress loading device to dynamically monitor and image the deformation of rock microstructure, thereby providing an effective way to visualize the deformation evolution law of deep reservoir rock microstructure.

[0093] Understanding the microstructure deformation law of reservoir rocks helps to accurately analyze the physical properties of the reservoir, such as pore radius distribution and mean value, throat radius distribution and mean value, pore throat ratio, coordination number, and tortuosity of rock pore structure characteristic parameters. By analyzing these rock pore structure characteristic parameters, the storage and flow capacity of oil and gas can be understood, which is of great significance for in-depth revealing the variation law of fluid flow in the reservoir, accurately analyzing the stability of the reservoir, and scientifically guiding the development of oil and gas resources

[0094] In the prior art, a small number of two-dimensional rock thin section images are usually used to reconstruct a three-dimensional digital core model of the rock by using numerical reconstruction methods such as simulated annealing, multi-point geostatistics, or Markov chain Monte Carlo, so as to realize the characterization of the rock micro-pore structure.

[0095] However, the reconstruction effect of these numerical reconstruction methods depends heavily on the resolution and image size of the two-dimensional rock thin section images. In actual applications, due to the limitations of these factors, numerical reconstruction methods are difficult to accurately restore the real rock micro-pore structure, especially in complex pore systems and strongly heterogeneous reservoirs, which limits the accuracy and reliability of subsequent related research.

[0096] Therefore, the embodiments of the present application propose a method, device and equipment for visualizing the deformation law of rock microstructure, which can be used in the field of oil and gas exploration, and aims to solve the above technical problems in the prior art. By pre-training a neural network model, using scanning gray scale data of the internal structure of the rock under different load conditions, a corresponding three-dimensional digital core model is constructed, and then the deformation law of the rock microstructure under different loads is accurately characterized based on the model, so as to more truly and accurately reflect the actual situation of the rock micro-pore structure.

[0097] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0098] Figure 1 This is a flowchart illustrating a method for visualizing the deformation patterns of rock microstructure, provided in an embodiment of this application. Figure 1 As shown, the method includes:

[0099] S101. Obtain scanned grayscale data of the internal structure of rocks under different load conditions.

[0100] In this embodiment of the application, the execution subject of a visualization characterization method for the deformation law of rock microstructure can be a server. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server, etc. No specific restrictions are made here.

[0101] Specifically, different load conditions refer to applying different stress conditions to rocks; here, the load refers to dynamic load. Rocks refer to rock samples from geological reservoirs, especially those from deep reservoirs. Scanning grayscale images can be CT grayscale images. Scanning grayscale image data refers to grayscale data obtained by performing tomographic scanning (such as CT scans) on rock samples under various effective stress loading conditions using high-resolution imaging equipment, reflecting the distribution of pores and fractures within the rock.

[0102] Furthermore, the process of acquiring grayscale scan data may include: acquiring the real-time stress value when the stress loading device applies stress to the rock at a preset constant displacement speed; when the real-time stress value is detected to reach a fixed stress value, the stress loading device pauses loading, and then the rock scanning device acquires grayscale scan data of the internal structure of the rock under the load condition of the fixed stress value; returning to the step of acquiring the real-time stress value when the stress loading device applies stress to the rock at a preset constant displacement speed, until the rock core fractures, and obtaining grayscale scan data of the internal structure of the rock under different load conditions, stress change data, and core axial displacement change data through the rock scanning device; and processing the stress change data and core axial displacement change data to obtain the stress-strain relationship of the rock.

[0103] The rock scanning device is a micro-focus CT device. Specifically, deep reservoir rocks are in a complex geological environment of high temperature, high pressure, and high ground stress. The evolution of the internal micro-pore structure and crack morphology of the deep reservoir rocks has an important influence on the migration, accumulation, and exploitation of oil and gas resources. The micro-focus CT scanning technology has the advantages of high resolution and non-destructive testing, and can clearly obtain the microstructure information inside the rock, which helps to deeply understand the deformation law and internal structure change of the deep reservoir rocks under different dynamic loads.

[0104] For the rock samples of typical deep reservoirs, first, a pre-experiment of in-situ stress loading can be performed at a constant displacement speed to understand the mechanical response of the rock under different stress states. According to the stress-strain relationship, appropriate stress points for CT scanning and the rock fracture pressure can be determined for subsequent in-situ stress loading micro-focus CT scanning experiments.

[0105] Specifically, the first rock sample can be placed in the in-situ stress loading experimental device, and an axial pressure can be applied to the first rock sample at a constant displacement speed, while the stress-strain relationship is monitored and recorded in real time. As the stress gradually increases, the first rock sample will undergo elastic deformation, plastic deformation, and eventually break. According to the recorded stress-strain curve, the deformation characteristics of the first rock sample at different stress stages are analyzed, and the mechanical parameters such as the elastic limit and yield strength of the first rock sample are determined. At the same time, the key points on the stress-strain curve are selected as the stress points for subsequent CT scanning, and the fracture pressure of the first rock sample is determined to provide a reference for the complete experiment.

[0106] Subsequently, an in-situ stress loading micro-focus CT scanning experiment is performed to scan the rock sample under in-situ stress loading at the determined stress points, and to obtain CT grayscale image data reflecting the deformation characteristics of the internal micro-pore structure and crack morphology of the deep reservoir rock under different dynamic loads, thereby deeply analyzing the evolution law of the internal structure of the rock.

[0107] Specifically, the second rock sample can be installed in the in-situ stress loading experimental device, and the position and state of the second rock sample can be ensured to be consistent with those of the first rock sample in the pre-experiment. The second rock sample and the first rock sample are taken from the same rock and have consistent initial characteristics. Then, different dynamic loads are applied to the second rock sample according to the stress points determined in the pre-experiment. At each stress point, the stress is kept stable for a period of time, and then the micro-focus CT scanning equipment is used to scan the second rock sample to obtain the CT grayscale image data inside the second rock sample under the stress state. During the scanning process, the stress value, scanning time, and other parameters of each stress point are accurately recorded to ensure the integrity and accuracy of the data.

[0108] A real rock sample from a deep oil and gas reservoir was used as the research object. The rock sample was cylindrical, with dimensions of 2.5 cm in diameter and 5 cm in height. The rock sample had a density of 2.56 g / cm³, a porosity of 4.46%, and a gas permeability of 1.48 × 10⁻⁶. -3 Square micrometers. Two smaller cylindrical rock samples (cores) can be drilled from the actual rock samples of this deep oil and gas reservoir, each with a diameter of 4 mm and a height of 8 mm. One cylindrical rock sample will be used for in-situ stress loading pre-experiment to determine the rock fracture pressure; the other cylindrical rock sample will be used for in-situ stress loading microfocus CT scanning experiment to obtain CT grayscale images of the internal fracture morphology of deep rocks under different dynamic loading conditions.

[0109] The apparatus for in-situ stress loading microfocus CT scanning experiments can employ a high-resolution three-dimensional X-ray microscope to facilitate non-destructive analysis of the internal structure of rock samples. Figure 2 A schematic diagram of the apparatus for in-situ stress loading microfocus CT scanning experiments provided in an embodiment of this application. Figure 2 As shown, this three-dimensional X-ray microfocus CT can employ a two-stage magnification method. The first stage uses geometric magnification (similar to conventional CT), while the second stage uses optical magnification (similar to synchrotron radiation source X-ray CT). The distance between the X-ray source 210 and the stress loading device 220 is d1, the distance between the X-ray source 210 and the scintillator 230 is d2, and the distance between the scintillator 230 and the detector 240 is d3. Geometric magnification is used between the X-ray source 210 and the scintillator 230, while optical magnification is used between the scintillator 230 and the detector 240. This two-stage magnification method reduces the reliance on geometric magnification, enabling high-resolution images of cylindrical rock samples even when they are far from the X-ray source. Meanwhile, the experimental setup can withstand a maximum allowable axial stress of 5000 Newtons (N), and for a cylindrical rock sample with a diameter of 4 mm, the maximum loading pressure can reach 398 Megapascals (MPa). This allows it to effectively meet the requirements for high-resolution three-dimensional imaging of the compression-internal fracture process of cylindrical rock samples under harsh working conditions, as well as dynamic monitoring of the stress-strain process. In the in-situ stress loading microfocus CT scanning experiment, the loading method can be controlled by a preset constant displacement velocity, ranging from 0.03 mm / min to 3.0 mm / min. Furthermore, the compression distance can be 15 mm to 5 mm between the top and bottom plates; the test temperature range is -20℃ to 160℃.

[0110] Furthermore, the specific steps for conducting in-situ stress-loading microfocus CT scanning experiments on the core are as follows:

[0111] ① For the cylindrical rock sample with a diameter of 4 mm and a length of 8 mm, the upper end surface and the lower end surface of the cylindrical rock sample are polished smooth by using an angle grinder.

[0112] ② The cylindrical rock sample is wrapped with a plastic wrap to prevent the rock from falling apart and causing the image of the fractured core to be unobservable during the in-situ stress loading microfocus CT scanning experiment, but the plastic wrap should not be wrapped too tightly to avoid changing the mechanical properties of the rock and introducing large errors.

[0113] ③ The wrapped cylindrical rock sample is placed in the rock clamping device of the CT equipment, the sealing cover is covered, and the screw is tightened.

[0114] ④ The front and rear positions of the X-ray source are adjusted to make the spatial resolution of the microfocus CT image meet the requirements. During the experiment, in order to observe the crack morphology and deformation law of the rock under different dynamic load conditions, the spatial resolution of the microfocus CT scanning can be set to 11.27 microns.

[0115] ⑤ Turn on the stress loading device to load stress at a preset constant displacement speed, observe the stress change state, pause loading when a fixed stress value is reached, and perform microfocus CT imaging under the condition of the fixed stress value.

[0116] ⑥ Scanning is performed using a cone-beam X-ray source with a cone angle of 5.11°. In each scanning step, an X-ray passes through the cylindrical rock sample to record a vertical projection, and then 1800 projections are obtained after 360° rotation at an interval angle of 0.2°, so that the CT grayscale image data of the cylindrical rock sample under the stress loading condition can be obtained.

[0117] ⑦ After CT scanning, continue to load stress at a preset constant displacement speed, repeat steps ⑤-⑥, and until the core is broken.

[0118] ⑧ After the in-situ stress loading is completed, the CT scanning images of the cylindrical rock sample at different stress nodes and the stress change and core axial displacement change during loading can be obtained, and the core stress-strain relationship curve can be obtained through further data processing.

[0119] Figure 3 The core stress-strain relationship curve provided by the embodiments of the present application is shown in the schematic diagram. As shown in Figure 3 , the graph in the middle position of Figure 3 is the core stress-strain relationship curve. In Figure 3In the middle of the figure, the porosity is 4.463% and the permeability is 1.4811 milliDarcy (mD). The horizontal axis represents rock strain in percentage (%), and the vertical axis represents the loading pressure in MPa. Each red dot represents a different scanning stage in the in-situ stress loading microfocus CT scanning experiment. Figure 3 The left and right images in the table are digital core models of fractures in a cylindrical rock sample at different scanning stages. Figure 3 As shown, the stress-strain relationship curve of the deep rock sample conforms to general laws. During the entire stress loading process, the compaction and linear elastic deformation stages of the deep rock sample take a relatively long time. Subsequently, under strong stress loading conditions, plastic nonlinear deformation occurs, inducing cracks and their propagation. When the axial stress exceeds the fracture pressure, the rock undergoes irreversible shear failure. Here, strong stress refers to a loading pressure greater than a fixed preset stress value, such as 18 MPa. Compared with the stress-strain relationship curve of the in-situ stress loading pre-experiment, in the in-situ stress loading microfocus CT scan experiment, the core stress-strain relationship curve at the end of each scan stage shows a slight downward shift. Analysis shows that the in-situ stress loading microfocus CT scan experiment adopted a constant-rate loading mode (e.g., a speed of 0.03 mm / min), maintaining a constant stress during microfocus CT scanning. After the scan, stress continues to be applied. During this period, the rock undergoes yield strain due to the load, resulting in a slight decrease in the actual stress compared to before the CT scan.

[0120] Figure 4 This is a schematic diagram of three-dimensional image data of rock cores under different loading pressures, provided in an embodiment of this application. Figure 4 In (a) to (f), the permeability is 1.48 mD. For example... Figure 4 As shown, Figure 4 (a) in the image is a three-dimensional image of the core when the loading pressure is equal to 0 MPa; Figure 4 (b) in the image is a three-dimensional image of the core when the loading pressure is equal to 17.4 MPa; Figure 4 (c) in the image is a three-dimensional image of the core when the loading pressure is equal to 25.8 MPa; Figure 4 (d) in the image is a three-dimensional image of the core when the loading pressure is equal to 35.1 MPa; Figure 4 (e) in the image is a three-dimensional image of the core when the loading pressure is equal to 48 MPa; Figure 4 (f) in the image is a three-dimensional image of the rock core when the rock sample fractured.

[0121] Figure 5 This is a schematic diagram of two-dimensional image data of rock cores under different loading pressures, provided in an embodiment of this application. Figure 5 In (a) to (f), the permeability is 1.48 mD. For example... Figure 5Shown, Figure 4 (a) to (f) in FIG. 8 are slice images of the rock sample in the X-axis direction-Y axis direction. Among them, Figure 5 (a) in FIG. 8 is a two-dimensional image of the core when the loading pressure is equal to 0MPa; Figure 5 (b) in FIG. 8 is a two-dimensional image of the core when the loading pressure is equal to 17.4MPa; Figure 5 (c) in FIG. 8 is a two-dimensional image of the core when the loading pressure is equal to 25.8MPa; Figure 5 (d) in FIG. 8 is a two-dimensional image of the core when the loading pressure is equal to 35.1MPa; Figure 5 (e) in FIG. 8 is a two-dimensional image of the core when the loading pressure is equal to 48MPa; Figure 5 (f) in FIG. 8 is a two-dimensional image of the core when the rock sample is broken.

[0122] S102, according to the scanning gray scale data and the pre-trained neural network, a three-dimensional digital core model of the rock under different load conditions is established.

[0123] Specifically, the server can input the obtained scanning gray scale data into the pre-trained deep neural network model, thereby generating a three-dimensional digital core model of the rock corresponding to each load condition, to accurately reconstruct the spatial distribution characteristics of the microstructure inside the rock.

[0124] The pre-trained neural network is obtained after model training using historical sample data, wherein the historical sample data is scanning gray scale sample image data of the training sample rock under different load conditions. Further, the model training process of the pre-trained neural network includes: first, in response to the selection operation of the user, part of the image in the scanning gray scale sample image data is extracted from the historical sample data as a training sample set; then, based on the full convolutional neural network and the training sample set, the model is trained in cycles for the rock pores, cracks and skeletons until the accuracy and error rate of the neural network obtained by the last model training meet the preset training end condition, then the neural network obtained by the last model training is taken as the pre-trained neural network.

[0125] Specifically, the server can extract representative image data from the scanning gray scale sample image data under different dynamic load conditions as a training sample set according to the selection operation of the user. Then, based on the full convolutional neural network, the cycle training accuracy and error rate are taken as the training target, a deep learning training model of the rock pores, cracks and skeletons is established, the rock pores, cracks and skeletons under different stress loading stages are subjected to multi-value semantic segmentation, and a three-dimensional digital core model of the rock under different stress loading stages is established based on the superposition algorithm.

[0126] Figure 6 The different loading pressure under the fracture digital core model provided by the embodiment of the present application is shown in the figure. AsFigure 6 (a) in FIG. 1 is a three-dimensional image of the core under a loading pressure equal to 0 MPa; Figure 6 (b) in FIG. 1 is a three-dimensional image of the core under a loading pressure equal to 17.34 MPa; Figure 6 (c) in FIG. 1 is a three-dimensional image of the core under a loading pressure equal to 26.01 MPa; Figure 6 (d) in FIG. 1 is a three-dimensional image of the core under a loading pressure equal to 34.68 MPa; Figure 6 (e) in FIG. 1 is a three-dimensional image of the core under a loading pressure equal to 47.69 MPa; Figure 6 (f) in FIG. 1 is a three-dimensional image of the core when the rock sample is broken. Figure 6 (g) in FIG. 2 is an image of the three-dimensional fracture digital core model under a loading pressure equal to 0 MPa; Figure 6 (b) in FIG. 2 is an image of the three-dimensional fracture digital core model under a loading pressure equal to 17.34 MPa; Figure 6 (c) in FIG. 2 is an image of the three-dimensional fracture digital core model under a loading pressure equal to 26.01 MPa; Figure 6 (d) in FIG. 2 is an image of the three-dimensional fracture digital core model under a loading pressure equal to 34.68 MPa; Figure 6 (e) in FIG. 2 is an image of the three-dimensional fracture digital core model under a loading pressure equal to 47.69 MPa; Figure 6 (f) in FIG. 2 is an image of the three-dimensional fracture digital core model when the rock sample is broken. Figure 6 Further, the process of establishing the three-dimensional digital core model of the rock includes: first, performing threshold segmentation processing according to the scanning grayscale image data and the pre-trained neural network to obtain a two-dimensional distribution image of the skeleton and the fracture-void of the rock; then, performing superposition processing on the two-dimensional distribution image through a superposition algorithm to establish the three-dimensional digital core model of the rock under different loading conditions. The superposition algorithm is a moving cubic algorithm, and the two-dimensional distribution image is an X-axis direction-Y-axis direction slice image. Therefore, the X-axis direction-Y-axis direction slice image can be superimposed along the Z-axis direction through the moving cubic algorithm to establish the three-dimensional digital core model of the rock under different loading conditions.

[0127] Specifically, after threshold segmentation, the two-dimensional distribution image of the skeleton and the fracture-void of the rock can be obtained, and all X-axis direction-Y-axis direction slice images can be superimposed along the Z-axis direction through a superposition algorithm such as the moving cubic algorithm to establish the three-dimensional digital core model of the rock under different stress loading stages.

[0128]

[0129] ​S103, perform a pore structure feature extraction process according to the rock three-dimensional digital core model, obtain a rock pore structure feature parameter, and represent a dynamic deformation rule of the rock micro-pore structure with effective stress loading according to the rock pore structure feature parameter.

[0130] Specifically, the rock pore structure feature parameter includes a pore radius distribution and mean value, a throat radius distribution and mean value, a pore-throat ratio, a coordination number, and a tortuosity, and the like. Specifically, the server can extract a rock spatial topology network model at different stress loading stages according to the rock three-dimensional digital core model based on topology geometry, analyze the rock pore structure feature parameter, and represent the dynamic deformation rule of the rock micro-pore structure with effective stress loading. Figure 7 As shown in FIG. 6, there are multi-scale cracks in the core, a large number of micro-pores and a small number of cracks exist in the initial state, with the gradual increase of the loading pressure, the effective stress loading, the micro-pores are gradually compacted, the original small number of cracks gradually expand into new cracks, and finally converge to form a complex crack network.

[0131] S104, perform a fracture morphology feature extraction process according to the rock three-dimensional digital core model, obtain a rock fracture morphology feature parameter system, and represent a rock internal fracture morphology deformation evolution rule according to the rock fracture morphology feature parameter system.

[0132] Specifically, the rock fracture morphology feature parameter system mainly includes a fracture volume, an opening, a connectivity rate, a shape factor, a fractal dimension, an Euler coefficient, and a tortuosity, and the like. Specifically, the server can represent the internal fracture morphology deformation evolution rule of the deep reservoir rock under the influence of strong stress based on the integral geometry theory according to the rock fracture morphology feature parameter system.

[0133] Further, the fracture volume refers to the volume occupied by the cracks in the core, the larger the fracture volume, the higher the rock permeability and the better the connectivity. In the established crack digital core model, the resolution of the image is the length of each voxel unit of the digital core, and therefore the sum of the volumes of all voxels is the size of the volume occupied by the cracks. The calculation formula of the fracture volume is as follows:

[0134]

[0135] wherein, is the fracture volume, is the length of the voxel unit, and N is the number of voxels in the crack digital core.

[0136] Figure 7 is a curve diagram provided by the present embodiment, which shows the change of the proportion of the fracture volume in the core with stress. Figure 7 The core permeability in FIG. 6 is 1.48 mD, as shown in FIG. 7. Figure 8As shown, in the initial state without stress loading, only a small amount of micro-fractures developed inside the core, and the fracture volume ratio was 0.127%; with the continuous loading of effective stress, the micro-fractures generated inside the core gradually expanded, the fracture volume increased, and the fracture ratio grew slowly before the rock broke; when the effective stress was greater than the rock breaking pressure, the micro-fractures converged and the fracture opening became larger, and the fracture volume ratio increased sharply to 2.942%. Therefore, the fracture volume can effectively represent the deformation evolution law of the fracture morphology inside the rock.

[0137] The shape factor is a dimensionless parameter used to characterize the fracture shape in the core, which is used to describe the regularity of the fracture. The closer the shape factor is to 1, the closer the fracture shape is to a sphere, and the higher the value indicates the more irregular the shape. The calculation formula of the shape factor is as follows:

[0138]

[0139] Where F is the shape factor, A is the surface area of the fracture, and V is the volume of the fracture.

[0140] Fractal dimension is a parameter used to describe the complexity of the fracture geometry, which can describe the roughness of the three-dimensional fracture geometry. The fractal dimension is generally between 1.5-2, and the larger the fractal dimension, the more complex the fracture geometry. The three-dimensional box counting method is usually used to calculate the fractal dimension of the fracture geometry, and the specific calculation steps are as follows:

[0141] ① Box covering: Place the fracture digital core in a smallest three-dimensional box that contains it. Then, divide the box into smaller sub-boxes.

[0142] ② Count the number of boxes: Count the number of boxes that cover the core at each scale. The scale can be adjusted by reducing the size of the sub-boxes.

[0143] ③ Scale and box number relationship: Record the number of boxes required to cover the core at different scales. Represented as a function of scale N(m), where m represents the size of the box.

[0144] ④ Fractal dimension calculation: Calculate the fractal dimension of the object based on the logarithmic function relationship. The commonly used logarithmic relationship can be described as:

[0145]

[0146] Where n is the fractal dimension.

[0147] ⑤ Linear regression: Take the logarithm of N(m) and m, and then get the slope by linear regression, which is the fractal dimension.

[0148] Euler coefficient is used to describe the complex spatial topological connectivity of the fracture. For three-dimensional fractures, the calculation formula of Euler coefficient is as follows:

[0149]

[0150] Where χ is the Euler coefficient. Let E be the number of vertices, E be the number of edges, and F be the number of faces. Figure 8 This is a schematic diagram illustrating the topological connectivity of objects under different Euler coefficients, as provided in embodiments of this application. Figure 8 As shown, Figure 8 In (a), the Euler coefficient is 1; Figure 8 In (b), the Euler coefficient is 2; Figure 8 The Euler coefficient in (c) is 0; Figure 8 The (d) Euler coefficient is -1; Figure 8 The Euler coefficient (e) in the equation is -4. Figure 9 It can be seen that a positive Euler coefficient indicates poor connectivity of objects, while a negative Euler coefficient indicates good connectivity of objects.

[0151] Tortuosity refers to the ratio of the actual flow path of the core to the overall length of the core, reflecting the connectivity of the core. In actual calculations, the outline length of the three-dimensional digital core can be obtained first, then the shortest path can be selected from the paths with the same endpoint, and the average value of all shortest paths can be calculated, which is the tortuosity.

[0152] Figure 9 This is a schematic diagram showing the variation of rock fracture morphology parameters with stress, as provided in an embodiment of this application. Figure 9 As shown, Figure 9 (a) in the figure is a schematic diagram of the shape factor of the rock core as a function of stress; Figure 9 (b) in the figure is a schematic diagram of the fractal dimension of the rock core as a function of stress; Figure 9 (c) in the figure is a schematic diagram of the Euler coefficient of the core as a function of stress; Figure 9 (d) in the figure is a schematic diagram of the curve of the core tortuosity as a function of stress.

[0153] Specifically, such as Figure 9 As shown in (a), under low-stress loading conditions, the closure of internal fractures and the rearrangement of rock particles increase the shape factor. As the effective stress gradually increases, the expansion or contraction of internal fractures becomes more drastic, leading to fluctuations in the shape factor and consequently, a gradual increase in the fractal dimension of the fractures. Figure 9 As shown in (b) in the figure. This is due to the increased complexity of the rock's microstructure caused by plastic nonlinear deformation and crack initiation and propagation.

[0154] like Figure 9As shown in (c) and (d), in the initial stage of stress loading, the core undergoes linear elastic deformation, resulting in slight changes in the Euler coefficient and tortuosity of the cracks. As the effective stress is gradually applied, the core is affected by plastic nonlinear deformation, which leads to the generation of microcracks and greater displacement between particles, resulting in a rapid increase in the Euler coefficient and a significant increase in tortuosity.

[0155] like Figure 10 As shown in (a) to (d), when the effective stress is greater than the fracture pressure, the microcracks inside the rock further expand and converge to form a crack network. The complexity of the crack structure increases and the topological connectivity is enhanced. Specifically, this is manifested in the rapid decrease of the crack shape factor, the increase of the fractal dimension, and the sharp decrease of the Euler coefficient and tortuosity.

[0156] Figure 10 This is a schematic diagram of the spatial distribution of fractures inside a rock core, provided as an embodiment of this application. Figure 10 As shown, by further extracting the distribution of fractures inside the core and measuring parameters such as fracture aperture, length, and dip angle, the types of three main fractures were classified: fracture F1 is a vertical central fracture; fracture F2 is a vertical large fracture; and fracture F3 is a low-angle central fracture. A localized fracture zone CR1 was also found inside the core. Figure 10 This reflects the spatial distribution of the three main fractures and fractured zones inside the core.

[0157] Table 1 Characteristic parameters of the main fracture inside the core.

[0158]

[0159] Table 1 provides the following information: Figure 11 The characteristic parameters corresponding to the three main fractures are shown in Table 1. These three main fractures have large openings, all exceeding 20 micrometers, classifying them as medium or large fractures. Fractures F1 and F2 are both relatively long, penetrating the entire core, and both have large dip angles, being vertical fractures. However, fracture F3 has a smaller dip angle, classifying it as a low-angle fracture. Analysis shows that fractures F1 and F2 are along the direction of axial stress. When axial load is applied, the core experiences shear slip, causing the axial fracture opening to gradually increase, forming interlocking fractures. Fracture F3, due to its lower dip angle, experiences a gradual decrease in opening due to increased overburden pressure when subjected to strong stress, and the overburden pressure limits its further extension. Therefore, fracture F3 has a shorter overall length and a smaller opening.

[0160] In addition, in an optional embodiment of this application, characterizing the evolution law of internal rock crack morphology deformation based on the rock crack morphology characteristic parameter system can be a process of visualizing the rock crack morphology characteristic parameter system in the form of a graph or a data list.

[0161] Further, by acquiring stress change data and core axial displacement change data of the rock in the conversion process under different load conditions; then, according to the stress change data and the core axial displacement change data, a core stress and strain relationship curve is generated; then, according to a preset strain equivalent assumption, a deep rock damage deformation constitutive relationship suitable for strong stress influence is established; finally, according to the deep rock damage deformation constitutive relationship and the core stress and strain relationship curve, fitting calculation processing is performed, so that the initial elastic modulus, peak stress and peak strain of the rock in the conversion process under different load conditions can be obtained, thereby facilitating the analysis of the mechanical action mechanism.

[0162] Specifically, according to the strain equivalent assumption, a deep rock damage deformation constitutive relationship suitable for strong stress influence can be established, and by fitting the core stress and strain relationship curve in the in-situ stress loading micro-focus CT scanning experiment, the initial elastic modulus, peak stress and peak strain in the stress loading process can be calculated, so that different mechanical response stages can be divided in the core stress and strain relationship curve, thereby facilitating the analysis of the mechanical action mechanism of the microstructure deformation law of the deep rock under strong stress influence.

[0163] According to the strain equivalent assumption, the evolution equation of the damage variable can be introduced into the constitutive relationship to establish a deep rock damage deformation constitutive relationship suitable for strong stress influence. The constitutive relationship under the strain equivalent assumption of the elastic-brittle material can be expressed as:

[0164]

[0165] Further, the evolution equation of the damage variable D can be expressed as a function of the strain, which is as follows:

[0166]

[0167] By substituting formula (6) into formula (5), the following formula is obtained:

[0168]

[0169] The formula 7 is simplified as the relationship between stress and strain, and the following formula is obtained:

[0170]

[0171] Formula (8) is the basic equation of the stress and strain relationship of the deep rock, and the three coefficients A, B and C in the basic equation are obtained, that is, the stress and strain constitutive relationship of the deep rock is determined.

[0172] For formula (8), the parameters A, B, C therein can be determined by the method of undetermined coefficients. First, the derivative of formula (8) is obtained, and the expression of the tangent modulus about the strain is obtained:

[0173]

[0174] At the initial moment of stress loading, the strain is 0, and formula (9) is obtained by substituting formula (9) into formula (8):

[0175]

[0176] is the initial elastic modulus.

[0177] At the peak point of stress loading, the stress reaches the maximum value , and the strain at this time is the peak strain . Substituting formula (8) into formula (8) obtains:

[0178]

[0179] is the peak stress, is the peak strain. The peak point is the maximum point of the stress-strain curve, and at this time, the slope of the curve is 0, so the value of formula (9) is 0, that is:

[0180]

[0181] By combining formula (10), formula (11) and formula (12), the values of the three coefficients A, B and C can be obtained, and the specific values are as follows:

[0182]

[0183] Substituting formula (13) into formula (8) obtains the constitutive relationship of the damage deformation of deep rock as follows:

[0184]

[0185] is the initial elastic modulus, is the peak stress, is the peak strain. When the initial elastic modulus is obtained, the slope of the stress-strain curve at the initial moment can be obtained by selecting the continuous stress points in a short time at the beginning of stress loading, dividing the change amount of the stress by the change amount of the strain, and the initial elastic modulus is obtained.

[0186] Figure 11 is the fitting curve diagram of the constitutive model of the damage mechanics of the core provided by the embodiment of the application. In Figure 11 ​​​In the figure, the permeability is 1.4811 mD, L1 is the curve connected by experimental data, and L2 is the fitting curve. Figure 11 In the figure, the damage deformation characteristics of the core during the stress loading process can be obtained. Figure 12 As shown in the figure, the core is compacted at the initial stage of stress loading due to the presence of micro-pores and micro-cracks. Since the core has fewer micro-pores and micro-cracks and is composed of larger gravel, the compacting and linear elastic deformation stages last for a long time during the entire stress loading process. Then, the micro-pores and micro-cracks reopen and expand under high stress conditions. Due to frictional sliding, the micro-cracks expand and converge, ultimately leading to the rupture of the core.

[0187] The method provided by the embodiment comprises the following steps: first, obtaining scanning gray scale data of the internal structure of the rock under different load conditions; then, establishing a three-dimensional digital core model of the rock under different load conditions according to the scanning gray scale data and a pre-trained neural network; then, performing pore structure feature extraction processing according to the three-dimensional digital core model of the rock to obtain rock pore structure feature parameters, and representing the dynamic deformation law of the rock micro-pore structure with effective stress loading according to the rock pore structure feature parameters; finally, performing crack morphology feature extraction processing according to the three-dimensional digital core model of the rock to obtain a rock crack morphology feature parameter system, and representing the deformation evolution law of the internal crack morphology of the rock according to the rock crack morphology feature parameter system.

[0188] The following technical effects are achieved: by using the scanning gray scale data of the internal structure of the rock under different load conditions, a three-dimensional digital core model of the rock under different load conditions is established by means of the pre-trained neural network. Then, the rock microstructure deformation law is represented according to the three-dimensional digital core model of the rock, which accurately reflects the real rock micro-pore structure, and solves the technical problems that the deep internal cracks, matrix and skeleton of the rock cannot be accurately segmented, and the dynamic deformation of the microstructure is difficult to accurately represent.

[0189] The embodiment of the present application can divide the function modules of the electronic device or the host device according to the above method examples. For example, each function module can be divided according to each function, or two or more functions can be integrated in one processing unit. The integrated unit can be realized in the form of hardware or in the form of a software function module. It should be noted that the division of the modules in the embodiment of the present application is illustrative, and is only a logical function division. When actually implemented, another division method can be used.

[0190] Figure 12 A structural schematic diagram of a rock microstructure deformation law visualization representation device provided by the embodiment of the present application is shown in Figure 13As shown, the device comprises: an acquisition module 1210; a core model establishing module 1220 and a visual representation module 1230.

[0191] The acquisition module 1210 is configured to acquire scanning grayscale data of the internal structure of the rock under different load conditions.

[0192] The core model establishing module 1220 is configured to establish a three-dimensional digital core model of the rock under different load conditions according to the scanning grayscale data and a pre-trained neural network.

[0193] The visual representation module 1230 is configured to perform pore structure feature extraction processing according to the three-dimensional digital core model of the rock, to obtain rock pore structure feature parameters, and to represent the dynamic deformation law of the microscopic pore structure of the rock with effective stress loading according to the rock pore structure feature parameters.

[0194] The visual representation module 1230 is configured to perform fracture morphology feature extraction processing according to the three-dimensional digital core model of the rock, to obtain a rock fracture morphology feature parameter system, and to represent the deformation evolution law of the internal fracture morphology of the rock according to the rock fracture morphology feature parameter system.

[0195] In a possible design, the acquisition module 1210 comprises: an acquisition stress value module, a data collection module, a cyclic pressurization module and a data processing module.

[0196] The acquisition stress value module is configured to acquire real-time stress values when the stress loading device loads the rock at a preset constant displacement speed.

[0197] The data collection module is configured to, when the stress loading device is detected to pause loading when the real-time stress value reaches a fixed stress value, collect scanning grayscale data of the internal structure of the rock under the load condition of the fixed stress value through a rock scanning device.

[0198] The cyclic pressurization module is configured to return to the step of acquiring real-time stress values when the stress loading device loads the rock at a preset constant displacement speed, until the core of the rock is broken, and to obtain scanning grayscale data of the internal structure of the rock, stress change data and core axial displacement change data under different load conditions through a rock scanning device.

[0199] The data processing module is configured to perform data processing according to the stress change data and the core axial displacement change data, to obtain a stress-strain correspondence of the rock.

[0200] In a possible design, the core model establishing module 1220 comprises: a segmentation module and a superposition module.

[0201] The segmentation module is configured to perform threshold segmentation processing on the scanned grayscale image data and the pre-trained neural network to obtain a skeleton of the rock and a two-dimensional distribution image of cracks and voids.

[0202] The superposition module is configured to perform superposition processing on the two-dimensional distribution image by using a superposition algorithm to establish a three-dimensional digital core model of the rock under different load conditions.

[0203] In a possible design, the superposition algorithm is a moving stereoscopic algorithm, and the two-dimensional distribution image is an X-axis direction-Y-axis direction slice image.

[0204] Correspondingly, the superposition module is further configured to perform superposition on the X-axis direction-Y-axis direction slice image along the Z-axis direction by using the moving stereoscopic algorithm to establish the three-dimensional digital core model of the rock under different load conditions.

[0205] In a possible design, the pre-trained neural network is obtained by using historical sample data to train a model, where the historical sample data is scanned grayscale sample image data of training sample rocks under different load conditions.

[0206] The core model establishment module 1220 further includes a model training module, and the model training module includes an extraction module and a loop training module.

[0207] The extraction module is configured to extract part of the images in the scanned grayscale sample image data as a training sample set from the historical sample data in response to a selection operation of a user.

[0208] The loop training module is configured to perform loop model training on the rock pores, cracks and skeleton based on the full convolutional neural network and the training sample set, until the accuracy and error rate of the neural network obtained by the last model training meet a preset training end condition, and then the neural network obtained by the last model training is taken as the pre-trained neural network.

[0209] In a possible design, the apparatus further includes an acquisition change data module, a generation module, a constitutive relationship establishment module and a fitting calculation module.

[0210] The stress change data and the core axial displacement change data of the rock in the conversion process under different load conditions are acquired.

[0211] The core stress-strain relationship curve is generated according to the stress change data and the core axial displacement change data.

[0212] The deep rock damage deformation constitutive relationship suitable for strong stress influence is established according to the preset strain equivalent assumption.

[0213] According to the deep rock damage deformation constitutive relation and the core stress and strain relation curve fitting calculation processing, the initial elastic modulus, the peak stress and the peak strain of the rock in the conversion process under different load conditions are obtained to analyze the mechanics mechanism.

[0214] In one possible design, in the stress value acquisition module, the preset constant displacement speed is 0.03 millimeters per minute to 3.0 millimeters per minute.

[0215] The embodiment provides a device for visualizing rock microstructure deformation law, which can execute the method for visualizing rock microstructure deformation law, has similar implementation principles and technical effects, and details are not described herein again.

[0216] In the specific implementation of the device for visualizing rock microstructure deformation law, each module can be implemented as a processor, and the processor can execute computer execution instructions stored in the memory, so that the processor executes the method for visualizing rock microstructure deformation law.

[0217] Figure 13 The electronic device provided in the embodiment of the present application is shown in a structural schematic diagram. As shown in the figure, the electronic device includes at least one processor 1310 and a memory 1320. The electronic device also includes a communication component 1330. The processor 1310, the memory 1320, and the communication component 1330 are connected through a bus 1340. ​

[0218] In the specific implementation process, the at least one processor 1310 executes computer execution instructions stored in the memory 1320, so that the at least one processor 1310 executes the method for visualizing rock microstructure deformation law as executed by the electronic device side.

[0219] The specific implementation process of the processor 1310 can refer to the method embodiments described above, which has similar implementation principles and technical effects, and details are not described herein again.

[0220] ​In the above embodiments, it should be understood that the processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the method disclosed in combination with the application can be directly embodied as hardware processor execution, or executed by a combination of hardware and software modules in the processor.

[0221] The memory can include a high-speed RAM memory, and can also include a non-volatile storage NVM, such as at least one disk memory.

[0222] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, the bus in the drawings of the present application does not limit to only one bus or one type of bus.

[0223] The functions realized by the electronic device and the master device described above are introduced for the scheme provided by the embodiments of the present application. It can be understood that the electronic device or the master device contains the hardware structure and / or software module corresponding to the execution of each function in order to realize the above functions. The units and algorithm steps of each example described in combination with the embodiments disclosed in the embodiments of the present application can be realized in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed by hardware or computer software driven hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the technical solution of the embodiments of the present application.

[0224] The present application also provides a computer readable storage medium, the computer readable storage medium stores computer execution instructions, when the processor executes the computer execution instructions, for realizing the above rock microstructure deformation rule visualization characterization method.

[0225] The aforementioned readable storage medium can be realized by any type of volatile or nonvolatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0226] An exemplary readable storage medium is coupled to the processor, so that the processor can read information from the readable storage medium, and can write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in an electronic device or a host device.

[0227] The present application also provides a computer program product, the computer program product comprising a computer program stored in a readable storage medium, at least one processor of an electronic device can read the computer program from the readable storage medium, and the at least one processor executes the computer program to enable the electronic device to perform the scheme provided by the above-mentioned embodiments.

[0228] Those skilled in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware. The aforementioned program can be stored in a computer readable storage medium. The program, when executed, executes steps including the above-mentioned method embodiments; and the aforementioned storage medium includes ROM, RAM, magnetic disk or optical disk and various storage medium that can store program codes.

[0229] So far, the technical scheme of the present application has been described in combination with the preferred embodiments shown in the drawings, but those skilled in the art can easily understand that the protection scope of the present application is obviously not limited to these specific embodiments, and the above embodiments are only used to illustrate the technical scheme of the present application, but not to limit it; although the present application has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that they can still modify the technical scheme recorded in the above-mentioned embodiments, or make equivalent replacement to part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical scheme deviate from the scope of the technical scheme of the embodiments of the present application.

Claims

1. A method for visualizing the deformation law of the microstructure of a rock, characterized in that, The method comprises the following steps: acquiring scanning grayscale data of internal structure of the rock under different load conditions; establishing a three-dimensional digital core model of the rock under different load conditions according to the scanning grayscale data and a pre-trained neural network; performing pore structure feature extraction processing according to the three-dimensional digital core model of the rock to obtain rock pore structure feature parameters, and representing a dynamic deformation rule of the rock micro-pore structure with effective stress loading according to the rock pore structure feature parameters; performing crack morphology feature extraction processing according to the three-dimensional digital core model of the rock to obtain a rock crack morphology feature parameter system, and representing a rock internal crack morphology deformation evolution rule according to the rock crack morphology feature parameter system.

2. The method of claim 1, wherein, The step of acquiring scanning grayscale data of internal structure of the rock under different load conditions comprises the following steps: acquiring real-time stress values when a stress loading device loads the rock at a preset constant displacement speed; suspending loading of the stress loading device when it is detected that the real-time stress values reach a fixed stress value, and then acquiring scanning grayscale data of internal structure of the rock under a load condition of the fixed stress value by a rock scanning device; returning to the step of acquiring real-time stress values when the stress loading device loads the rock at a preset constant displacement speed until the core of the rock is broken, and then acquiring scanning grayscale data of internal structure of the rock, stress change data and core axial displacement change data of the rock under different load conditions by the rock scanning device; performing data processing according to the stress change data and the core axial displacement change data to obtain a stress-strain corresponding relationship of the rock.

3. The method of claim 1, wherein, The step of establishing a three-dimensional digital core model of the rock under different load conditions according to the scanning grayscale data and a pre-trained neural network comprises the following steps: performing threshold segmentation processing according to the scanning grayscale data and the pre-trained neural network to obtain a two-dimensional distribution image of the skeleton and crack-void of the rock; performing superposition processing on the two-dimensional distribution image by a superposition algorithm to establish a three-dimensional digital core model of the rock under different load conditions.

4. The method of claim 3, wherein, The superposition algorithm is a moving stereoscopic algorithm, and the two-dimensional distribution image is an X-axis direction-Y-axis direction slice image. Correspondingly, the step of performing superposition processing on the two-dimensional distribution image by a superposition algorithm to establish a three-dimensional digital core model of the rock under different load conditions comprises the following steps: performing superposition on the X-axis direction-Y-axis direction slice image along the Z-axis direction by the moving stereoscopic algorithm to establish a three-dimensional digital core model of the rock under different load conditions.

5. The method according to any one of claims 1 to 4, characterized in that, The pre-trained neural network is obtained after model training using historical sample data, wherein the historical sample data is scanning grayscale sample data of training sample rocks under different load conditions. The model training process of the pre-trained neural network comprises the following steps: extracting part of the images in the scanning grayscale sample data as a training sample set in response to a selection operation of a user; The full convolutional neural network and the training sample set are used to perform cycle model training on rock pores, cracks and skeletons until the accuracy and error rate of the neural network obtained through the last model training meet preset training end conditions, and the neural network obtained through the last model training is taken as a pre-training neural network.

6. The method of claim 1, wherein, Also includes: Obtain the stress change data and core axial displacement change data of the rock during the conversion process under different load conditions; According to the stress change data and core axial displacement change data, generate a core stress-strain relationship curve; According to the preset strain equivalent assumption, establish a deep rock damage deformation constitutive relationship suitable for strong stress influence; According to the deep rock damage deformation constitutive relationship and the core stress-strain relationship curve, perform fitting calculation processing to obtain the initial elastic modulus, peak stress and peak strain of the rock during the conversion process under different load conditions, and analyze the mechanics action mechanism.

7. The method of claim 2, wherein, The preset constant displacement speed is 0.03mm / min to 3.0mm / min.

8. A device for visualizing the deformation law of the microstructure of a rock, characterized in that it comprises: Includes: The acquisition module is used to acquire scanning gray scale data of the internal structure of the rock under different load conditions; The core model establishment module is used to establish a three-dimensional digital core model of the rock under different load conditions according to the scanning gray scale data and the pre-training neural network; The visualization representation module is used to perform pore structure feature extraction processing on the three-dimensional digital core model of the rock, obtain rock pore structure feature parameters, and represent the dynamic deformation law of rock micro-pore structure with effective stress loading according to the rock pore structure feature parameters; The visualization representation module is used to perform crack morphology feature extraction processing on the three-dimensional digital core model of the rock, obtain a rock crack morphology feature parameter system, and represent the rock internal crack morphology deformation evolution law according to the rock crack morphology feature parameter system.

9. An electronic device, comprising: Includes: Memory, processor; The memory stores computer execution instructions; The processor executes the computer execution instructions stored in the memory, so that the processor executes the method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by the processor to implement the method of any one of claims 1-7.

Images