A numerical simulation method for triaxial compression of heterogeneous bedding rocks

Abstract

This invention relates to a triaxial compression numerical simulation method for heterogeneous layered rocks, comprising the following steps: Step S1, establishing a rock numerical model; Step S2, establishing models for different bedding angles; Step S3, generating contact between particles; Step S4, generating contact for different groupings; Step S5, generating random number parameters following a Weiper distribution; Step S6, using the micromechanical parameters following a Weiper distribution generated in Step S5, generating parameter models following a Weiper distribution for the rock matrix-rock matrix contact, bedding-bedding contact, rock-bedding contact, rock-wall contact, and bedding-wall contact in Step S4, respectively; Step S7, simulating failure of the parameter models established in Step S6. This method combines flexibility and scalability, allowing for more refined simulations by adjusting the particle shape and distribution parameters when dealing with different types of rocks or different loading conditions.

Description

Technical Field

[0001] This invention belongs to the fields of geotechnical engineering and computational mechanics, specifically relating to a triaxial compression numerical simulation method for heterogeneous layered rocks. Background Technology

[0002] Rock is a highly discontinuous and complex natural geological material, composed of numerous widely distributed natural or stress-induced defects, including weak interfaces, microcracks, and pores. In-depth research on these defects is of significant practical importance for promoting the development of rock mechanics and ensuring the safety and long-term stability of engineering structures. Numerous laboratory test results and experimental observations demonstrate that the heterogeneity of rocks is largely controlled by these defects, thereby causing complex deformation responses associated with microfracture modes. Therefore, heterogeneity is an important factor to consider when studying the micromechanical behavior of rocks.

[0003] Currently, there are several methods for establishing heterogeneous rock models, including using Thiessen polygons to generate irregular polygonal particles, controlling the distribution of polygonal shapes to create rocks with different homogeneity, and generating particles with random radii to generate rocks with different homogeneity. However, all of these methods are based on the physical morphology and distribution of rock particles, and the same mechanical parameters are still used to assign values ​​to the mechanical properties of rock particles. The heterogeneity of rocks depends not only on particle morphology and size, but also on the mechanical properties of the rock's microstructure. Therefore, considering only the heterogeneity of the rock particle structure cannot accurately simulate the heterogeneity of rocks, and the simulated stress-strain curves and failure characteristics differ significantly from actual experiments. Secondly, in establishing models of bedding rocks, smooth joint models are often used to simulate rock bedding. However, most experimental results show that smooth joint models are poor at simulating rock bedding.

[0004] To address the above limitations, the following technical issues need to be resolved: 1) When establishing numerical models for heterogeneous rocks, in addition to considering the heterogeneity of particle distribution and morphology, the heterogeneity of microscopic particle mechanical parameters also needs to be considered. 2) When establishing numerical models for bedding rocks, a bedding mechanics model more suitable for real physical experiments needs to be found. Therefore, a triaxial compression numerical simulation method for heterogeneous bedding rocks is urgently needed at this stage. Summary of the Invention

[0005] To address the aforementioned shortcomings in existing technologies, this invention provides a triaxial compression numerical simulation method for heterogeneous bedding rocks, comprising:

[0006] An improved method for triaxial compression numerical simulation of heterogeneous layered rocks, comprising the following steps:

[0007] Step S1: Establish a numerical model of the rock;

[0008] Step S2: Generate bedding with thickness a, spacing h, and dip angle α°, and establish models for different bedding dip angles; construct a three-dimensional rectangular coordinate system xyz with the center of the rock numerical model as the origin, where the x-axis is the horizontal axis, the y-axis is the vertical axis, and the z-axis is the vertical axis.

[0009] The initial stratification 1 is rotated about the y-axis by the following formula;

[0010]

[0011] Here, x, y, and z represent the values ​​at the [x, y, z] position of the normal vector. The normal vector command [x, y, z] can be used to rotate the bedding planes.

[0012] Step S3: Create contact between particles;

[0013] Step S4: Generate contacts in different groups;

[0014] Step S5: Generate random number parameters that follow a Weiber distribution;

[0015] Step S6: Using the micromechanical parameters following a Weiber distribution generated in Step S5, generate Weiber-distributed parameter models for the rock matrix-rock matrix contact, bedding-bedding contact, rock-bedding contact, rock-wall contact, and bedding-wall contact in Step S4 respectively: generate a Weiber-distributed pb_ten parameter model; generate a Weiber-distributed pb_coh parameter model; generate a Weiber-distributed pb_fa parameter model.

[0016] Step S7: Simulate the destruction of the pb_ten parameter model, the pb_coh parameter model, and the pb_fa parameter model established in step S6.

[0017] Step S1 includes:

[0018] Step S1-1: Observe and statistically analyze the microstructure of the layered rock to determine the grain type, grain size and content of the rock under study, as well as the width and distribution of the layered structure of the rock under study, and obtain the micromechanical parameters of the rock matrix and the layering.

[0019] Step S1-2: Based on the observation and statistical results obtained in step S1-1, the homogeneity of the rock is classified, and a numerical model is established based on the grain size, content, width and distribution of the bedding structure. The shape of the numerical model includes, but is not limited to, cylinders and cubes.

[0020] Step S2 includes:

[0021] Step S2-1: The initial bedding 1 is rotated around the y-axis to obtain rotated bedding 2 with different included angles α°, so as to realize the establishment of bedding dip angle models;

[0022] Step S2-2: Group the rock matrix and bedding of the rock numerical model: When α° is less than 70°, starting from the bottom 5 of the rock numerical model, take the first rotated bedding 2 obtained by rotating the initial bedding 1 by α°, and add the next rotated bedding towards the top of the rock numerical model with the distribution of the rotated bedding 2 structure as the interval.

[0023] Step S2-3: Repeat step S2-2, and add the next rotating bedding layer to the top of the rock numerical model in a cyclical manner until the bottom plane height of the next rotating bedding layer is greater than the height of the top 6 of the model, thus completing the grouping of the rock matrix and bedding.

[0024] Step S2-4: Group the rock matrix and bedding of the rock numerical model: When α° is greater than 70°, starting from the left side of the x-axis of the bottom 5 of the rock numerical model, starting from the first rotated bedding 2 obtained by rotating the initial bedding 1 by α°, add the next rotated bedding in the horizontal direction of the x-axis of the rock numerical model at intervals based on the distribution of the rotated bedding 2 structure.

[0025] Step S2-5: Repeat step S2-4, and cyclically add the next rotating bedding in the horizontal direction of the x-axis of the rock numerical model until the x-axis corresponding to the left plane of the next rotating bedding is outside the rock numerical model, thus completing the grouping of rock matrix and bedding.

[0026] Step S3 includes: regenerating the particles of the rock matrix and bedding according to the rock matrix and bedding grouping obtained in step S2.

[0027] The step S4 includes: step S4-1, generating contacts for the particles regenerated in step S3 according to groups, the generated contacts are divided into rock matrix-rock matrix contact, bedding-bedding contact, rock-bedding contact, rock-wall contact, bedding-wall contact, and setting the contacts between different particles and the wall to contacts with the same parameter properties;

[0028] Step S4-2: Assign the micromechanical parameters of the rock matrix and bedding from step S1 to rock matrix-rock matrix contact, bedding-bedding contact, rock-bedding contact, rock-wall contact, and bedding-wall contact, respectively, to generate the above different grouped contacts.

[0029] Step S5 includes: using the FISH language to traverse all contacts, determine the contact type, obtain the micromechanical parameters of the contacts, and modify the micromechanical parameters as shown in the following formula:

[0030] Bi =b i *(-log(1-Q)) (1 / w)

[0031] In the formula, b i Here are the microscopic mechanical parameters; Q is a random number generated in the FISH language that follows a uniform distribution; w is the homogeneity coefficient, the larger the homogeneity coefficient, the more uniform the material; B i These are the modified micromechanical parameters that follow a Weiber distribution.

[0032] The pb_ten parameter model established in step S6 was subjected to simulated failure under confining pressure; the pb_coh parameter model was subjected to simulated failure under cohesion parameters; and the pb_fa parameter model was subjected to simulated failure under friction parameters, and experimental results were obtained.

[0033] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0034] This invention provides a numerical simulation method for triaxial compression of heterogeneous layered rocks. By generating non-uniform particle morphology and distribution, it achieves a heterogeneous distribution of rock micromechanics, fully reflecting the complexity of the rock's microstructure. The diverse and non-uniformly distributed particle characteristics make the simulation results more representative, realistically reproducing the heterogeneous characteristics of rocks in nature, thus improving the simulation's realism and reliability. This method accurately captures the mechanical behavior of different regions within the rock, revealing various mechanical modes and deformation mechanisms that may occur in layered rocks during triaxial compression by modeling the heterogeneity. This is of great significance for understanding the mechanical properties, fracture behavior, and long-term load-bearing capacity of rocks. The method is flexible and scalable; by adjusting the particle shape and distribution parameters, more refined simulations can be achieved when dealing with different types of rocks or different loading conditions. Attached Figure Description

[0035] Figure 1 The flowchart of the triaxial compression numerical simulation method for heterogeneous layered rocks involved in this application is shown below;

[0036] Figure 2 This application relates to a schematic diagram of layered rotation;

[0037] Figure 3 This application relates to a schematic diagram of the rock matrix and bedding groupings;

[0038] Figure 4 This application relates to the rock matrix and bedding grouping;

[0039] Figure 5 This application relates to a schematic diagram of the rock matrix and 90° bedding groupings;

[0040] Figure 6 This application relates to the grouping of rock matrix and bedding contact;

[0041] Figure 7 The parameters involved in this application follow a Weiber distribution pb_ten model;

[0042] Figure 8 The parameters involved in this application follow a Weiber distribution pb_coh model;

[0043] Figure 9 The parameters involved in this application follow a Weiber distribution pb_fa model;

[0044] Figure 10 The final destructive consequences involved in this application;

[0045] Among them, 1. Initial bedding; 2. Rotational bedding; 3. Center of rock numerical model; 4. Rock numerical model; 5. Bottom of rock numerical model; 6. Top; 7. Plane. Detailed Implementation

[0046] To better understand this invention, the following description, in conjunction with the accompanying drawings and examples, will further illustrate the invention.

[0047] The present invention adopts the following technical solution:

[0048] A triaxial compression numerical simulation method for heterogeneous layered rocks includes: establishing a cylindrical rock numerical model; generating bedding planes with thickness *a*, spacing *h*, and dip angle *α*°; generating inter-particle contacts and contacts of different groupings; determining the generated particle groupings; assigning the physical and mechanical parameters of the rock matrix to the contacts between rock matrix particles, and assigning the physical and mechanical parameters of the bedding planes to the contacts between bedding plane particles; simulating weakened contact between rock matrix particles and bedding plane particles, with the contact physical parameters being the average of the two different contact types; generating random numbers following a Weiper distribution; and multiplying the generated random numbers by the expected values ​​of the micromechanical parameters for different grouping types to obtain micromechanical parameters following a Weiper distribution.

[0049] Subsequently, the established numerical model was preloaded, the confining pressure was applied to a preset value, the speed of the upper and lower walls was controlled, and the model was loaded until it became unstable and failed, thus obtaining the experimental results.

[0050] Specifically, such as Figure 1 As shown, the triaxial compression numerical simulation method for heterogeneous bedding rocks involved in this application includes the following steps:

[0051] Step S1, establish a rock numerical model, including:

[0052] Step S1-1 involves observing and statistically analyzing the microstructure of the layered rock to determine the grain types, sizes, and contents of the rock. Simultaneously, it determines the width and distribution of the layering structure and obtains the micromechanical parameters of the rock matrix and layering. Specifically, the grain type can be determined by observing characteristics under single-polarized light (e.g., mineral color, transparency, cleavage, luster) using a polarizing microscope. Alternatively, it can be determined by observing characteristics under crossed-polarized light (e.g., interference colors, extinction types, twinning) using a polarizing microscope. Grain size can be directly measured by installing a micrometer in the microscope eyepiece to measure the major axis, minor axis, or equivalent diameter of the grains; or by using image analysis software to process the micrographs, automatically measuring the grain size and statistically analyzing its distribution. Grain content can be determined by randomly selecting the field of view under a microscope using a grid eyepiece or software, and statistically analyzing the mineral types at each grid point. Alternatively, image analysis software can be used to calculate the percentage area of ​​each mineral in the field of view, thus obtaining the rock grain content. Simultaneously, microscopic features reflecting the bedding structure can be obtained from rock slices under an electron microscope, such as grain surface morphology, pore structure, and minute layered features, as well as fine bedding structures. Energy dispersive spectroscopy (EDS) analysis can then determine the compositional differences between different microlayers. If the bedding structure is caused by compositional variations, EDS analysis can more accurately identify bedding boundaries and measure their widths. By analyzing multiple regions, the distribution characteristics of micro-bedding widths can be summarized, thereby enabling the inference of macroscopic bedding distribution patterns.

[0053] Step S1-2: Based on the observations and statistical results obtained in step S1-1, the rocks are classified according to their homogeneity. Numerical models are established based on grain size, content, bedding width, and distribution patterns. The shapes of the numerical models include, but are not limited to, cylinders and cubes. Specifically, rock homogeneity classification refers to classifying rocks according to the uniformity of their internal structure, composition, and other characteristics. Rocks with high homogeneity have relatively consistent physical properties and chemical composition throughout their interior; while rocks with low homogeneity exhibit significant inhomogeneity.

[0054] Based on the collected data, grain size distribution parameters can be set as the basis for generating grain sizes in the model. A random number generator combined with a normal distribution or other suitable distribution functions can be used to simulate the randomness of grain size. The content of different components or phases can be used as input parameters to determine the volume fraction or weight of each phase in the model. Based on the measured width data of the layered structure, parameters such as layer thickness and spacing can be set. If the layered structure has a periodic distribution, its spatial distribution can be described by defining a periodic function. For non-periodic layered structures, piecewise functions or other suitable mathematical methods can be used for approximation.

[0055] Step S2: Generate bedding planes with thickness 'a', spacing 'h', and dip angle 'α', and establish models for different bedding plane dip angles. Construct a three-dimensional rectangular coordinate system xyz with the center of the rock numerical model as the origin, where the x-axis is the horizontal axis, the y-axis is the vertical axis, and the z-axis is the vertical axis; initial bedding plane 1 is a parallel bedding plane that passes through the center of the rock numerical model and is parallel to the bottom surface of the rock numerical model.

[0056] The initial stratification 1 is rotated about the y-axis using the following formula:

[0057]

[0058] Here, x, y, and z represent the values ​​at the [x, y, z] position of the normal vector. The normal vector command [x, y, z] can be used to rotate the bedding planes.

[0059] Specifically, such as Figure 2 and Figure 3 As shown, in order to achieve bedding rotation, the initial bedding 1 is rotated around the y-axis at an angle of α° with the center 3 of the rock numerical model as the origin, i.e. the reference point, to obtain the rotated bedding 2 with a dip angle of α°.

[0060] Both initial bedding 1 and rotated bedding 2 consist of two parallel planes 7, and the particles between the two planes 7 are defined as bedding.

[0061] Step S2-1: The initial bedding 1 is rotated around the y-axis to obtain rotated bedding 2 with different included angles α°, thereby establishing bedding models with different dip angles. Specifically, conditional statements are written using the FISH language: when the dip angle α is less than 45° or greater than 135°, the bedding thickness is h; when α is between 45° and 70° or between 110° and 135°, the bedding thickness is 1.1h; when α is between 70° and 90° or between 90° and 110°, the bedding thickness is 1.2h.

[0062] Step S2-2: Group the rock matrix and bedding of the rock numerical model: When α° is less than 70°, starting from the bottom 5 of the rock numerical model, take the first rotated bedding 2 obtained by rotating the initial bedding 1 by α°, and add the next rotated bedding towards the top of the rock numerical model with the distribution of the rotated bedding 2 structure as the interval.

[0063] Step S2-3: Repeat step S2-2 to add the next rotating bedding layer to the top of the rock numerical model in a cyclical manner until the bottom plane height of the next rotating bedding layer is greater than the model height, thus completing the grouping of the rock matrix and bedding layers.

[0064] Specifically, the rock numerical model is grouped into rock matrix and bedding. When α° is less than 70°, starting from the bottom 5 of the rock numerical model, two planes 7 are used to divide the rock numerical model 4. The particles between the two planes are defined as bedding. According to the distribution pattern of the bedding structure, the above operation is repeated cyclically with the measured spacing of the bedding structure as the interval. That is, the bedding only needs to be moved in the vertical direction until the height of the bottom plane of the two planes is greater than the height of the top 6 of the model. This is considered as completing the grouping of rock matrix and bedding.

[0065] Optionally, to control the starting point at different locations of the layer, different depths can be set by controlling the first segmentation starting point, and the number of loops can be increased to achieve this.

[0066] Step S2-4: Group the rock matrix and bedding of the rock numerical model: When α° is greater than 70°, starting from the left side of the x-axis of the bottom 5 of the rock numerical model, take the first rotated bedding 2 obtained by rotating the initial bedding 1 by α° as the starting point, and add the next rotated bedding in the horizontal direction of the x-axis of the rock numerical model at intervals based on the distribution of the rotated bedding 2 structure.

[0067] Step S2-5: Repeat step S2-4, and cyclically add the next rotating bedding in the horizontal direction of the x-axis of the rock numerical model until the x-axis corresponding to the left plane of the next rotating bedding is outside the rock numerical model, thus completing the grouping of rock matrix and bedding.

[0068] Specifically, the rock numerical model is divided by two planes, and the grains between the two planes are defined as bedding. The spacing between the measured bedding structures is used as the interval according to the distribution pattern of the bedding structure. For example... Figure 3 As shown, when α° is less than 70°, the bedding can be completed simply by moving along the vertical direction. Figure 4 As shown, when α° is greater than 70°, the bedding moves horizontally from the left starting point of the x-axis at the bottom of the rock model. Figure 5 As shown, when α° is 90°, the bedding is perpendicular to the bottom 5 and moves horizontally until the bedding separates from the x-axis. This process is repeated until the height of the bottom plane of both planes is greater than the height of the top surface 6 of the model, or the bedding separates from the x-axis, completing the grouping of the rock matrix and bedding.

[0069] Step S3: Generate contact between particles. Based on the rock matrix and bedding grouping obtained in step S2, regenerate the particles of the rock matrix and bedding.

[0070] Specifically, based on the actual physical test results, the rock matrix and bedding obtained in step S2 are grouped and input with the actual physical test results, such as particle size and density, and the particles of the rock matrix and bedding are regenerated.

[0071] Step S4: Generate contacts for different groups. Specifically, generate contacts for the particles regenerated in step S3 according to their groups, that is, assign the physical and mechanical parameters of the rock matrix to the contacts between the rock matrix particles, and assign the physical and mechanical parameters of the bedding to the contacts between the bedding particles.

[0072] The generated contacts are categorized into rock matrix-rock matrix contact, bedding-bedding contact, rock-bedding contact, rock-wall contact, and bedding-wall contact, setting the contacts between different particles and the wall to have the same parameter properties. Simultaneously, the micromechanical parameters of the rock matrix and bedding from step S1 are assigned to rock matrix-rock matrix contact, bedding-bedding contact, rock-bedding contact, rock-wall contact, and bedding-wall contact, respectively, generating the above-mentioned different grouped contacts, resulting in... Figure 6 The rock matrix and bedding contact groupings are shown.

[0073] Specifically, the `fish` programming language is used to traverse all contacts and determine the contact type, including: grouping particle contacts using `fish`, obtaining all contact information (each contact has two pointers, a head and a tail), and determining the type of the head and tail pointers.

[0074] If both the first and last pointers are of type ball-ball, then continue the judgment: if the classification of the first pointer is rock matrix and the classification of the last pointer is rock matrix, then the contact is judged to be a rock-rock matrix contact; similarly, if the classification of the first pointer is rock matrix and the classification of the last pointer is bedding, then the contact is judged to be a rock-bedding contact; if the classification of the first pointer is bedding and the classification of the last pointer is bedding, then the contact is judged to be a bedding-bedding matrix contact.

[0075] If the type of the first and last pointers is ball-wall or wall-ball, then continue to make a judgment: if the classification of the first pointer is rock matrix and the last pointer is wall, then the contact is judged to be rock-wall contact; if the classification of the first pointer is bedding and the last pointer is wall, then the contact is judged to be bedding-wall contact.

[0076] Step S5: Generate random number parameters following a Weiber distribution for the rock matrix-rock matrix contact, bedding-bedding contact, rock-bedding contact, rock-wall contact, and bedding-wall contact from step S4, including:

[0077] The micromechanical parameters of the contact are obtained, and then modified to generate micromechanical parameters that follow a Weiber distribution, as shown in the following equation:

[0078] B i =b i *(-log(1-Q)) (1 / w)

[0079] In the formula, b i The micromechanical parameters are as follows: b1 represents the micromechanical parameters of the rock matrix-rock matrix contact; b2 represents the micromechanical parameters of the bedding-bedding contact; b3 represents the micromechanical parameters of the rock-bedding contact; b4 represents the micromechanical parameters of the rock-wall contact; and b5 represents the micromechanical parameters of the bedding-wall contact. Q is a uniformly distributed random number generated in the FISH language. w is the homogeneity coefficient; a larger homogeneity coefficient indicates a more homogeneous material. B i To represent the modified micromechanical parameters conforming to a Weiper distribution, B1 represents the micromechanical parameters of the rock matrix-rock matrix contact conforming to a Weiper distribution; B2 represents the micromechanical parameters of the bedding-wall contact conforming to a Weiper distribution; B3 represents the micromechanical parameters of the rock-bedding contact conforming to a Weiper distribution; B4 represents the micromechanical parameters of the rock-wall contact conforming to a Weiper distribution; and B5 represents the micromechanical parameters of the bedding-wall contact conforming to a Weiper distribution. Optionally, the micromechanical parameters can be tensile strength, cohesion, or friction angle.

[0080] Step S6: Using the micromechanical parameters that follow a Weiber distribution generated in step S5, generate Weiber distribution parameter models for the rock matrix-rock matrix contact, bedding-bedding contact, rock-bedding contact, rock-wall contact, and bedding-wall contact in step S4 respectively: generate a Weiber distribution pb_ten parameter model; generate a Weiber distribution pb_coh parameter model; generate a Weiber distribution pb_fa parameter model.

[0081] Specifically, such as Figure 7 As shown, the pb_ten parameter model is used to simulate the heterogeneity and strength properties of materials. Certain key parameters of the rock (such as tensile strength and elastic modulus) are assumed to follow a Weiber distribution. Through random sampling, each contact in the model is assigned a Weiber-distributed micromechanical parameter (i.e., tensile strength parameter value) generated in step S5, thereby simulating the heterogeneity of the rock material. In numerical simulations of laboratory rock mechanics experiments, using the pb_ten model following a Weiber distribution can more accurately reproduce the mechanical behavior of rocks under different loading conditions, such as uniaxial compression, triaxial compression, and tension experiments.

[0082] like Figure 8As shown, the pb_coh parameter model is a particle flow simulation used to model the cohesive properties between particles. Making the parameters follow a Weiper distribution better reflects the heterogeneity within the rock material. In the pb_coh model, cohesion parameters (such as cohesion force, cohesion strength, and cohesion stiffness) are made to follow a Weiper distribution. That is, each contact in the model is assigned a Weiper-distributed micromechanical parameter, i.e., a cohesion parameter value, generated in step S5, to simulate the changes in the cohesive properties within the rock, thereby obtaining the actual situation of the rock material.

[0083] like Figure 9 As shown, in particle flow simulation, the pb_fa parameter model is used to describe the frictional contact characteristics between particles. Making the parameters in the model follow a Weiber distribution better reflects the non-uniformity within the material. In the pb_fa model, frictional contact parameters (such as friction coefficient and frictional stiffness) are made to follow a Weiber distribution. That is, each contact in the model is assigned a Weiber-distributed micromechanical parameter value generated in step S5, i.e., a frictional contact parameter value, to simulate the differences in frictional characteristics within the rock, thereby obtaining the actual situation of the rock material.

[0084] Step S7 involves simulating failure of the pb_ten parameter model, pb_coh parameter model, and pb_fa parameter model established in step S6. Specifically, the pb_ten parameter model established in step S6 is subjected to confining pressure for simulated failure; the pb_coh parameter model is subjected to cohesion parameters for simulated failure; and the pb_fa parameter model is subjected to friction parameters for simulated failure, yielding experimental results.

[0085] Specifically, the established pb_ten parameter model, which follows a Weiber distribution, is pre-loaded to bring the confining pressure to a preset value. The pre-loading method involves moving the top wall of the model, controlling the speed of the upper and lower walls to cause them to displace downwards, and displacing the walls around the cylinder inwards. The two walls are loaded using a servo-based method. When the stress on a particular element or particle exceeds its own strength parameter, that element or particle will fail. Loading stops when the rock simulation specimen fails and the strain reaches the expected value. Figure 10 As shown, the horizontal axis represents the number of loading steps, and the vertical axis represents the stress value. This yields the model's failure mode, and the resulting curves are more consistent with real rock mass conditions.

[0086] The established pb_coh parameter model, which follows a Weiber distribution, is loaded with cohesion parameters until the simulated rock specimen fails. Loading is stopped once the strain reaches the expected value. This method is used to study the crack initiation, propagation, and failure process of rocks under stress due to the non-uniformity of cohesion properties.

[0087] Friction parameters are applied to the established pb_fa parameter model, which follows a Weiber distribution, until the simulated rock specimen fails. Loading is stopped once the strain reaches the expected value. This method is used to study the particle movement, deformation, and failure processes of materials under stress due to the non-uniformity of frictional properties.

[0088] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0089] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0090] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0091] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0092] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of the claims of the present invention pending approval.

Claims

1. A triaxial compression numerical simulation method for heterogeneous layered rocks, characterized in that, The method includes the following steps: Step S1: Establish a numerical model of the rock; Step S2: Generate bedding with thickness a, spacing h, and dip angle α°, and establish models for different bedding dip angles; construct a three-dimensional rectangular coordinate system xyz with the center of the rock numerical model as the origin, where the x-axis is the horizontal axis, the y-axis is the vertical axis, and the z-axis is the vertical axis. The initial stratification 1 is rotated about the y-axis by the following formula; ， Here, x, y, and z represent the values ​​at [x, y, z] of the normal vector, respectively; the rotation of the layering is achieved using the normal vector command [x, y, z]. Step S3: Create contact between particles; Step S4: Generate contacts in different groups; Step S5: Generate random number parameters that follow a Weiber distribution; Step S6: Using the micromechanical parameters following a Weiber distribution generated in Step S5, generate Weiber-distributed parameter models for the rock matrix-rock matrix contact, bedding-bedding contact, rock-bedding contact, rock-wall contact, and bedding-wall contact in Step S4 respectively: generate a Weiber-distributed pb_ten parameter model; generate a Weiber-distributed pb_coh parameter model; generate a Weiber-distributed pb_fa parameter model. Step S7: Simulate the destruction of the pb_ten parameter model, the pb_coh parameter model, and the pb_fa parameter model established in step S6. Step S2 includes: Step S2-1: The initial bedding (1) is rotated around the y-axis to obtain rotated bedding (2) with different included angles α°, so as to realize the establishment of different bedding dip angle models; Step S2-2, group the rock matrix and bedding of the rock numerical model: when α° is less than 70°, starting from the bottom (5) of the rock numerical model, start with the first rotating bedding (2) obtained by rotating the initial bedding (1) at an angle of α°, and add the next rotating bedding towards the top of the rock numerical model with the distribution of the rotating bedding (2) structure as the interval; Step S2-3: Repeat step S2-2, and add the next rotating bedding layer to the top of the rock numerical model in a cyclic manner until the bottom plane height of the next rotating bedding layer is greater than the height of the top of the model (6), thus completing the grouping of the rock matrix and bedding layers; Step S2-4: Group the rock matrix and bedding of the rock numerical model: When α° is greater than 70°, starting from the left side of the x-axis of the bottom (5) of the rock numerical model, starting from the first rotating bedding (2) obtained by rotating the initial bedding (1) at an angle of α°, add the next rotating bedding in the horizontal direction of the x-axis of the rock numerical model at intervals based on the distribution of the rotating bedding (2) structure. Step S2-5: Repeat step S2-4, and cyclically add the next rotating bedding in the horizontal direction of the x-axis of the rock numerical model until the x-axis corresponding to the left plane of the next rotating bedding is outside the rock numerical model, thus completing the grouping of rock matrix and bedding.

2. The triaxial compression numerical simulation method for heterogeneous layered rocks as described in claim 1, characterized in that, Step S1 includes: Step S1-1: Observe and statistically analyze the microstructure of the layered rock to determine the grain type, grain size and content of the rock under study, as well as the width and distribution of the layered structure of the rock under study, and obtain the micromechanical parameters of the rock matrix and the layering. Step S1-2: Based on the observation and statistical results obtained in step S1-1, the homogeneity of the rock is classified, and a numerical model is established according to the grain size, content, width and distribution of the bedding structure. The shape of the numerical model includes cylinder or cube.

3. The triaxial compression numerical simulation method for heterogeneous layered rocks as described in claim 1, characterized in that, Step S3 includes: regenerating the particles of the rock matrix and bedding according to the rock matrix and bedding grouping obtained in step S2.

4. The triaxial compression numerical simulation method for heterogeneous layered rocks as described in claim 1, characterized in that, The step S4 includes: step S4-1, generating contacts for the particles regenerated in step S3 according to groups, the generated contacts are divided into rock matrix-rock matrix contact, bedding-bedding contact, rock-bedding contact, rock-wall contact, bedding-wall contact, and setting the contacts between different particles and the wall to contacts with the same parameter properties; Step S4-2: Assign the micromechanical parameters of the rock matrix and bedding from step S1 to rock matrix-rock matrix contact, bedding-bedding contact, rock-bedding contact, rock-wall contact, and bedding-wall contact, respectively, to generate the above different grouped contacts.

5. The triaxial compression numerical simulation method for heterogeneous layered rocks as described in claim 1, characterized in that, Step S5 includes: traversing all contacts, determining the contact type, obtaining the micromechanical parameters of the contacts, and modifying the micromechanical parameters as shown in the following formula: ， In the formula, b i For microscopic mechanical parameters; Q is the number of random numbers generated following a uniform distribution; w is the homogeneity coefficient, the larger the homogeneity coefficient, the more uniform the material; B i These are the modified micromechanical parameters that follow a Weiber distribution.

6. The triaxial compression numerical simulation method for heterogeneous layered rocks as described in claim 1, characterized in that, The pb_ten parameter model established in step S6 was subjected to simulated failure under confining pressure; the pb_coh parameter model was subjected to simulated failure under cohesion parameters; and the pb_fa parameter model was subjected to simulated failure under friction parameters, and experimental results were obtained.

Images