Discontinuous graded soil large-scale discrete element model generation method

Abstract

The invention provides a discontinuous graded soil large-scale discrete element model generation method. The method comprises the following steps: (1) creating a soil particle material and setting soil intrinsic parameters and contact model mesoscopic parameters; (2) creating a 3D penetrometer geometric model for assisting model hierarchical generation and impact simulation; (3) importing the geometric model into EDEM software; (4) creating a soil particle factory for hierarchically generating the model; (5) generating particles with different scale factors in the soil particle factory; (6) selecting different geometric model combinations for assisting model hierarchical generation in the step (2) to generate a model for simulation; and (7) importing the geometric model of the 3D penetrometer in the step (2) into the model generated in the step (6) to carry out impact penetration dynamics simulation. According to the method, construction of the discontinuous graded soil large-scale discrete element model can be realized, and the problem that effective calculation cannot be carried out due to too many particles of a large-scale model generated by fine-grained soil can be effectively solved.

Description

Technical Field

[0001] The invention relates to the technical field of soil dynamic response analysis, and in particular to a method for generating a large-scale discrete element model of discontinuously graded soil. Background Art

[0002] The dynamic response analysis of soil is one of the most complex and challenging problems in computational geomechanics. Penetrometers are widely used as in-situ measurement tools to evaluate the mechanical properties of soil. During the penetration process, the contact surface between the penetrometer and the soil is constantly changing. It is necessary to consider the relative incompressibility of the soil in a short time interval, the inhomogeneity of stress-strain behavior, and the rate dependence of soil stiffness and strength. The dynamic response and mechanism research are the basis for quantitative mechanical relationships and model construction and correction. The use of simulation methods is an indispensable means to carry out mechanism analysis. Simulation simulation can also carry out working conditions that cannot be achieved experimentally and reduce experimental costs. Finite element simulation is based on the concept of continuous media and can only describe the overall force behavior of the soil. It cannot accurately obtain the large deformation process and microscopic information changes of the soil. The discrete element method can simulate the complex mechanical behavior of particles and their movement process by analyzing the force and motion laws between soil particles at a microscopic scale. It has become a common method for studying particle dynamics and is widely used to simulate the interaction between soil and tools.

[0003] The commonly used software for discrete element simulation are PFC and EDEM. PFC is mainly implemented by commands, which is relatively flexible, but it is difficult to simulate complex geometric bodies, and it cannot use GPU parallel computing, and the particle model generation efficiency is low. EDEM has a mature operation interface, which can quickly realize complex geometric structure modeling by importing CAD geometric models, and has a secondary development interface (API interface) based on C++ language, which supports custom complex dynamic models.

[0004] The discrete element model used for engineering analysis is close to the actual model size, which can eliminate the influence of size effect and boundary effect and provide more accurate simulation results. However, when generating large-scale models of discontinuously graded fine-grained soil, there will be too many particles and it will be impossible to perform efficient calculations. Especially when it comes to the simulation of impact penetration involving cones or pile-soil interaction, there will be a significant particle size effect at the tip of the cone due to the extremely uneven particles. Most of the models used in existing discrete element simulations are scaled down through PFC command operations to generate smaller-scale models, or simplified by enlarging the particle size through particle size scaling, or by scaling the particle size by a large multiple to generate a half-section, 1 / 4 size or smaller model to reduce the total number of particles, such as Fig.11 (a), but this method is not suitable for materials with particle grading, because the randomly generated particle size distribution will be non-uniform. Fig.11 As shown in (b), even if the particle refinement method is adopted, although the final large-scale model is generated, the particle size is large and the particles are of a single particle size, and the particle gradation is not considered. The implementation scheme closest to the present invention is to implement particle refinement through the PFC command method, which takes the particle gradation into consideration, but the particle gradation distribution is relatively uniform, and there is no need to consider the particle size effect. Fig.11 (c) In addition, due to the disadvantage of low PFC operating efficiency, this method is not suitable for generating large-scale models, and it is also unable to achieve modeling of complex geometric bodies. Moreover, the selection of the stratification method and the scale factor of the particle size scaling during the refinement process need to be adjusted and verified through trial and error to find the appropriate parameters, which is inefficient. At present, there is still a blank in the method of generating large-scale discrete element models for discontinuous graded soils based on EDEM technology.

[0005] The problem of generating large-scale discrete element models for discontinuously graded soils, especially fine-grained soils with extremely uneven particle distribution, has not been solved. To generate discrete element models for such soils or granular materials, it is crucial to refine the relevant models and determine the scaling parameters. On the basis of generating a suitable number of particles for both computational efficiency and consideration of boundary effects and particle size effects, it is necessary to simultaneously generate discrete element models for such soils or granular materials. Summary of the invention

[0006] In view of the technical problems existing in the above-mentioned background technology, the present invention proposes a method for generating a large-scale discrete element model of discontinuous graded soil. The method has a reasonable concept and can realize the construction of a large-scale discrete element model of discontinuous graded soil, and can effectively solve the problem that the number of particles in the large-scale model generated by fine-grained soil is too large to be effectively calculated.

[0007] In order to solve the above technical problems, the present invention provides a method for generating a large-scale discrete element model of discontinuously graded soil, which mainly includes the following steps:

[0008] (1) Create soil particle materials and set soil intrinsic parameters and contact model microscopic parameters;

[0009] (2) Create a 3D penetrometer geometry model to assist in model layer generation and impact simulation;

[0010] (3) Importing the geometric model used to assist model layer generation into the EDEM software;

[0011] (4) Create a soil particle factory for layered generation model;

[0012] (5) Generate particles of different scale factors in a soil particle factory;

[0013] (6) selecting different geometric models used to assist in model hierarchical generation in step (2) to combine and generate a model for simulation;

[0014] (7) Importing the 3D penetrometer geometric model in step (2) above into the model generated in step (6) above to perform impact penetration dynamics simulation, using the same velocity and simulation parameters, and outputting the resistance value of the cone tip for comparison, so as to select and verify a suitable model.

[0015] The method for generating a large-scale discrete element model of discontinuously graded soil, wherein the specific process of step (1) is as follows: first, a new soil particle material model is created, the name of the soil particle material is edited, and the intrinsic parameters and contact model parameters of the soil particle material model used to generate soil particles are set; then, the particle shape required for simulation is added to the soil particle material model and the particle parameters are set, and the particle parameters include the initial particle radius size and particle size distribution; for discontinuously graded particles, on the basis of particle size scaling, the percentage of each particle size is set by the user according to the particle grading experimental results, so as to complete the setting corresponding to the particle size distribution of the real material.

[0016] The method for generating a large-scale discrete element model of discontinuous graded soil, wherein: the intrinsic parameters include Poisson's ratio, true density and shear modulus, which can be determined based on particle material physical property experiments; the contact model parameters include restitution coefficient, static friction coefficient and rolling friction coefficient.

[0017] The method for generating a large-scale discrete element model of discontinuously graded soil, wherein: the step (1) can also use the Import function of BulkMaterial-Material-particle to import an external Excel spreadsheet of relevant particle grading distribution information, thereby completing the setting corresponding to the actual material particle size distribution.

[0018] The method for generating a large-scale discrete element model of discontinuous graded soil, wherein the specific process of step (2) is as follows: a geometric model for assisting model layer generation is established through three-dimensional modeling software, such a geometric model plays an isolating role for discrete element models generated layer by layer using different scale factors, and at the same time a 3D penetrometer geometric model for impact simulation is constructed.

[0019] The method for generating a large-scale discrete element model of discontinuous graded soil, wherein: in the step (2), when generating the geometric model, the inner circle of the geometric model with the smallest diameter is made hollow, so that particles are generated in the innermost circle area by setting up a particle factory.

[0020] The method for generating a large-scale discrete element model of discontinuously graded soil, wherein the specific process of step (3) is: using the geometry import function of the EDEM software to import the geometric model generated in step (2), forming different geometric model combinations according to simulation requirements, and uniformly adjusting the length unit to mm during import, and setting the Poisson's ratio, density, shear modulus of the imported geometric model and the contact parameters between the geometric model and soil particles of different scale factors.

[0021] The method for generating a large-scale discrete element model of discontinuously graded soil, wherein the specific process of step (4) is as follows:

[0022] (4.1) For the innermost region of the geometric model combination formed in step (3), firstly, a geometric body for generating particles is added through EDEM software to form an innermost particle factory, wherein the shape and size of the added geometric body are consistent with the central region of the geometric model with the smallest diameter in step (2), and then the geometric body is subjected to a volume filling particle generation method to generate soil particles in the innermost region;

[0023] (4.2) For areas outside the innermost circle where different geometric model combinations are formed according to simulation requirements, the volume filling particle generation method is used directly on each geometric model used to assist in model layer generation to generate soil particles in each area according to simulation requirements.

[0024] The method for generating a large-scale discrete element model of discontinuous graded soil, wherein the specific process of step (5) is as follows: after the particle generation method is set in step (4), the soil type ratio and solid volume ratio parameters in the volume filling particle generation method are set in sequence starting from the central area with the smallest inner circle diameter; a particle refinement method is used to achieve the smallest particles generated in the innermost circle and the particle sizes of the particles generated in the outer layer area become larger in sequence.

[0025] The method for generating a large-scale discrete element model of discontinuously graded soil, wherein the specific process of performing the impact penetration dynamics simulation in step (7) is as follows:

[0026] The total calculation time of the impact penetration dynamics simulation is set to 0.1s, and 0.005s is used as the storage time interval of data and calculation results. In the discrete element simulation, the Rayleigh time step T R Solve by the following formula:

[0027]

[0028] In the above formula, R is the radius of the particle, ρ is the density of the particle, G is the shear modulus, and v is the Poisson's ratio;

[0029] Rayleigh time step T RDuring the solution process, the changes in the resistance value at the cone tip and the number of contacts between the cone tip and the particles during the entire penetration process are solved simultaneously, where the cone tip resistance is the force component along the axis Z direction at the cone tip during the penetration process; the number of contacts between particles at the cone tip is calculated using the coordination number, which refers to the number of contacts between each particle and another particle at any given time point, and the average coordination number is expressed as:

[0030]

[0031] Where M i is the contact number of the ith particle.

[0032] By adopting the above technical solution, the present invention has the following beneficial effects:

[0033] The method for generating a large-scale discrete element model of discontinuously graded soil of the present invention is reasonably conceived and can fill the gap in the method for generating a large-scale discrete element model of discontinuously graded soil based on EDEM technology. It is suitable for any fine-grained soil or granular material, and is also suitable for generating a large-scale discrete element model of soil or granular material with any particle gradation.

[0034] The auxiliary geometric model combination scheme of the present invention can realize the generation and effective comparison of various particle sizes. Compared with the method of generating large-scale discrete element models in PFC discrete element software, the model generation process and calculation efficiency of the present invention are higher.

[0035] The particle refinement method of the present invention can realize the construction of a large-scale discrete element model of fine-grained soil or granular material, and can effectively solve the problem of too many particles in the large-scale discrete element model generated by fine-grained soil, resulting in low operation and calculation efficiency.

[0036] The different combination schemes of the geometric model generated by the auxiliary model layer in the present invention can effectively solve the problem that there are too many particles in the generation of a large-scale model of a granular material with extremely uneven particle size distribution, resulting in inefficient calculation.

[0037] Different combination schemes of the geometric model generated by the auxiliary model layering in the present invention can effectively solve the boundary effect problem and the particle size effect problem of the discrete element simulation calculation of the interaction between the cone and the granular material. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 A flow chart of a method for generating a large-scale discrete element model of discontinuously graded soil according to the present invention;

[0040] Figure 2 A particle gradation curve diagram involved in the method for generating a large-scale discrete element model of discontinuously graded soil according to the present invention;

[0041] Figure 3 It is a schematic diagram of a generated geometric model involved in the method for generating a large-scale discrete element model of discontinuously graded soil according to the present invention;

[0042] Figure 4 A schematic diagram of an imported geometric model involved in the method for generating a large-scale discrete element model of discontinuously graded soil according to the present invention;

[0043] Figure 5 It is a schematic diagram of the geometric body of the added central area involved in the method for generating a large-scale discrete element model of discontinuously graded soil according to the present invention;

[0044] Figure 6 It is a schematic diagram of particle refinement involved in the method for generating a large-scale discrete element model of discontinuously graded soil according to the present invention;

[0045] Figure 7 A schematic diagram of the final generated model and the total number of particles involved in the method for generating a large-scale discrete element model of discontinuously graded soil according to the present invention;

[0046] Figure 8 A comparison diagram of calculation results related to the method for generating a large-scale discrete element model of discontinuously graded soil according to the present invention;

[0047] Fig. 9 It is a flow chart of the simulation of the impact penetration model involved in the method for generating a large-scale discrete element model of discontinuously graded soil according to the present invention;

[0048] Fig.10 The cone tip resistance involved in the method for generating a large-scale discrete element model of discontinuously graded soil according to the present invention is a schematic diagram of the force component at the cone tip along the axis Z direction during the penetration process;

[0049] Fig.11 Schematic diagram of generating a discrete element model for the PFC software. DETAILED DESCRIPTION

[0050] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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.

[0051] The present invention is further explained below in conjunction with specific implementation modes.

[0052] The present embodiment provides a method for generating a large-scale discrete element model of discontinuously graded soil, which can solve the problem that when generating a large-scale model of fine-grained soil with extremely uneven particle size distribution, there are too many particles, resulting in the inability to perform efficient calculations, and can avoid boundary effects and particle size effects of cones and particles.

[0053] Taking the cone-shaped body penetrating discontinuously graded silty clay as an example, firstly, through the limit moisture content test, the curve of cone sinking depth and moisture content is obtained, so as to obtain the liquid and plastic limits of the soil and determine that the soil is silty clay. The particle gradation curve is obtained by the screening method, such as Figure 2 As shown in the results, it can be seen that the particles are very small, D 50 The D of coarse-grained soil such as sandy soil is about 40um. 50 Generally several hundred microns (D 50 Refers to the particle size corresponding to when the cumulative particle size distribution percentage of a sample reaches 50%). The curvature coefficient C is obtained by calculation C is 0.6, the non-uniformity coefficient C U The result shows that the soil particle size distribution is discontinuous and the particles are extremely uneven. The large-scale model generation process for this type of fine-grained soil is shown in Figure 1 :

[0054] S100, create soil particle material and set soil intrinsic parameters and contact model microscopic parameters

[0055] First, create a new soil particle material model, edit the material names: soil1, soil2, soil3, soil4, soil5, and set the intrinsic parameters such as Poisson's ratio, true density, shear modulus, etc. (Poisson's ratio, true density, shear modulus, etc. can be determined based on the physical property experiment of the particle material) and contact model parameters (including restitution coefficient, static friction coefficient and rolling friction coefficient) used to generate the material model of soil particles; then right-click on the material model and click Add particle adds the particle shape required for simulation (the software itself has different forms of particle models to choose from or builds a particle model through 3D software according to simulation requirements, and then imports the self-built particle model through displayTemplates-Import in the Tools-options-particledisplay option in the menu bar, and supports importing STL type files) and sets particle parameters, including the initial particle radius size and particle size distribution; for discontinuous grading particles, on the basis of particle size scaling (particle size scaling is to enlarge based on the actual particle grading distribution, and this process is completed synchronously during the generation of the model, that is, the simultaneous generation of all particles in the model), according to the particle grading test results (a conventional physical property experiment for measuring the size distribution or proportion of particles in geotechnical engineering, which needs to be completed before the simulation research work, as the basis for particle generation in the simulation work), the percentage of particles of each particle size is set by the user, and the external excel table of particle grading distribution information can be imported through the Import function in the Size Distribution module in Bulk Material-Material-particle, so as to complete the setting corresponding to the actual material particle size distribution.

[0056] S200, Generate 3D penetrometer geometry model for auxiliary model layer generation and impact simulation

[0057] Through CAD, solidworks, Rhi no and other three-dimensional modeling software, a geometric model for assisting model layer generation is established. This type of geometric model plays an isolating role for discrete element models generated by layers with different scale factors. At the same time, a 3D penetrometer geometric model for impact simulation can be constructed, as shown below Figure 3 As shown, the purpose of the geometric model required for stratification is to isolate the particles generated by different scale factors. The inner circle of the minimum diameter geometric model is made hollow and is used to generate particles in the innermost circle area by setting up a particle factory (the shape and size of the geometric model can be adjusted according to specific needs).

[0058] S300, import the geometric model used to assist model layering generation into EDEM software

[0059] Import the geometry model designed in the above step S200 through the geometry import function (Import Geometry) in the Creator Tree module in EDEM, and form different geometry model combinations according to simulation requirements; as follows Figure 4 As shown, the length unit is uniformly adjusted to mm during import, and the Poisson's ratio, density, shear modulus of the imported geometric model and the contact parameters between the geometric model and soil particles of different scale factors are set.

[0060] S400, Creating a Soil Particle Factory for Hierarchical Generative Modeling

[0061] The particle generation method of the particle factory can be selected according to specific needs. First, you can choose the dynamic generation method, which requires the total mass or total number of particles, generation speed, expected number or mass of particles generated per second, etc.; the second is the static generation method such as Volume packing, which requires the proportion of particles to be filled, the filling amount and the particle generation time. Here, taking the Volume packing generation method as an example, particle generation is divided into two parts:

[0062] First, for the innermost circle (i.e., the central cylindrical area with the smallest diameter) of the geometric model combination formed in step S300, firstly, a geometry Factory for generating particles is added through the EDEM software (specifically, by right-clicking the Geometries module in the EDEM software and selecting the AddGeometry function) to form a particle factory in the innermost circle. The shape and size of the added geometry Factory are consistent with the central area of ​​the geometric model with the smallest diameter in step (2). Then, the geometry Factory is used to generate soil particles in the innermost circle by using a volume filling particle generation method (specifically, this is achieved by the Add volume packing function of the EDEM software), as shown in FIG. Figure 5 As shown;

[0063] Second, for the area outside the innermost circle formed by different geometric model combinations according to the simulation requirements in the above step S300, the soil particles in each area are generated by directly using the volume filling particle generation method on each geometric model used for auxiliary model layer generation according to the simulation requirements (such as Figure 4 The 0.1-0 in the figure is realized by the Addvolume packing function of the EDEM software).

[0064] S500, generating particles of different scale factors in a soil particle factory

[0065] After the particle generation method is set in step S400, the soil type ratio and solid volume ratio parameters in the volume filling particle generation method are set in sequence starting from the central area with the smallest diameter in the innermost circle (the soil type selection is from the inside to the outside as soil1, soil2, soil3, soil4, soil5, and the ratio is 100%), and the particle types set in step S100 are selected, which are soi l 1 (innermost circle), soi l2, soi l 3, soi l4, soi l5... In order to reduce the total number of particles and improve the calculation efficiency, the present invention adopts a particle refinement method, that is, the particles generated in the innermost circle are the smallest, and the particle sizes of the particles generated in the outer layer area are successively larger; the purpose is to reduce the total number of particles of the overall model without affecting the calculation results, so as to improve the calculation efficiency of the large-scale discrete element model; in the model of generating larger particles in the outer layer area with very small interaction with the contact body, according to the literature and the verification of this method, a scale factor of 1.5 is used to avoid the migration of small particles to the large particle layer, that is, the particle size of the basic particles in soi l2 is soi l 1. 1.5 times of 1, taking the particle size distribution magnified 20 times as an example, the particle size of the innermost circle basic particles is 2mm, such as Figure 6 As shown, and so on.

[0066] S600, selecting different geometric models used to assist in model layer generation in step S200 to combine and generate a model for simulation

[0067] Start solving and generating discrete element models under different geometric model combinations. Set the time step for solution in the solution calculation module in EDEM. The time step for generating the model can be selected to be automatic, and the time can be 1s. If CPU calculation is selected, the grid size needs to be estimated. Click the Estimate Cell Size button in the Simulator Setting to estimate the grid size. Finally, click the Start button on the visualization interface to generate.

[0068] According to the above step S600, three discrete element models are generated for verifying the experimental effect of the model generation method of the present invention, that is, different cylinder geometric models are selected to generate different models. The generated 5C model is to use Figure 3 All the geometric models in the form of 5 particle generation areas, the model used to generate 4C Figure 3 The geometric models in (b), (c), and (d) are used to generate the 3C model. Figure 3 The geometric models in (c) and (d) are shown in the figure. The final generated model and the total number of particles are shown in Figure 7 As shown, the total number of particles is 808393, 2042028, and 4143195 respectively. If the layered generation method is not adopted, the total number of particles will be as high as more than 8 million.

[0069] S700, import the 3D penetrometer model in step (2) into the model generated in step (6) above to perform impact penetration dynamics simulation, use the same speed and simulation parameters, and output the resistance value of the cone tip for comparison, so as to select and verify the appropriate model. The calculation results are as follows: Figure 8 As shown in the figure, the model calculation effect obtained by the 4C processing method is almost the same as that of the 3C, and the maximum resistance values ​​are 599.5N and 600.2N respectively, and the error can be basically ignored. However, due to the small diameter of the middle action zone of the 5C model, the inner circle model will have a large strain under the impact force, that is, the boundary effect affects the calculation result, and the model is not suitable. By comparison, it is found that the total number of particles in the 4C model is appropriate, which meets the requirements of high-efficiency calculation and can avoid boundary effects. The 3C model has relatively too many particles. Through literature review, the rationality of the 4C model can be further proved, that is, when the width of the inner layer of soil is greater than or equal to 23 times the radius of the penetrometer, the boundary effect can be reduced to an acceptable range. At present, the inner circle width of the 3C model is 0.6m, which is close to 34 times the diameter of the penetrometer; the inner circle width of the 4C model is 0.4m, which is close to 23 times the diameter of the penetrometer. In addition, to obtain a stable cone tip resistance curve, the number of particles in contact with the cone tip should be no less than 13, such as Figure 8 As shown in (c), the total number of particles at the cone tip is 47 to 62.

[0070] The specific process of performing the impact penetration dynamics simulation in the above step S700 is as follows:

[0071] After completing the above steps S100-S700, the EDEM software has the model and basic parameters required for simulation; here the total calculation time of the impact penetration dynamics simulation is set to 0.1s, and 0.005s is used as the storage time interval for data and calculation results;

[0072] In discrete element simulation, the Rayleigh time step T R is a key number; this refers to the time it takes for the shear wave to propagate through the solid particles; since the discrete element is an explicit calculation, the step size will directly determine the update of the particle velocity and position. If the time step is too large, it will affect the result. The calculation formula is:

[0073]

[0074] In the above formula, R is the radius of the particle, ρ is the density of the particle, G is the shear modulus, and v is Poisson's ratio. This formula assumes that the relative velocity between contacting particles is very small. Except for quasi-static systems, a fraction of this maximum value is used in practice. For high coordination numbers (4 and above), a typical time step is 0.2T. R (20%) has been shown to be suitable; when the coordination number is low, 0.4T R(40%) is more appropriate; in order to make the calculation results more stable, the time step here is set to 0.1T R ;

[0075] During the calculation process, the resistance value at the cone tip and the change in the number of contacts between the cone tip and the particles during the entire penetration process are calculated simultaneously. The cone tip resistance is the force component at the cone tip along the axis Z direction during the penetration process, such as Fig.10 shown.

[0076] The number of contacts between particles at the cone tip can be calculated using the coordination number, which refers to the number of contacts between each particle and another particle at any given time point. It is an important indicator for studying the contact characteristics of the particle system and is related to the stability of the structural system. It reflects the microscopic force structure characteristics inside the soil and the looseness and density of the model during loading. The average coordination number can be expressed as:

[0077]

[0078] Where M i is the contact number of the ith particle.

[0079] The present invention can effectively determine an appropriate model generation mode and refinement method, and is used to solve the problem of large-scale discrete element model generation of discontinuous graded soil, especially fine-grained soil with extremely uneven particle distribution.

[0080] 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 method for generating a large-scale discrete element model of discontinuously graded soil, characterized in that: The main steps include: (1) Create soil particle materials and set soil intrinsic parameters and contact model microscopic parameters; (2) Create a 3D penetrometer geometry model to assist in model layer generation and impact simulation; (3) Importing the geometric model used to assist model layer generation into the EDEM software; (4) Create a soil particle factory for layered generation model; (5) Generate particles of different scale factors in a soil particle factory; (6) selecting different geometric models used to assist in model hierarchical generation in step (2) to combine and generate a model for simulation; (7) Importing the 3D penetrometer geometric model in step (2) above into the model generated in step (6) above to perform impact penetration dynamics simulation, using the same velocity and simulation parameters, and outputting the resistance value of the cone tip for comparison, so as to select and verify a suitable model.

2. The method for generating a large-scale discrete element model of discontinuously graded soil according to claim 1, characterized in that: The specific process of step (1) is as follows: first, a new soil particle material model is created, the name of the soil particle material is edited, and the intrinsic parameters and contact model parameters of the soil particle material model used to generate soil particles are set; then, the particle shape required for simulation is added to the soil particle material model and the particle parameters are set, and the particle parameters include the initial particle radius size and the particle size distribution; for discontinuously graded particles, on the basis of particle size scaling, the percentage of each particle size is set by the user according to the particle grading experimental results, so as to complete the setting corresponding to the particle size distribution of the real material.

3. The method for generating a large-scale discrete element model of discontinuously graded soil according to claim 2, characterized in that: The intrinsic parameters include Poisson's ratio, true density and shear modulus, which can be determined based on the physical property experiments of the granular materials; The contact model parameters include the coefficient of restitution, the coefficient of static friction and the coefficient of rolling friction.

4. The method for generating a large-scale discrete element model of discontinuously graded soil according to claim 2, characterized in that: The step (1) can also use the Import function of Bulk Material-Material-particle to import an external Excel spreadsheet of relevant particle gradation distribution information, thereby completing the setting corresponding to the actual material particle size distribution.

5. The method for generating a large-scale discrete element model of discontinuously graded soil according to claim 1, characterized in that: The specific process of step (2) is as follows: a geometric model for assisting model layer generation is established through three-dimensional modeling software. Such a geometric model plays an isolating role for discrete element models that use different scale factors for layer generation. At the same time, a 3D penetrometer geometric model for impact simulation is constructed.

6. The method for generating a large-scale discrete element model of discontinuously graded soil according to claim 1, characterized in that: In the step (2), when generating the geometric model, the inner circle of the geometric model with the smallest diameter is made hollow, so that particles are generated in the innermost circle area by setting up a particle factory.

7. The method for generating a large-scale discrete element model of discontinuously graded soil according to claim 1, characterized in that: The specific process of step (3) is as follows: the geometric model generated in step (2) is imported using the geometry import function of the EDEM software, and different geometric model combinations are formed according to the simulation requirements. During the import, the length unit is uniformly adjusted to mm, and the Poisson's ratio, density, shear modulus of the imported geometric model and the contact parameters between the geometric model and soil particles of different scale factors are set.

8. The method for generating a large-scale discrete element model of discontinuously graded soil according to claim 7, characterized in that: The specific process of step (4) is as follows: (4.1) For the innermost region of the geometric model combination formed in step (3), firstly, a geometric body for generating particles is added through EDEM software to form an innermost particle factory, wherein the shape and size of the added geometric body are consistent with the central region of the geometric model with the smallest diameter in step (2), and then the geometric body is subjected to a volume filling particle generation method to generate soil particles in the innermost region; (4.2) For areas outside the innermost circle where different geometric model combinations are formed according to simulation requirements, the volume filling particle generation method is used directly on each geometric model used to assist in model layer generation to generate soil particles in each area according to simulation requirements.

9. The method for generating a large-scale discrete element model of discontinuously graded soil according to claim 1, characterized in that: The specific process of step (5) is as follows: after the particle generation method is set in step (4), the soil type ratio and solid volume ratio parameters in the volume filling particle generation method are set in sequence starting from the central area with the smallest inner circle diameter; a particle refinement method is used to achieve the smallest particles generated in the innermost circle and the particle sizes of the particles generated in the outer layer area become larger in sequence.

10. The method for generating a large-scale discrete element model of discontinuously graded soil according to claim 1, characterized in that: The specific process of performing the impact penetration dynamics simulation in step (7) is as follows: The total calculation time of the impact penetration dynamics simulation is set to 0.1s, and 0.005s is used as the storage time interval of data and calculation results. In the discrete element simulation, the Rayleigh time step T R Solve by the following formula: In the above formula, R is the radius of the particle, ρ is the density of the particle, G is the shear modulus, and v is the Poisson's ratio; Rayleigh time step T R During the solution process, the changes in the resistance value at the cone tip and the number of contacts between the cone tip and the particles during the entire penetration process are solved simultaneously, where the cone tip resistance is the force component along the axis Z direction at the cone tip during the penetration process; the number of contacts between particles at the cone tip is calculated using the coordination number, which refers to the number of contacts between each particle and another particle at any given time point, and the average coordination number is expressed as: Where M i is the contact number of the ith particle.

Images