A numerical simulation method for variable angle shear test based on particle flow

Abstract

The present invention provides a numerical simulation method for variable angle shear test based on particle flow, which belongs to the technical field of rock mechanics. The steps of the numerical simulation method for variable angle shear test include: preliminarily establishing a sample model in the particle flow; cyclically distributing points within the sample range, and using the formed lattice to perform Thiessen polygon area division; using the particle mass pointer to group the minerals, and complete the establishment of heterogeneous rocks; traversing the particle positions, and using the coordinate transformation matrix to rotate the rock sample to the angle to be tested; drawing a plane geometric model of a compression shear box, and importing the model into the particle flow for simulating compression shear box loading; using formula conversion to obtain a shear stress expression; using the compression shear box for loading, recording the specimen shear stress-displacement curve, and terminating the simulation after reaching the stress stop condition. After establishing a calculation model of a variable angle shear sample, the present invention implements a variable angle shear simulation test based on particle flow.

Description

Technical Field

[0001] The invention relates to the technical field of rock mass mechanics, and in particular to a numerical simulation method for a variable angle shear test based on particle flow. Background Art

[0002] Rock, as a solid aggregate composed of various mineral grains, exhibits significant anisotropic characteristics determined by its composition, content and micromechanical parameters. The variable angle shear test, also known as the wedge shear test, is a rock mechanics test method. In this test, a vertical load is applied to the specimen by a press, and the specimen is guided to shear along a predetermined shear plane with the help of special fixtures such as a shear box and an angle die. During the test, the normal stress and shear stress on the shear plane are analyzed according to the principle of static equilibrium by considering factors such as the number of rollers, the area of ​​the shear zone, and the peak strength of the rock. By analyzing the test data of multiple groups of specimens, according to Coulomb's law, the two key mechanical parameters of rock cohesion c and internal friction angle φ can be obtained. This test method has a wide range of applications in the field of geotechnical engineering, and its results provide a valuable theoretical basis for the design of support structures, the planning of construction strategies, and the optimization of engineering deployment. In view of this, developing a numerical simulation method for rock variable-angle shear that comprehensively considers multiple factors such as material properties and loading conditions can not only effectively compare with actual test results, but also deeply analyze the failure mode of rock under different loading conditions, which has significant practical value for improving the scientific nature of engineering design and the safety of the construction process.

[0003] Currently, Voronoi (i.e., Thiessen polygons) are mostly used to group grain structures to simulate the heterogeneity of rocks. Voronoi is a method of dividing space into several regions, each of which is defined by a generating point (or seed point) and contains all points that are closer to the generating point than other generating points. Due to the equal division characteristics of Thiessen polygons in space, they can be used to solve spatial analysis problems. In the process of establishing a rock model, Thiessen polygons can be used to divide the original sample, and different regions divided can be given different mineral properties, thereby simulating the mechanical behavior of heterogeneous rocks. However, there are still some shortcomings in using Voronoi plates for variable angle shear simulation:

[0004] (1) When using the Thiessen polygon method to spatially segment rock samples, the geometric distribution and spacing of the points are crucial. If the points are unevenly distributed or the spacing is improperly set, some spatial regions may not be effectively divided into any Thiessen polygons, causing these regions to be incorrectly classified as the "none" group. This classification error will weaken the accuracy of the model and affect the accuracy of subsequent numerical simulations;

[0005] (2) The current rock model construction process usually starts with randomly filling particles in a predefined area, and then applying the Voronoi diagram to spatially group these particles. Although this process provides a preliminary heterogeneous structure for the simulation, it has significant limitations when performing multi-angle simulations of variable angle shear tests. The randomness of the Voronoi diagram leads to inconsistency in the results of each division, which means that the same rock sample may not be used in different simulations, that is, the simulation results cannot be guaranteed to be repeatable and comparable;

[0006] (3) After the region is divided by Voronoi, the particles are grouped according to the actual mineral composition and content to simulate different mineral grains. However, the existing grouping methods have some shortcomings: although the method of importing an external rigid cluster template can simulate the characteristics of different minerals, this process is laborious and not efficient in operation, and its application effect in large-scale and complex models is poor; although the method of using random numbers for grouping is simple, it ignores the spatial structure defined by Voronoi division, resulting in the inability to accurately simulate the spatial distribution characteristics of minerals;

[0007] (4) In traditional particle flow simulation, loading is usually achieved through wall loading. However, this method of simulating boundary loading may not accurately simulate the complex stress conditions of compression shear box loading in actual tests, and is different from the compression shear box loading method used in indoor variable angle shear tests;

[0008] The limitations of the above simulation methods are extremely unfavorable for the simulation of variable-angle shear tests, limiting the application value of the simulation in engineering design and scientific research, especially when considering key factors such as different mineral components of rocks and shear angles. Summary of the invention

[0009] In order to solve the problems mentioned in the background technology, the present invention provides a numerical simulation method for variable angle shear test based on particle flow, which is used to establish a variable angle shear model with specified mineral components, contents, shear angles, bonding types and other parameters, and can be used to study the mechanical parameters of rocks such as strength, internal friction angle, cohesion and rock failure modes at different shear angles.

[0010] The method should have the following characteristics:

[0011] (1) Parametric input: allows users to flexibly specify key parameters such as mineral composition, content, and desired shear angle based on actual sample parameters;

[0012] (2) Spatial distribution control: by using the Voronoi spatial partitioning method, the spatial distribution of mineral particles is ensured to reflect the heterogeneity of the actual rock;

[0013] (3) Accurate modeling of contact information: simulating the contact and interaction between particles, including friction and adhesion, to accurately analyze the mechanical behavior of rocks;

[0014] (4) Dynamic loading and failure simulation: Reproduction of the loading path and crack development and evolution process of rock during variable angle shearing;

[0015] (5) Integration of experimental data: combining experimental observations with simulation results to verify and calibrate the simulation model and improve its prediction accuracy.

[0016] The present invention is achieved through the following technical solutions, specifically:

[0017] A numerical simulation method for variable angle shear test based on particle flow, the steps of which include:

[0018] S1. Set the sample particle model in the particle flow and assign corresponding mechanical parameters;

[0019] S2, servo the wall velocity in the particle flow so that the sample obtains the confining pressure and reaches the set value through iterative calculation;

[0020] S3. A loop function is set in the particle flow to distribute points on the sample to generate a dot map, and the dot map is divided according to the Thiessen polygon to preliminarily form a heterogeneous rock model; the loop function uses the variable i to circulate, and distributes points slightly beyond the sample range to prevent the subsequent appearance of a "none" area;

[0021] S4. Traverse the particle grouping pointer and use the particle mass pointer to group different particles of the heterogeneous rock model; when the particles of this type reach the corresponding content, jump out of the loop and start the next mineral division until all particles are divided into the corresponding group. Due to the problems of program contact judgment and Voronoi area overlap, some "small points" will not be correctly classified into the corresponding group. The particles that meet the conditions in each judgment process can be temporarily stored in a group, such as "tg_new". After all the loops are completed, use the grouping command to put all the particles in the "tg_new" group into another different group. All minerals are operated in sequence to solve the problem.

[0022] S5. Traverse the particle contact and assign corresponding contact force parameters to the same mineral and between different minerals, including contact model, mechanical parameters, etc., to improve the heterogeneous rock model;

[0023] S6. Traverse the positions of all particles and update the positions of particles in the heterogeneous rock model using a coordinate transformation matrix; the coordinate transformation matrix is ​​a function that is defined to complete the conversion between the angle system and the radian system, and to achieve subsequent rotation transformation by changing the angle;

[0024] S7. Draw the corresponding image in CAD based on the shear box, export it to a dxf file, and use the clump command to interact with the particle flow; use the sample position pointer to identify the positions where the upper and lower shear boxes need to be placed, and then import the shear boxes to the upper and lower sides of the sample model;

[0025] S8. Use a compression shear box to load the specimen model, and convert the monitored contact force to obtain the corresponding shear stress τ and normal stress σ.

[0026] Furthermore, the mechanical parameters in S1 include the effective stiffness of the linear contact part, the normal-to-tangential stiffness ratio, the particle friction coefficient, the effective stiffness of the parallel bonding part, the normal-to-tangential stiffness ratio, the tensile strength, the cohesion, etc.

[0027] Furthermore, the wall velocity servo process in S2 is as follows: setting a stress threshold, traversing the wall contact force, and continuing the servo if the corresponding contact stress calculated is less than the stress threshold, otherwise stopping the servo, and iterating in sequence until the sample reaches a static equilibrium state.

[0028] Furthermore, the loop function in S3 uses the plot_mesh function and uses the variable i to loop.

[0029] Furthermore, in S4, a distribute_group function is written to traverse the ball particle list to obtain the ball grouping and mass pointers of all particles; in the loop process, ball_mass = ball_mass + ball.mass (bp),

[0030] ball.mass(bp) is the particle mass obtained through the particle pointer;

[0031] The particle mass can be gradually accumulated by iterative calculation; when the particle mass reaches a certain mineral mass fraction, the loop is exited and the next mineral grouping identification is continued, and so on, all mineral types can be grouped.

[0032] Furthermore, the following instructions are added to the S4 group:

[0033]

[0034] Helps solve the "small dot" doping phenomenon caused by program recognition problems.

[0035] Furthermore, the fenzu_jiaojie function is written in S5 to traverse the contact list between local spherical particles and use contact.end1(cp) and contact.end2(cp) to obtain the two particle groups contacted by the contact pointer. The particle group will play a role in the next step of identifying particles. On the premise that the contact groups of the two minerals are the same, if the two particle groups are both a certain mineral component, then the ore component is assigned to the group of the contact pointer; the groups of the remaining minerals can be determined in the same way; if the contact groups of the two minerals are different, the group of the contact pointer is recorded as "joints".

[0036] Furthermore, the coordinate transformation matrix in S6 is: Where α is the rotation angle.

[0037] Furthermore, S8 uses the position of the rigid cluster and the contact force pointer to identify the corresponding contact force, displacement, etc., and with the help of the coordinate transformation matrix in S6, the shear stress τ and the normal stress σ are converted according to the corresponding formula of variable angle shear.

[0038] Compared with the prior art, the present invention has the following beneficial features:

[0039] 1. The present invention solves the problem of "small spots" that cannot be correctly classified when using mass pointer grouping. The so-called "small spots" refer to the situation where the particle grouping is incorrect due to program misjudgment, spatial segmentation and other problems, such as the situation where a very few "plagioclase" particles are mixed into the "quartz" particle group;

[0040] 2. The present invention optimizes the modeling method for generating rock samples with different shear angles. Compared with most previous methods of rotating and dividing first, the present invention can rotate the sample to the set angle by adjusting the trigonometric function corresponding to the shear angle in the coordinate change matrix, which is more accurate and convenient.

[0041] 3. The present invention uses a compression shear box generated by a rigid cluster to simulate the loading conditions in the actual test. Compared with the traditional wall loading method, the compression shear box generated by the clump provides a more accurate and realistic loading environment, so that the simulation results can more intuitively reflect the experimental phenomena;

[0042] 4. The present invention can record the position and number of tensile and shear transcrystalline cracks in the mineral by identifying the position and number of cracks. It can also record the position and number of tensile and shear transcrystalline cracks between different minerals (such as between plagioclase and quartz, between plagioclase and mica, etc.), so as to more accurately analyze the development process of cracks.

[0043] The numerical simulation method of the variable angle shear test proposed in the present invention can provide more accurate and reasonable simulation results, which is of great significance for understanding the mechanical behavior of rock under multi-angle shear. By comprehensively comparing and analyzing the results of the numerical simulation with the test results under actual engineering conditions, the crack propagation evolution and failure mode of the rock under different shear angles can be observed and recorded in detail, providing theoretical support for more accurate evaluation of the damage and failure process of rock materials in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0045] Figure 1 It is a schematic flow chart of a numerical simulation method for a variable angle shear test based on particle flow provided by the present invention;

[0046] Figure 2 It is a schematic diagram of forming a dot matrix by defining a function and looping using a variable i in an embodiment of the present invention;

[0047] Figure 3 is a schematic diagram of dividing a sample using Thiessen polygons according to a dot map in an embodiment of the present invention;

[0048] Figure 4 This is a schematic diagram of solving the "small dot doping" problem when using quality indicator grouping in an embodiment of the present invention;

[0049] Figure 5 It is a schematic diagram of solving the problem of "small dot doping" when using quality indicator grouping in an embodiment of the present invention;

[0050] Figure 6 This is a schematic diagram of adding bonding to the mineral components inside the rock mass in an embodiment of the present invention; a parallel bonding contact model is used inside the same type of minerals, and a smooth joint contact model is used between different types of minerals;

[0051] Figure 7 is a schematic diagram of a sample at different shear angles in a particle flow simulation in an embodiment of the present invention;

[0052] Figure 8 is a schematic diagram of loading a sample at different shear angles using upper and lower shear boxes in a particle flow simulation in an embodiment of the present invention;

[0053] Fig. 9 is a schematic diagram comparing the experimental and simulated shear stress-displacement curves in an embodiment of the present invention;

[0054] Fig.10 It is a schematic diagram comparing the sample failure modes at a shear angle of 65° in actual experiments and particle flow simulation in an embodiment of the present invention.

[0055] Fig.11 It is a schematic diagram of the variation of the number of cracks such as transgranular tensile cracks and transgranular shear cracks with time and the shear stress-time curve of the sample under a simulated shear angle of 65° in an embodiment of the present invention; DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described in combination with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] like Figure 1 As shown, the numerical simulation method of variable angle shear test based on particle flow provided by the present invention comprises the following steps:

[0058] S1. Set the sample particle model in the particle flow and assign corresponding mechanical parameters;

[0059] S2, servo the wall velocity in the particle flow so that the sample obtains the confining pressure and reaches the set value through iterative calculation;

[0060] S3. A loop function is set in the particle flow to distribute points on the sample to generate a dot map, and the dot map is divided according to the Thiessen polygon to preliminarily form a heterogeneous rock model; the loop function uses the variable i to circulate, and distributes points slightly beyond the sample range to prevent the subsequent appearance of a "none" area;

[0061] S4. Traverse the particle grouping pointer and use the particle mass pointer to group different particles of the heterogeneous rock model; when the particles of this type reach the corresponding content, jump out of the loop and start the next mineral division until all particles are divided into the corresponding group. Due to the problems of program contact judgment and Voronoi area overlap, some "small points" will not be correctly classified into the corresponding group. The particles that meet the conditions in each judgment process can be temporarily stored in a group, such as "tg_new". After all the loops are completed, use the grouping command to put all the particles in the "tg_new" group into another different group. All minerals are operated in sequence to solve the problem.

[0062] S5. Traverse the particle contact and assign corresponding contact force parameters to the same mineral and between different minerals, including contact model, mechanical parameters, etc., to improve the heterogeneous rock model;

[0063] S6. Traverse the positions of all particles and update the positions of particles in the heterogeneous rock model using a coordinate transformation matrix; the coordinate transformation matrix is ​​a function that is defined to complete the conversion between the angle system and the radian system, and to achieve subsequent rotation transformation by changing the angle;

[0064] S7. Draw the corresponding image in CAD based on the shear box, export it to a dxf file, and use the clump command to interact with the particle flow; use the sample position pointer to identify the positions where the upper and lower shear boxes need to be placed, and then import the shear boxes to the upper and lower sides of the sample model;

[0065] S8. Use a compression shear box to load the specimen model, and convert the monitored contact force to obtain the corresponding shear stress τ and normal stress σ.

[0066] The terms appearing in the embodiments of the present invention are explained to facilitate understanding of the present invention by those skilled in the art.

[0067] Terminology explanation:

[0068] Particle Flow Code (PFC) software is an application example of discrete element method, which can effectively simulate the complex dynamic process of granular media, including the movement, collision and other contact effects of particles. PFC software relies on discrete element method to handle complex large displacement and rotation phenomena inside the rock mass, and even simulate the block separation process, providing a practical simulation method for studying the nonlinear large deformation characteristics in jointed rock mass.

[0069] Traversal: In particle flow, "traversal" is a key computational process that involves systematic inspection and data processing of key components of the model, including balls, contacts, walls, and rigid clusters. By executing the "loop foreach" command, the software is able to visit each element within the defined range along a preset search path and extract its key properties, such as position and velocity, as data pointers for further analysis.

[0070] For example, traversing the particle list (such as "ball.list") allows obtaining the state parameters of all particles, which are very important for performing subsequent spatial coordinate transformations and size adjustments; similarly, traversing the contact list (such as "contact.list") can identify and classify all contact types, which is important for applying adhesion models and distinguishing contact properties; traversing the particle grouping ("ball.group") provides a means to group and classify particles; and traversing the wall contact map ("wall.contactmap") provides data support for calculating the forces and stresses interacting with the wall.

[0071] The calculation process of particle flow is based on Newton's second law. It calculates the motion state of particles, such as position and velocity, and then updates the contact force according to the force-displacement relationship between particles. This process is iterated through time steps until the entire model reaches rapid convergence and forms a (quasi) static equilibrium. Therefore, traversal is not only a basic operation of particle flow, but also a core link to ensure simulation accuracy and efficiency.

[0072] Pointers: In the core mechanism of PFC particle flow simulation, "pointers" play a vital role. Pointers are special variables used to track and reference particles or their collections, providing direct access to specific data points in the particle data structure. This mechanism enables the simulation program to quickly retrieve and adjust key properties of particles, such as their position, velocity, and force state. The use of pointers is not limited to static data access. They can also dynamically reference newly generated particles or mark those that are about to be removed from the simulation. When performing large-scale parallel calculations, another key role of pointers is to coordinate the distribution of tasks between different processors or computing nodes. Especially when dealing with complex models containing millions of particles, by using pointers intelligently, the use of computing resources can be optimized and the simulation process can be accelerated. In addition, pointers also support data export functions, allowing the state and interaction information of particles to be exported to text files (such as txt) or data files (such as dat). This step is crucial to achieve real-time monitoring of simulation results and subsequent detailed analysis, providing researchers with powerful data visualization and a means of in-depth understanding of simulation behavior. In summary, pointers are key tools for achieving efficient data processing and optimizing simulation performance in PFC particle flow simulation. They make the simulation process more flexible and responsive, and provide convenience for subsequent analysis.

[0073] Servo: In granular flow, servo control is a key mechanism used to precisely adjust parameters such as the position, state, and stress of the specimen to achieve consistency with the preset target, which is of great significance for simulating the stress state that the specimen is subjected to in the actual engineering environment. Servo control is usually performed in the early stages of modeling to preload the specimen to reproduce its original stress conditions in the formation. To this end, servo commands are used to finely control the movement speed of the simulated boundary (such as the wall). The displacement of the wall is then transmitted to the contacting particles, causing relative displacement and overlap between the particles. Based on the overlap of the particle contact and the predefined contact stiffness, the contact force between the particles can be calculated, and the total contact force acting on the specimen is obtained by accumulating the contact forces applied by all particles in contact with the wall. Then, the contact stress between the specimen and the boundary is determined by appropriate conversion. This process involves repeated iterative calculations until a predetermined stress threshold is reached. If the contact stress is less than the set threshold, the servo control will continue to adjust until the stress reaches or exceeds the threshold, at which time the servo control stops and the preloading process is completed. Pre-pressing not only makes the sample structure more compact, but also ensures sufficient contact between particles, which is crucial for the stability and authenticity of subsequent simulations. In addition, pre-pressing helps release the unbalanced forces inside the sample, laying a solid foundation for further simulations.

[0074] clump: In particle flow simulation software, "clump" is a module for creating and manipulating rigid clusters. These rigid clusters are filled with particles called "pebbles". Unlike conventional particles (balls), "pebbles" do not undergo relative displacement or deformation and exhibit rigidity. Therefore, they are used in simulations to represent structural units that will not be destroyed by internal forces. The "clump" module not only supports the generation of simple rigid structures through the built-in instructions of the software, but also can create more complex and accurate rigid components by importing external geometric model files, such as DXF or STL formats. This function greatly enhances the flexibility and realism of the simulation, allowing engineers to design and implement more realistic loading schemes according to specific engineering conditions and loading requirements. In the present invention, the "clump" module is used to simulate the compression shear box in the variable angle shear test. Compared with the traditional wall loading method, the compression shear box created using "clump" can more accurately simulate the loading conditions and boundaries in the actual test, thereby providing a more intuitive and realistic simulation effect. This method not only improves the accuracy of the simulation, but also provides more reliable data support for material behavior research and engineering design in geotechnical engineering.

[0075] Embodiment 1:

[0076] Step 1: Establish a numerical model of heterogeneous rock in the particle flow numerical simulation software: the size of the heterogeneous rock numerical model is 50mm*50mm, the minimum diameter of the spherical particles used to fill the sample is 0.23mm, the maximum and minimum particle size ratio is 1.5, the particle density is 0.07, and finally 8736 spherical particles are generated;

[0077] Step 2: Preliminarily assign mechanical parameters such as effective stiffness and normal-tangential stiffness ratio for subsequent modeling;

[0078] Step 3: Servo the wall speed to generate a confining pressure of 0.1 MPa on the sample, which can release part of the internal force of the model while compacting it, and can better add contact later;

[0079] Step 4: Write the plot_mesh function, and the variable i is looped from 1 to 500 to arrange the points, as shown in the following figure. Figure 2 shown.

[0080] Generate point horizontal coordinates:

[0081] x1=-0.025*1.2+0.05*1.2*math.random.uniform,

[0082] Point vertical coordinate:

[0083] y1=-0.025*1.2+0.05*1.2*math.random.uniform.

[0084] Among them, -0.025 and 0.025 are the upper and lower positions that the sample can reach on the x / y axis; math.random.uniform is a random number in the range of 0 to 1 that comes with the software; 0.05*1.2*math.random.uniform is the x / y coordinate in a slightly larger range than the sample. This operation can prevent the problem of "none" area appearing during subsequent Thiessen polygon division;

[0085] Step 5: Use the "geometry tessellate Voronoi" command of the particle flow to divide the area according to the division rules of Thiessen polygons in the dot map generated in step 4. The effect after the division is completed is shown in the attached figure. Figure 3 As shown;

[0086] Step 6: Write the "distribute_group" function, traverse the list of ball particles, and obtain the ball grouping and mass pointers of all particles.

[0087] During the loop, ball_mass = ball_mass + ball.mass(bp), where ball.mass(bp) is the particle mass obtained through the particle pointer. By iteratively calculating in sequence, the particle mass can be gradually accumulated. When the particle mass reaches a certain mineral mass fraction, the loop is exited and the grouping identification of the next mineral is continued;

[0088] if ball_mass < xiechangshi_content * total_ballmass then name = string('xiechangshi'). For example, this command can group the particles reaching the corresponding content of plagioclase as "plagioclase". And so on, the grouping of all mineral species can be completed.

[0089] Step 7: Add instructions in the grouping of Step 6 to help solve the "dot" doping phenomenon caused by program recognition problems.

[0090] if ball_mass < xiechangshi_content * total_ballmass then

[0091] c_g = ball.group(bp,1)

[0092] c_g_new = int(c_g)

[0093] name = string('xiechangshi')

[0094] endif

[0095] ball group 'xiechangshi' slot 2 range group[c_g_new] slot 1

[0096] Put the previously assigned ball particles into a new group c_g_new first, and then reassign the plagioclase group originally belonging to c_g_new to slot2, and this problem can be solved. Operate in the same way for the remaining 3 minerals, and the "dot" doping problem can be solved. The effect comparison diagram is as shown in the appendix Figure 4 、 5 shown.

[0097] Step 8: Write the "fenzu_jiaojie" function, traverse the contact list between local spherical particles, and use contact.end1(cp) and contact.end2(cp) to obtain the two particle groups that the contact pointer contacts. The particle group will play a role in the next step of identifying particles. Under the premise that the contact groups of the two minerals are the same, if both particle groups are plagioclase, then "plagioclase" is assigned to the group of the contact pointer. Similarly, the groups of the remaining three groups of minerals can be determined; if the contact groups of the two minerals are different, the group of the contact pointer is recorded as "joints". After completing all the above operations, the establishment of heterogeneous rocks can be achieved.

[0098] Step 9: Define three arrays deform_arr(2,2), pb_deforam_arr(2,2), and bond_arr(1,2) to store the effective modulus, stiffness ratio and other parameters of the linear part and parallel bonding part for subsequent assignment.

[0099] if contact.group(cp,2)='xiechangshi'

[0100] then contact.model(cp)='linearpbond'

[0101] If the contact group is identified as "plagioclase", its contact pointer type is assigned as "linear parallel bonding model"; then according to the properties and modes of the linear parallel bonding model, the array is used to assign parameters to define the effective modulus, stiffness ratio, tensile strength, cohesion, internal friction angle, friction coefficient and other parameters, and the mechanical parameter definition of the "plagioclase" mineral can be completed. Following the above method, the other three minerals and joint groups are assigned parameters to complete the complete setting of the internal contact model of the sample. The effect diagram is shown in the attached figure. Figure 6 shown.

[0102] Step 10: Calibrate the mesoscopic parameters by comparing with the actual physical experimental results, and give the numerical model of the filled jointed rock the mesoscopic parameters after calibration. The mesoscopic parameter table is as follows:

[0103] Table 1 Microscopic parameters

[0104] Mineral Group emod / GPa kratio Pb_emod / MPa Fric Pb_kratio Pb_ten / MPa Pb_coh / MPa Pb_fa / ° Bond_gap / m Plagioclase 10 1.2 10 0.1 1.2 21.1 57 11 0.23e-4 Potassium feldspar 11 1.1 11 0.1 1.1 18.5 50 15 0.23e-4 quartz 6 1.1 6 0.1 1.1 15 42 20 0.23e-4 Mica 5 1.3 5 0.2 1.3 6.9 20 20 0.23e-4 Joint Group Sj_kn / Gpa Sj_ks / GPa Sj_fric Sj_fa / ° Sj_coh / MPa Sj_ten / MPa Bond_gap / m Smooth joints 10e3 7e3 0.1 15 8 4 0.46e-4

[0105] Step 11: Define the "zhuanhua" function, and use the calculation formula of radians = angle * π / 180° to complete the conversion between angle system and radian system. Define the xz function, traverse all particle lists to obtain information such as the coordinates of each particle, and update the x / y coordinates of each point using the coordinate transformation matrix for subsequent tests at different angles. The coordinate transformation matrix is: The effect after rotating 35°, 50°, 65°, 80° is shown in the attached figure. Figure 7 shown.

[0106] Step 12: In AutoCAD, draw a plane geometry according to the shape of the shear box, convert it to a dxf file, and use the geometry import yajian_box1_65.dxf command to import it into the particle flow. Use the "clumptemplate create" command to generate the dxf file as a template in the working interface. Define the "clump_pos" function, first use the clump1 = clump.find(1) command to find the position of the inserted clump, and then use the clump position pointer to fix the shear box to the upper and lower sides of the specimen. The completed effect diagram is shown in the attached figure. Figure 8 As shown:

[0107] Step 13: Define the force function, identify the contact force of the upper and lower shear boxes, take the average value as the axial load, and then combine it with the variable angle shear conversion formula:

[0108] τ=P / A(sinα-fcosα) can complete the definition of shear stress τ.

[0109] In the formula τ=P / A(sinα-fcosα), P is the axial pressure; α is the shear angle; f is the friction coefficient, f=1 / (n*d), n is the number of rollers in the variable angle shear loading device, d is the diameter of the rollers in the variable angle shear loading device (mm); α is the shear angle.

[0110] Define the axial_strain_wall function and use the clump displacement pointer to monitor the vertical displacement of the upper and lower shear boxes for subsequent observation of the shear stress-displacement curve.

[0111] Step 14: Use the shear box introduced in step 12 to load up and down, fix its vertical speed and rotation, and have no limit on the horizontal speed, to simulate the effect of the upper and lower loading devices sliding along the rollers during the variable angle shear test. Finally, load according to the time step. During the loading process, parameters such as stress, displacement, and number of cracks can be detected. Stop loading when the corresponding stop condition is reached, and the variable angle shear simulation ends.

[0112] Combined with the above method, the actual test of the sample failure at a 65° shear angle was carried out, and the sample failure at a 65° shear angle was simulated. The results are as follows: Figure 9-10 As shown, it can be seen that the simulation data of Example 1 is close to the actual experimental data. Further, the evolution law of the number of cracks such as transgranular tensile cracks and transgranular shear cracks over time can be obtained, and the effect diagram is shown in the attached figure. Fig.11 shown.

[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A numerical simulation method for variable angle shear test based on particle flow, characterized in that: The steps include: S1. Set the sample particle model in the particle flow and assign corresponding mechanical parameters; S2, servo the wall velocity in the particle flow so that the sample obtains the confining pressure and reaches the set value through iterative calculation; S3, setting a circulation function in the particle flow to distribute the sample to generate a dot map, and dividing the dot map according to Thiessen polygons to preliminarily form a heterogeneous rock model; S4, traversing the particle grouping pointers, and using the particle mass pointers to group different particles of the heterogeneous rock model; S5. Traverse the particle contacts, assign corresponding contact force parameters within the same mineral and between different minerals, and improve the heterogeneous rock model; S6, traversing the positions of all particles, and updating the positions of particles in the heterogeneous rock model using a coordinate transformation matrix; S7. Draw the corresponding image in CAD based on the shear box, export it to a dxf file, and use the clump command to interact with the particle flow; use the sample position pointer to identify the positions where the upper and lower shear boxes need to be placed, and then import the shear boxes to the upper and lower sides of the sample model; S8. Use a compression shear box to load the specimen model, and convert the monitored contact force to obtain the corresponding shear stress τ and normal stress σ.

2. The numerical simulation method of variable angle shear test based on particle flow according to claim 1, characterized in that: The mechanical parameters in S1 include one or more of the effective stiffness of the linear contact part, the normal-tangential stiffness ratio, the particle friction coefficient, the effective stiffness of the parallel bonding part, the normal-tangential stiffness ratio, the tensile strength and the cohesion.

3. The numerical simulation method of variable angle shear test based on particle flow according to claim 1, characterized in that: The wall velocity servo process in S2 is as follows: set the stress threshold, traverse the wall contact force, and continue the servo if the corresponding contact stress calculated is less than the stress threshold, otherwise stop the servo, and iterate in sequence until the sample reaches a static equilibrium state.

4. The numerical simulation method of variable angle shear test based on particle flow according to claim 1, characterized in that: The loop function in S3 uses the plot_mesh function and uses the variable i to loop.

5. The numerical simulation method of variable angle shear test based on particle flow according to claim 1, characterized in that: Write the distribute_group function in S4 to traverse the list of ball particles and obtain the ball grouping and mass pointers of all particles; In the loop, ball_mass = ball_mass + ball.mass (bp), ball.mass(bp) is the particle mass obtained through the particle pointer; The particle mass can be gradually accumulated by iterative calculation; when the particle mass reaches a certain mineral mass fraction, the loop is exited and the next mineral grouping identification is continued, and so on, all mineral types can be grouped.

6. The numerical simulation method of variable angle shear test based on particle flow according to claim 5, characterized in that: Add the following instructions to the S4 group: if ball_mass <xiechangshi_content*total_ballmass then c_g = ball.group(bp,1) c_g_new = int(c_g) name = string('xiechangshi') endif ball group'xiechangshi'slot 2range group[c_g_new]slot 1.

7. The numerical simulation method of variable angle shear test based on particle flow according to claim 1, characterized in that: In S5, write the fenzu_jiaojie function to traverse the contact list between local spherical particles and use contact.end1(cp) and contact.end2(cp) to obtain the two particle groups that the contact pointer contacts. The particle group will play a role in the next step of identifying particles. On the premise that the contact groups of the two minerals are the same, if the two particle groups are both a certain mineral component, then the ore component is assigned to the group of the contact pointer; the groups of the remaining minerals can be determined in the same way; if the contact groups of the two minerals are different, the group of the contact pointer is recorded as "joints".

8. The numerical simulation method of variable angle shear test based on particle flow according to claim 1, characterized in that: The coordinate transformation matrix in S6 is:

9. The numerical simulation method of variable angle shear test based on particle flow according to claim 8, characterized in that: S8 uses the position of the rigid cluster and the contact force pointer to identify the corresponding contact force and displacement, and with the help of the coordinate transformation matrix in S6, the shear stress τ and the normal stress σ are converted according to the corresponding formula of variable angle shear.

Images