Construction method of loess particle interaction parameter library and loess collapsibility prediction method

Abstract

The present invention belongs to the technical field of geotechnical engineering geological survey, and relates to a method for constructing a parameter library of inter-particle interaction of loess and a method for predicting the collapsibility of loess. The method of the present invention reconstructs a three-dimensional structural model of original loess under different pressures and water contents, constructs a particle numerical model of original loess according to basic physical property parameters and the three-dimensional structural model of original loess, and adjusts the inter-particle interaction parameters of the particle numerical model to obtain the porosity ratio-pressure curve and pressure-collapse coefficient curve of the particle numerical model under different pressures and water contents, so that the porosity ratio-pressure curve and pressure-collapse coefficient curve corresponding to the particle numerical model and the original loess are respectively the same, and based on different depth strata, the above steps are repeated to obtain the inter-particle interaction parameters of original loess in different depth strata, construct a parameter library of inter-particle interaction of loess, and use the parameter library of inter-particle interaction of loess to achieve rapid and accurate prediction of the collapsibility coefficient of the original loess to be tested.

Description

Technical Field

[0001] The invention belongs to the technical field of geotechnical engineering geological survey, and relates to a method for constructing a parameter library of interactions between loess particles and a method for predicting loess collapsibility. Background Art

[0002] The colors of loess include gray-yellow, brown-yellow and brown-red. It is characterized by being mostly silt particles, being porous, having poor cementation and having developed vertical joints. It belongs to the Quaternary sediments. Loess is widely distributed, accounting for about one-tenth of the global land area. Influenced by the specific formation, sedimentary environment and source composition, the naturally deposited undisturbed loess has the special structural characteristics of being porous, weakly cemented and under-compacted, and is mostly in an unsaturated state. Collapsible loess is a special type of soil with relatively uniform soil quality, loose structure and developed pores. When not soaked in water, it generally has high strength and low compressibility. When soaked in water under a certain pressure, the soil structure will be rapidly destroyed, resulting in large additional subsidence and rapid decrease in strength. When carrying out engineering in loess areas, obtaining an accurate loess collapsibility coefficient is a prerequisite for ensuring the safety of engineering construction and operation. However, loess has significant spatial variability. The collapsibility coefficients of loess at different horizontal positions and burial depths in the same area may be significantly different. This feature brings great difficulties to the survey of loess collapsibility.

[0003] Studies have shown that the macroscopic collapsible behavior of loess is closely related to its microstructure. The microstructural levels and elements of loess, including particles, pores, contact relationships and cementing materials, restrict and influence each other, and essentially determine the macroscopic mechanical properties of loess, and there is a certain correspondence between them. Therefore, a deep understanding of the microstructural characteristics of loess materials plays a vital role in revealing the internal mechanism of the macroscopic performance of loess bodies. With the deepening of microstructural research and the development of advanced testing technology, the basic principle of computed tomography technology is that when X-rays penetrate the object being tested, the intensity of the X-rays decays according to a certain law as the object being tested absorbs the X-rays, and a complete CT image can be obtained by brightness imaging. The introduction of this technology makes it possible to observe the dynamic, non-destructive and three-dimensional microstructures. Many scholars use AVIZO, ImageJ and other three-dimensional visualization and analysis systems to reconstruct two-dimensional CT images into three-dimensional models, quantitatively analyze the changes in the microstructural parameters of particles and pores, such as volume, surface area, long axis direction, shape coefficient and connectivity during the loess collapse process, and thus establish the correlation between the loess microstructure and its macroscopic mechanical properties. AVIZO is an advanced three-dimensional visualization and analysis software launched by VSG, specifically for the fields of earth geology, materials science and computer-aided engineering calculations. ImageJ is an open source image processing software based on Java.

[0004] The particle discrete element numerical simulation technology can simulate the true shape of soil particles, particle size distribution, stacking structure and consolidation deformation process, and plays an important role in the study of loess macro and micro characteristics. The particle discrete element numerical simulation can decouple the influence of soil particle structure and inter-particle interaction on the macroscopic mechanical properties of soil, calibrate the common inter-particle interaction parameters of loess samples at different spatial locations in the same area, and carry out numerical simulation tests to determine the pressure-wetting coefficient curves of loess samples with different microstructural characteristics.

[0005] With the development of particle flow numerical simulation technology, numerical simulation technology based on discrete element method plays an important role in the study of loess macro and micro characteristics. Compared with physical experiments, particle flow numerical simulation experiments based on discrete element method have great advantages in studying loess-related particle size, gradation and other micro-studies. Although a large number of scholars have used discrete elements to numerically simulate physical tests such as triaxial shear and three-dimensional direct shear of loess, most simulations are performed on spherical particles. The three-dimensional porosity ratio obtained by simulation cannot be directly verified with the three-dimensional porosity ratio of the real original loess sample, resulting in insufficient prediction accuracy. At the same time, in order to use spheres to simulate the mechanical behavior of real-shaped particles, the number of inter-particle interaction parameters that need to be set increases greatly. The more complex the parameter settings, the more cumbersome the calculations, resulting in increased computational complexity.

[0006] In summary, the existing loess collapsibility prediction methods have problems of complex calculations and low prediction accuracy. Summary of the invention

[0007] The purpose of the present invention is to provide a method for constructing a parameter library of interactions between loess particles and a method for predicting the collapsibility of loess, so as to solve the technical problems of complex calculations and low prediction accuracy in existing methods for predicting the collapsibility of loess. The present invention can achieve accurate prediction of the collapsibility of loess and simplify the prediction process.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] In a first aspect, the present invention discloses a method for constructing a parameter library of loess particle interactions, comprising the following steps:

[0010] Obtain the basic physical properties and porosity-pressure curve of the original loess, obtain the pressure-wetting coefficient curve of the original loess when it is saturated with water under different pressures and the corresponding two-dimensional slice photos;

[0011] Reconstruct the 3D structural model of the original loess under different pressures and water contents based on the 2D slice photos;

[0012] According to the basic physical property parameters of the original loess and the three-dimensional structure model of the original loess, the particle numerical model of the original loess is constructed;

[0013] Adjust the inter-particle interaction parameters of the particle numerical model of the original loess to obtain the porosity ratio-pressure curve and pressure-wetting coefficient curve of the particle numerical model under different pressure and water content conditions, so that the porosity ratio-pressure curve of the particle numerical model is the same as that of the original loess, and the pressure-wetting coefficient curve of the particle numerical model is the same as that of the original loess;

[0014] Based on strata at different depths, the above steps are repeated to obtain the inter-particle interaction parameters of the original loess at different depths, and to construct a loess inter-particle interaction parameter library.

[0015] Furthermore, the basic physical property parameters include natural density, moisture content, specific gravity and particle size grading curve.

[0016] Furthermore, the particle numerical model of the original loess is constructed according to the basic physical property parameters of the original loess and the three-dimensional structure model of the original loess, as follows:

[0017] The single particles and single pores in the three-dimensional structural model of the original loess are segmented, and the segmented single particles and single pores are classified and counted according to the particle size to obtain the characteristic parameters of single particles with different particle sizes, and the local coordinates of the single particle surface point cloud are obtained;

[0018] The extended hyperellipsoid equation is used to fit the local coordinates of all single-particle surface point clouds to obtain the parameters of the single-particle extended hyperellipsoid equation;

[0019] A particle numerical model is constructed based on single particle characteristic parameters, single particle extended hyperellipsoid equation parameters and basic physical property parameters.

[0020] Furthermore, the particle numerical model is constructed according to the single particle characteristic parameters, the single particle extended hyperellipsoid equation parameters and the basic physical property parameters, as follows:

[0021] An initial particle numerical model is constructed according to single particle characteristic parameters and single particle extended hyperellipsoid equation parameters;

[0022] The basic physical property parameters of the initial particle numerical model are adjusted until the basic physical property parameters of the initial particle numerical model are consistent with the basic physical property parameters of the original loess, thereby obtaining the particle numerical model.

[0023] Furthermore, the inter-particle interaction parameters of the particle numerical model include: normal contact stiffness, tangential contact stiffness, sliding friction coefficient, rolling friction coefficient, non-contact gravity constant, non-contact gravity critical coefficient, inter-particle bond tensile strength, inter-particle bond cohesion and inter-particle bond internal friction angle.

[0024] Furthermore, the inter-particle interaction parameters of the particle numerical model of the original loess are adjusted to obtain the porosity ratio-pressure curve and pressure-wetting coefficient curve of the particle numerical model under different pressure and water content conditions, as follows:

[0025] Adjust the inter-particle interaction parameters of the particle numerical model of the original loess, and conduct a simulation test on the collapsibility test of the particle numerical model;

[0026] During the simulation test, plane wall units are set at the upper and lower ends of the particle numerical model, and the specified pressure is applied through the plane wall units. Fixed annular wall units are set around the particle numerical model. In the numerical experiment of the collapsibility test, the speed of the plane walls at the upper and lower ends of the particle numerical model is adjusted through the servo control function, so that the load acting on the upper and lower boundaries of the particle numerical model is always equal to the set axial pressure.

[0027] By changing the pressure and water content, the porosity ratio-pressure curve and pressure-wetting coefficient curve of the particle numerical model under different pressure and water content conditions are obtained.

[0028] Furthermore, in the step of adjusting the inter-particle interaction parameters of the particle numerical model of the original loess to obtain the porosity ratio-pressure curve and the pressure-wetting coefficient curve of the particle numerical model under different pressure and water content conditions, the pore size-volume fraction distribution curves of each single pore in the particle numerical model and the three-dimensional structural model of the original loess are respectively statistically analyzed, and the inter-particle interaction parameters between particles of different particle sizes are adjusted so that the particle numerical model and the three-dimensional structural model of the original loess under the same pressure and water content conditions have the same pore size distribution characteristics.

[0029] In a second aspect, the present invention discloses a method for predicting the collapsibility of loess, comprising the following steps:

[0030] Obtain the basic physical property parameters and two-dimensional slice photos of the original loess to be tested;

[0031] Reconstructing a three-dimensional structural model of the original loess to be tested according to the two-dimensional slice photos of the original loess to be tested;

[0032] A particle numerical model of the original loess to be tested is constructed according to the basic physical property parameters and the three-dimensional structure model of the original loess to be tested;

[0033] According to the loess particle interaction parameter library obtained by the construction method of the loess particle interaction parameter library, the particle interaction parameters of the particle numerical model of the original loess to be tested are set, and a numerical simulation test of the loess collapsibility coefficient test is carried out to obtain the pressure-collapse coefficient curve of the original loess to be tested, and the collapsibility coefficient of the original loess to be tested is predicted.

[0034] In a third aspect, the present invention discloses a system for constructing a parameter library of interactions between loess particles, comprising:

[0035] Parameter acquisition module: used to obtain the basic physical parameters and porosity-pressure curve of the original loess, the pressure-wetting coefficient curve of the original loess under different pressures and the corresponding two-dimensional slice photos;

[0036] 3D structural model building module: used to reconstruct the 3D structural model of the original loess under different pressures and water contents based on 2D slice photos;

[0037] Particle numerical model building module: used to build the particle numerical model of original loess according to the basic physical property parameters of original loess and the three-dimensional structure model of original loess;

[0038] Action parameter acquisition module: used to adjust the inter-particle action parameters of the particle numerical model of the original loess, obtain the porosity ratio-pressure curve and pressure-wetting coefficient curve of the particle numerical model under different pressure and water content conditions, so that the porosity ratio-pressure curve of the particle numerical model is the same as that of the original loess, and the pressure-wetting coefficient curve of the particle numerical model is the same as that of the original loess;

[0039] Database construction module: It is used to obtain the inter-particle interaction parameters of the original loess in different depths of strata based on the above modules, and to construct a loess inter-particle interaction parameter library.

[0040] In a fourth aspect, the present invention discloses a loess collapsibility prediction system, comprising:

[0041] Loess parameter acquisition module to be tested: used to obtain basic physical parameters and two-dimensional slice photos of the original loess to be tested;

[0042] The loess model construction module to be tested is used to reconstruct the three-dimensional structural model of the original loess to be tested according to the two-dimensional slice photos of the original loess to be tested;

[0043] The module for constructing the numerical model of the loess particles to be tested is used to construct the numerical model of the particles of the original loess to be tested according to the basic physical property parameters and the three-dimensional structure model of the original loess to be tested;

[0044] Collapsibility coefficient prediction module: It is used to set the inter-particle interaction parameters of the particle numerical model of the original loess to be tested according to the loess inter-particle interaction parameter library obtained by the construction method of the loess inter-particle interaction parameter library, carry out numerical simulation experiments on the loess collapsibility coefficient test, obtain the pressure-collapse coefficient curve of the original loess to be tested, and predict the collapsibility coefficient of the original loess to be tested.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] 1. The method for constructing the parameter library of inter-particle interaction of loess of the present invention ensures a comprehensive understanding of the physical properties of the original loess by obtaining the basic physical properties and porosity-pressure curve of the original loess, the pressure-wetting coefficient curve of the original loess under different pressures and the corresponding two-dimensional slice photos, and provides accurate input parameters for the subsequent three-dimensional structural reconstruction and particle numerical model construction. The three-dimensional structural model of the original loess under different pressures and water contents is reconstructed according to the two-dimensional slice photos. The three-dimensional structural model can intuitively display the internal structure of the loess, providing a more realistic geometric morphological basis for the construction of the particle numerical model. The particle numerical model of the original loess is constructed according to the basic physical properties of the original loess and the three-dimensional structural model of the original loess. The particle numerical model can simulate the mechanical response of the loess under different conditions, and provides a platform for the adjustment and optimization of the inter-particle interaction parameters. By adjusting the inter-particle interaction parameters of the particle numerical model of the original loess, the porosity ratio-pressure curve and pressure-collapse coefficient curve of the particle numerical model under different pressure and water content conditions are obtained, so that the particle numerical model has the same porosity ratio-pressure curve as the original loess, and the particle numerical model has the same pressure-collapse coefficient curve as the original loess, thereby verifying the accuracy and reliability of the model. At the same time, it also provides key data for the subsequent construction of the loess particle interaction parameter library. Based on different depth strata, the above steps are repeated to obtain the inter-particle interaction parameters of the original loess in different depth strata, and to construct the loess particle interaction parameter library. The variation law of the inter-particle interaction parameters of loess layers in different geological ages is clarified from the microscopic scale. The construction of the loess particle interaction parameter library provides rich data support for the study of loess mechanical properties. By querying the inter-particle interaction parameter database in the same area, the inter-particle interaction parameters of loess under different conditions in the area can be quickly obtained, which simplifies the prediction process of the collapsibility of other loess to be tested in the same area, reduces complex calculations, and also ensures the accuracy of the prediction of the collapsibility of the loess to be tested.

[0047] 2. The loess collapsibility prediction method of the present invention obtains the basic physical property parameters and two-dimensional slice photos of the original loess to be tested; reconstructs the three-dimensional structural model of the original loess to be tested according to the two-dimensional slice photos of the original loess to be tested; constructs the particle numerical model of the original loess to be tested according to the basic physical property parameters and the three-dimensional structural model of the original loess to be tested; sets the particle interaction parameters of the particle numerical model of the original loess to be tested according to the loess particle interaction parameter library obtained by the construction method of the loess particle interaction parameter library, carries out the numerical simulation test of the loess collapsibility coefficient test, obtains the pressure-collapse coefficient curve of the original loess to be tested, and predicts the collapsibility coefficient of the original loess to be tested. By combining the numerical simulation test with the particle interaction parameter library, the prediction process is simplified, and the accurate prediction of the loess collapsibility can be achieved.

[0048] 3. The construction system of the loess particle interaction parameter library of the present invention includes: a parameter acquisition module, a three-dimensional structure model construction module, a particle numerical model construction module, an action parameter acquisition module and a database construction module. The parameter acquisition module is used to obtain the basic physical properties and porosity-pressure curve of the original loess, and obtain the pressure-collapse coefficient curve of the original loess under saturated water under different pressures and the corresponding two-dimensional slice photos; the three-dimensional structure model construction module is used to reconstruct the three-dimensional structure model of the original loess under different pressures and water contents according to the two-dimensional slice photos; the particle numerical model construction module is used to construct the particle numerical model of the original loess according to the basic physical properties and the three-dimensional structure model of the original loess; the action parameter acquisition module is used to adjust the inter-particle action parameters of the particle numerical model of the original loess, and obtain the porosity-pressure curve and pressure-collapse coefficient curve of the particle numerical model under different pressures and water contents, so that the particle numerical model is the same as the porosity-pressure curve of the original loess, and the particle numerical model is the same as the pressure-collapse coefficient curve of the original loess; the database construction module is used to obtain the inter-particle action parameters of the original loess in different depths according to the above modules based on different depth formations, and construct a loess particle action parameter library. The various modules work together to build a database of inter-particle interaction parameters. By querying the database of inter-particle interaction parameters in the same area, the inter-particle interaction parameters of loess under different conditions in the area can be quickly obtained, which simplifies the prediction process of collapsibility of other loess to be tested in the same area, and also ensures the accuracy of the prediction of collapsibility of the loess to be tested.

[0049] 4. The loess collapsibility prediction system of the present invention comprises a module for acquiring parameters of loess to be tested, a module for constructing a model of loess to be tested, a module for constructing a numerical model of loess particles to be tested, and a module for predicting collapsibility coefficient. The module for acquiring parameters of loess to be tested is used to acquire basic physical property parameters and two-dimensional slice photos of the original loess to be tested; the module for constructing a model of loess to be tested is used to reconstruct a three-dimensional structural model of the original loess to be tested based on the two-dimensional slice photos of the original loess to be tested; the module for constructing a numerical model of loess particles to be tested is used to construct a particle numerical model of the original loess to be tested based on the basic physical property parameters and the three-dimensional structural model of the original loess to be tested; the module for predicting collapsibility coefficient is used to set the inter-particle interaction parameters of the particle numerical model of the original loess to be tested according to the loess inter-particle interaction parameter library obtained by the construction method of the loess inter-particle interaction parameter library, carry out a numerical simulation test for the loess collapsibility coefficient test, obtain the pressure-collapse coefficient curve of the original loess to be tested, and predict the collapsibility coefficient of the original loess to be tested. The various modules work together to achieve accurate prediction of loess collapsibility. The prediction process is simplified by combining numerical simulation experiments with the inter-particle interaction parameter library. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0051] Figure 2 A loess particle size distribution curve diagram of the area selected for the soil sample of the embodiment of the present invention;

[0052] Figure 3 This is a schematic diagram of XY two-dimensional slice data obtained based on CT scanning according to an embodiment of the present invention;

[0053] Figure 4 This is a schematic diagram of XZ two-dimensional slice data obtained based on CT scanning according to an embodiment of the present invention;

[0054] Figure 5 This is a schematic diagram of YZ two-dimensional slice data obtained based on CT scanning according to an embodiment of the present invention;

[0055] Figure 6 This is a characteristic parameter table for characterizing loess particles and pores according to an embodiment of the present invention;

[0056] Figure 7 This is a partial particle display diagram based on real single particle extraction according to an embodiment of the present invention;

[0057] Figure 8 A particle sphericity and particle size distribution diagram obtained based on a three-dimensional structure model of a loess soil sample according to an embodiment of the present invention;

[0058] Fig. 9 It is a flow chart of the method for constructing the parameter library of loess particle interactions of the present invention;

[0059] Fig.10 It is a flow chart of the loess collapsibility prediction method of the present invention;

[0060] Fig.11 A system module diagram for constructing a parameter library of interactions between loess particles according to the present invention;

[0061] Fig.12 This is a module diagram of the loess collapsibility prediction system of the present invention. DETAILED DESCRIPTION

[0062] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0063] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0064] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0065] See also Fig. 9 The present invention discloses a method for constructing a parameter library of loess particle interactions, comprising the following steps:

[0066] S1. Obtain the basic physical properties and porosity-pressure curve of the original loess, obtain the pressure-wetting coefficient curve of the original loess under different pressures and the corresponding two-dimensional slice photos;

[0067] Preferably, the corresponding two-dimensional slice photos include two-dimensional slice photos of original loess and two-dimensional slice photos of original loess under different pressures and saturated collapsible states. Each time the pressure is adjusted, two-dimensional slice photos are taken before pressurization, after the pressurization stabilizes, and after the saturated collapsible deformation stabilizes.

[0068] Preferably, the basic physical property parameters include natural density, moisture content, specific gravity and particle size grading curve.

[0069] S2, reconstruct the three-dimensional structural model of the original loess under different pressures and water contents based on the two-dimensional slice photos;

[0070] S3. According to the basic physical property parameters of the original loess and the three-dimensional structural model of the original loess, the particle numerical model of the original loess is constructed as follows:

[0071] The single particles and single pores in the three-dimensional structural model of the original loess are segmented, and the segmented single particles and single pores are classified and counted according to the particle size to obtain the characteristic parameters of single particles with different particle sizes, and the local coordinates of the single particle surface point cloud are obtained;

[0072] The extended hyperellipsoid equation is used to fit the local coordinates of all single-particle surface point clouds to obtain the parameters of the single-particle extended hyperellipsoid equation;

[0073] The particle numerical model is constructed based on the single particle characteristic parameters, single particle extended hyperellipsoid equation parameters and basic physical property parameters, as follows:

[0074] An initial particle numerical model is constructed according to single particle characteristic parameters and single particle extended hyperellipsoid equation parameters;

[0075] The basic physical property parameters of the initial particle numerical model are adjusted until the basic physical property parameters of the initial particle numerical model are consistent with the basic physical property parameters of the original loess, thereby obtaining the particle numerical model.

[0076] S4. Adjust the inter-particle interaction parameters of the particle numerical model of the original loess to obtain the porosity ratio-pressure curve and pressure-wetting coefficient curve of the particle numerical model under different pressure and water content conditions, so that the porosity ratio-pressure curve of the particle numerical model is the same as that of the original loess, and the pressure-wetting coefficient curve of the particle numerical model is the same as that of the original loess, as follows:

[0077] Adjust the inter-particle interaction parameters of the particle numerical model of the original loess, and conduct a simulation test on the collapsibility test of the particle numerical model;

[0078] During the simulation test, plane wall units are set at the upper and lower ends of the particle numerical model, and the specified pressure is applied through the plane wall units. Fixed annular wall units are set around the particle numerical model. In the numerical experiment of the collapsibility test, the speed of the plane walls at the upper and lower ends of the particle numerical model is adjusted through the servo control function, so that the load acting on the upper and lower boundaries of the particle numerical model is always equal to the set axial pressure.

[0079] By changing the pressure and water content, the porosity ratio-pressure curve and pressure-collapse coefficient curve of the particle numerical model under different pressure and water content conditions are obtained, so that the porosity ratio-pressure curve of the particle numerical model is the same as that of the original loess, and the pressure-collapse coefficient curve of the particle numerical model is the same as that of the original loess.

[0080] Preferably, the pore size-volume fraction distribution curves of each single pore in the particle numerical model and the three-dimensional structural model of the original loess are statistically calculated respectively, and the inter-particle interaction parameters between particles of different particle sizes are adjusted to ensure that the particle numerical model and the three-dimensional structural model of the original loess under the same pressure and moisture content conditions have the same pore size distribution characteristics.

[0081] Preferably, the inter-particle interaction parameters of the particle numerical model include: normal contact stiffness, tangential contact stiffness, sliding friction coefficient, rolling friction coefficient, non-contact gravity constant, non-contact gravity critical coefficient, inter-particle bond tensile strength, inter-particle bond cohesion and inter-particle bond internal friction angle.

[0082] S5. Based on strata at different depths, S1 to S4 are repeated to obtain the inter-particle interaction parameters of the original loess at different depths, and a loess inter-particle interaction parameter library is constructed.

[0083] See also Fig. 9 In another feasible embodiment of the present invention, the following is adaptively modified according to the situation. It includes the following steps:

[0084] S1. Obtain the basic physical properties and porosity-pressure curve of the original loess, obtain the pressure-collapse coefficient curve of the original loess under different pressures and the corresponding two-dimensional slice photos, to ensure a comprehensive understanding of the physical properties of the original loess, and provide accurate input parameters for the subsequent three-dimensional structure reconstruction and particle numerical model construction. The physical property parameters of loess are obtained through indoor experiments, and the corresponding parameters can be input during the simulation.

[0085] S2. Reconstruct the three-dimensional structural model of the original loess under different pressures and moisture contents based on the two-dimensional slice photos. The three-dimensional structural model can intuitively display the internal structure of the loess, provide a more realistic particle and pore geometry basis for the construction of the particle numerical model, and realize the extraction of the real soil particle shape.

[0086] S3. A particle numerical model of original loess is constructed based on the basic physical properties of original loess and the three-dimensional structural model of original loess. The particle numerical model can simulate the mechanical response of loess under different conditions and provide a platform for the adjustment and optimization of the interaction parameters between particles.

[0087] S4. Adjust the inter-particle interaction parameters of the particle numerical model of the original loess to obtain the porosity ratio-pressure curve and pressure-collapse coefficient curve of the particle numerical model under different pressure and moisture content conditions, so that the porosity ratio-pressure curve of the particle numerical model is the same as that of the original loess, and the pressure-collapse coefficient curve of the particle numerical model is the same as that of the original loess. By comparing the indoor test results with the simulation results, a comparative verification can be carried out, thereby verifying the accuracy and reliability of the model. At the same time, it also provides key data for the subsequent construction of a parameter library for inter-particle interaction of loess.

[0088] S5. Based on different depths of strata, repeat S1 to S4 to obtain the inter-particle interaction parameters of the original loess in different depths, and construct a loess inter-particle interaction parameter library. The variation law of inter-particle interaction parameters of loess layers in different geological ages is clarified from a microscopic scale. The construction of the loess inter-particle interaction parameter library provides rich data support for the study of loess mechanical properties. By querying the inter-particle interaction parameter database of the same area, the inter-particle interaction parameters of loess under different conditions in the area can be quickly obtained, which simplifies the prediction process of the collapsibility of other loess to be tested in the same area, and at the same time, it can also ensure the accuracy of the prediction of the collapsibility of the loess to be tested.

[0089] Based on the above method, the present invention also discloses a loess collapsibility prediction method, see Fig.10 , including the following steps:

[0090] S10, obtaining basic physical property parameters and two-dimensional slice photos of the original loess to be tested, thereby providing necessary input data for subsequent three-dimensional structure reconstruction and particle numerical model construction;

[0091] S20, reconstructing a three-dimensional structural model of the original loess to be tested according to the two-dimensional slice photos of the original loess to be tested, wherein the three-dimensional structural model provides an accurate geometric basis for the subsequent construction of a particle numerical model, so that the simulation results are closer to the actual situation;

[0092] S30, constructing a particle numerical model of the original loess to be tested according to the basic physical property parameters and the three-dimensional structure model of the original loess to be tested;

[0093] S40, according to the loess inter-particle action parameter library obtained by the construction method of the loess inter-particle action parameter library, set the inter-particle action parameters of the particle numerical model of the original loess to be tested, carry out the numerical simulation test of the loess collapsibility coefficient test, obtain the pressure-collapse coefficient curve of the original loess to be tested, and realize the prediction of the collapsibility coefficient of the original loess to be tested. By combining the numerical simulation test with the inter-particle action parameter library, the prediction process is simplified, and the accurate prediction of the loess collapsibility can be realized.

[0094] Embodiment 1:

[0095] See also Figure 1 This embodiment discloses a method for constructing a parameter library of loess particle interactions, which mainly includes: Figure 1 The steps are shown. First, the collapsibility of a few undisturbed loess samples in the area to be surveyed is measured; secondly, the particle numerical model of the undisturbed loess with known collapsibility is gradually established by the three-dimensional reconstruction technology of CT slice images. The particle numerical model has the same microstructure as the undisturbed loess. Numerical tests for collapsibility tests are carried out to calibrate the inter-particle interaction parameters of the particle numerical model before and after saturation, and a particle interaction parameter library for the undisturbed loess in the area to be surveyed is established. The particle interaction parameter library is the particle interaction parameter database; finally, for any undisturbed loess with unknown collapsibility in the area to be surveyed, a particle numerical model of the undisturbed loess to be tested is constructed, and the particle interaction parameters are set according to the established particle interaction parameter library, and a numerical simulation test for the loess collapsibility coefficient test is carried out to calculate the pressure-collapse coefficient curve of the undisturbed loess with unknown collapsibility, so as to realize the prediction of the collapsibility coefficient of the undisturbed loess to be tested. The specific steps are as follows:

[0096] Step 1: Based on the original loess in the selected area, the basic physical properties of the soil sample are first obtained through indoor mechanical experiments. The basic physical properties include the natural density of the soil , moisture content ,proportion and particle size distribution curve, see Figure 2 The following is the loess particle size distribution curve for the selected area. Secondly, the dry density of the soil is calculated based on the three-phase relationship of the soil. , initial porosity ratio At the same time, the collapsibility coefficient test was carried out using small cylindrical original loess samples with a diameter of 10mm~15mm and a height of 10mm~15mm. The single-line method was used to measure the natural moisture content of the original loess in the area when it was consolidated to stability under different loads such as 25kPa, 50kPa, 100kPa, 200kPa, 400kPa, 800kPa, etc. , to obtain the porosity ratio when the additional deformation is stable after immersion and saturation ; Calculate the collapsibility coefficient corresponding to each pressure level according to the porosity ratio-pressure curve , obtain the pressure-wetting coefficient curve. The basic physical property parameters in this step are used for the subsequent modeling of the particle numerical model, and the porosity-pressure curve and the pressure-wetting coefficient curve are used for the calibration of the inter-particle interaction parameters of the subsequent discrete element simulation test. The conventional calculation method of the wetting coefficient is as follows:

[0097]

[0098] in, The original loess sample is under load p The collapsibility coefficient under the action of represents the initial void ratio of the original loess soil sample; The original loess sample showing the natural moisture content under load p The porosity ratio when consolidating to stability is Indicates that the load p The porosity ratio of the original loess soil sample after saturation with water and additional subsidence stabilized under the action of water.

[0099] Step 2: Use CT scanning technology to obtain the microstructure data of the observation area in the small cylindrical soil sample after loading in step 1, that is, a two-dimensional slice photo. It should be noted that the CT observation area is a cylindrical area located in the center of the small cylindrical soil sample with a size of 1 mm in height and 1 mm in diameter. The resolution of the two-dimensional slice photos of the observation area taken by CT is 1.0 μm / pixel, such as Figure 3 , Figure 4 and Figure 5 shown.

[0100] Step 3: Based on the image 3D visualization analysis system, import more than 500 2D slice photos obtained by CT, and pre-process them through grayscale histogram equalization, smoothing filtering and threshold segmentation to accurately distinguish the pores and soil particles in the observation area. Finally, the pre-processed 2D slice photo sequence is 3D superimposed to construct a 3D structural model of the original loess in the observation area. According to the volume of the pores in the 3D structural model of the CT observation area With total volume Calculate the void ratio of the soil sample in the observation area , compared with the porosity ratio of the original loess sample measured in step 1, including and , verify the accuracy of the reconstructed 3D structural model. Based on the 3D structural model of the CT observation area, the 3D microstructural index system of loess can be obtained, which mainly includes the parameters of particles and pores in terms of size, morphology, orientation and connectivity. The specific parameters that can be obtained are shown in Figure 6 Among them, the porosity ratio of soil samples in the CT observation area is The calculation method is as follows:

[0101]

[0102] In the formula, is the void ratio of the three-dimensional structure model, It represents the volume of pores in the three-dimensional structural model of the CT observation area; Represents the total volume of the 3D structure model of the CT observation area.

[0103] Step 4: After constructing the three-dimensional structural model of the soil sample in the CT observation area according to step 3, segment the single particles and single pores in the three-dimensional structural model. Figure 7 As shown in the figure, different real soil particle models can be obtained after segmentation. The segmented single particles and single pores are classified and statistically analyzed according to different particle sizes to obtain the single particle characteristic parameters of different particle sizes, and the single particle characteristic parameter table is summarized. The single particle characteristic parameters include particle centroid coordinates, major axis length, median axis length, minor axis length, major axis inclination, major axis inclination, particle volume and particle surface area. The single particle characteristic parameters are used for subsequent particle discrete element modeling. Particle sphericity is a global morphological parameter that describes the degree of closeness of particles to smooth spheres. The method for obtaining it is to calculate the ratio of the spherical surface area with the same volume as the particle to the particle surface area. The calculation formula is as follows:

[0104]

[0105] In the formula, is the three-dimensional sphericity of the particle, is the surface area of ​​the spherical harmonic reconstruction particle, is the volume of the spherical harmonic reconstructed particle. The sphericity value ranges from 0 to 1. The closer the shape of the particle is to an ideal smooth sphere, the closer its sphericity value is to 1. Conversely, the closer the sphericity value is to 0, the greater the difference between the particle shape and the sphere.

[0106] like Figure 8 The particle size and sphericity distribution diagram shown in the figure is used to verify the correctness of the particle numerical model. Figure 2 The selected area loess particle size distribution curve corresponds to the selected area loess particle size distribution curve. At the same time, the coordinates of each single particle surface point cloud in the three-dimensional structure model are exported, and the local coordinates of each single particle surface point cloud are calculated with the coordinates of each particle centroid as the origin of the particle local coordinate system. The local coordinates of the single particle surface point cloud are used to subsequently calculate the extended hyperellipsoid equation parameters of each single particle. Similarly, the pore volume is measured based on the extracted single pore and pore surface area , calculate the equivalent pore diameter. The pore size-volume fraction distribution curve is statistically used for the calibration of the inter-particle interaction parameters in the discrete element simulation test. The calculation formula for the equivalent pore diameter is as follows:

[0107]

[0108] in, is the pore equivalent diameter, To measure the pore volume, is the pore surface area.

[0109] Step 5: Use the extended hyperellipsoid equation to fit the local coordinates of all single particle surface point clouds in step 4. The extended hyperellipsoid equation is given by , , , , , , ε 1 and ε 2 A total of 8 parameters describe the surface of an asymmetric ellipsoid particle in the Cartesian coordinate system. The extended hyperellipsoid equation parameters of all single particles obtained in step 4 are used to subsequently establish a particle numerical model with the real shape of the particle.

[0110]

[0111]

[0112]

[0113]

[0114] in: is the X-axis coordinate of any point on the particle surface, is the Y-axis coordinate of any point on the particle surface, is the Z-axis coordinate of any point on the particle surface, is the radius parameter of the particle on the principal axis X, is the radius parameter of the particle on the principal axis Y, is the radius parameter of the particle on the principal axis Z, for The value on the positive principal axis X, for The value on the positive principal axis Y, for The value on the positive principal axis Z, for The value on the negative principal axis X, for The value on the negative principal axis Y, for The value on the negative principal axis Z, and It is the control parameter of particle surface prism.

[0115] Step 6: Based on the particle flow numerical simulation technology, the single particle characteristic parameters of step 4 and the single particle extended hyperellipsoid equation parameters of step 5 are used to generate an initial particle numerical model in the discrete element software. The particle centroid position, major axis direction, and surface shape of the generated initial particle numerical model are consistent with the real loess particles in the CT observation area in step 2. The overall size of the generated initial particle numerical model is consistent with the size of the CT observation area in step 2, that is, the size of the reconstructed model scanned by CT is consistent with the simulated size. The basic physical properties of the initial particle numerical model should be consistent with the basic physical properties of the corresponding soil sample measured in step 1.

[0116] Furthermore, when the basic physical property parameters of the initial particle numerical model generated in step 6 are significantly different from the basic physical property parameters of the corresponding soil sample measured in step 1, the operation of adding or deleting particles of corresponding particle sizes should be performed according to the basic physical property parameters in step 1, especially the particle size grading curve, so that the basic physical property parameters of the initial particle numerical model are finally consistent with the corresponding soil sample in step 1, and the particle numerical model is obtained.

[0117] Step 7: Set conventional inter-particle interaction parameters for the real particle numerical model constructed in step 6, numerically simulate the loess collapsibility coefficient test in step 1, and obtain the porosity-pressure curve and pressure-collapse coefficient curve of the particle numerical model. The inter-particle interaction parameters of the particle numerical model of loess particles include: normal contact stiffness, tangential contact stiffness, sliding friction coefficient, rolling friction coefficient, non-contact gravity constant, non-contact gravity critical coefficient, inter-particle bond tensile strength, inter-particle bond cohesion, and inter-particle bond internal friction angle.

[0118] Preferably, in the collapsibility test simulation process in step 7, the same axial pressure as that in the indoor test in step 1 is set. The implementation method is to set plane wall units at the upper and lower ends of the particle numerical model generated in step 6 to apply a specified pressure, and set fixed annular wall units around the particle numerical model to simulate the ring knife that limits the lateral displacement of the sample in the indoor collapsibility coefficient test. In the numerical experiment of the collapsibility test, the speed of the plane walls at the upper and lower ends of the particle numerical model sample is adjusted by the servo control function, so that the load acting on the upper and lower boundaries of the particle numerical model is always equal to the set axial pressure.

[0119] Step 8: Compare the porosity ratio-pressure curve and the pressure-collapse coefficient curve obtained in step 1 and step 7, and adjust the values ​​of the inter-particle interaction parameters described in step 7 when the soil sample is in the natural moisture content and saturated water state, so that the porosity ratio-pressure curve and the pressure-collapse coefficient curve obtained by the numerical simulation test of the loess collapsibility coefficient test in step 7 are consistent with the porosity ratio-pressure curve and the pressure-collapse coefficient curve in step 1.

[0120] Furthermore, the single pores in the particle numerical model are segmented according to the method described in step 4, and the pore size-volume fraction distribution curves of each single pore in the particle numerical model are statistically analyzed, and the inter-particle interaction parameters between particles of different particle sizes are further fine-tuned, so that the particle numerical model and the three-dimensional structure model of step 4 have similar pore size distribution characteristics under the same pressure and water content conditions. So far, the inter-particle interaction parameters of loess in the survey area under the two states of natural water content and saturation have been obtained at the microscopic scale, and the collapsibility coefficient test numerical simulation test can be carried out in the future to predict the collapsibility coefficient of loess with different microscopic pore structures in the area.

[0121] Step 9: By implementing steps 1 to 8 on loess samples in the area to be surveyed, a database of inter-particle interaction parameters of undisturbed loess in different spatial locations in the area to be surveyed before and after collapse is established, and the variation patterns of inter-particle interaction parameters of loess layers in different geological ages are clarified from a microscopic scale.

[0122] Embodiment 2:

[0123] See also Figure 1The present embodiment discloses a method for predicting the collapsibility of loess. Based on the database of inter-particle interaction parameters obtained in the above embodiment 1, for any original loess with unknown collapsibility coefficient in the area, by implementing steps 2 to 7 in the embodiment 1, by taking CT two-dimensional slice photos and gradually reconstructing a particle numerical model with the same microstructure as the original loess with unknown collapsibility coefficient, according to the established database of inter-particle interaction parameters of original loess, the inter-particle interaction parameters of the particle numerical model of the original loess to be tested are set, and a numerical simulation test of the loess collapsibility coefficient test is carried out to obtain the pressure-collapse coefficient curve of the original loess with unknown collapsibility coefficient, so as to realize the prediction of the collapsibility coefficient of the original loess.

[0124] Compared with the prior art, the present invention mainly combines the three-dimensional reconstruction technology of CT slice images with the numerical simulation of particle discrete elements to establish a three-dimensional particle numerical model with the real microstructure of original loess. Based on the database of inter-particle interaction parameters of original loess before and after collapse constructed from a few original soil samples in the area to be surveyed, the inter-particle interaction parameters of the numerical model of the original loess to be tested are set, and the complete pressure-collapse coefficient curve of the original loess to be tested is obtained by carrying out numerical experiments on the loess collapsibility coefficient test.

[0125] Based on the above-mentioned loess inter-particle interaction parameter library construction method, the present invention also discloses a loess inter-particle interaction parameter library construction system, see Fig.11 ,include:

[0126] Parameter acquisition module: used to obtain the basic physical parameters and porosity-pressure curve of the original loess, the pressure-wetting coefficient curve of the original loess under different pressures and the corresponding two-dimensional slice photos;

[0127] 3D structural model building module: used to reconstruct the 3D structural model of the original loess under different pressures and water contents based on 2D slice photos;

[0128] Particle numerical model building module: used to build the particle numerical model of original loess according to the basic physical property parameters of original loess and the three-dimensional structure model of original loess;

[0129] Action parameter acquisition module: used to adjust the inter-particle action parameters of the particle numerical model of the original loess, obtain the porosity ratio-pressure curve and pressure-wetting coefficient curve of the particle numerical model under different pressure and water content conditions, so that the porosity ratio-pressure curve of the particle numerical model is the same as that of the original loess, and the pressure-wetting coefficient curve of the particle numerical model is the same as that of the original loess;

[0130] Database construction module: It is used to obtain the inter-particle interaction parameters of the original loess in different depths of strata based on the above modules, and to construct a loess inter-particle interaction parameter library.

[0131] The various modules work together to build a database of inter-particle interaction parameters. By querying the database of inter-particle interaction parameters in the same area, the inter-particle interaction parameters of loess under different conditions in the area can be quickly obtained, which simplifies the prediction process of collapsibility of other loess to be tested in the same area, and also ensures the accuracy of the prediction of collapsibility of the loess to be tested.

[0132] Based on the above loess particle interaction parameter library construction method, the present invention also discloses a loess collapsibility prediction system, see Fig.12 ,include:

[0133] Loess parameter acquisition module to be tested: used to obtain basic physical parameters and two-dimensional slice photos of the original loess to be tested;

[0134] The loess model construction module to be tested is used to reconstruct the three-dimensional structural model of the original loess to be tested according to the two-dimensional slice photos of the original loess to be tested;

[0135] The module for constructing the numerical model of the loess particles to be tested is used to construct the numerical model of the particles of the original loess to be tested according to the basic physical property parameters and the three-dimensional structure model of the original loess to be tested;

[0136] Collapsibility coefficient prediction module: It is used to set the inter-particle interaction parameters of the particle numerical model of the original loess to be tested according to the loess inter-particle interaction parameter library obtained by the construction method of the loess inter-particle interaction parameter library, carry out numerical simulation experiments on the loess collapsibility coefficient test, obtain the pressure-collapse coefficient curve of the original loess to be tested, and predict the collapsibility coefficient of the original loess to be tested.

[0137] The various modules work together to achieve accurate prediction of loess collapsibility. The prediction process is simplified by combining numerical simulation experiments with the inter-particle interaction parameter library.

[0138] The above contents are only for explaining the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A method for constructing a parameter library of interactions between loess particles, characterized in that: The following steps are involved: S1. Obtain the basic physical properties and porosity-pressure curve of the original loess, obtain the pressure-wetting coefficient curve of the original loess under different pressures and the corresponding two-dimensional slice photos; S2, reconstruct the three-dimensional structural model of the original loess under different pressures and water contents based on the two-dimensional slice photos; S3. According to the basic physical property parameters of the original loess and the three-dimensional structural model of the original loess, the particle numerical model of the original loess is constructed as follows: The single particles and single pores in the three-dimensional structural model of the original loess are segmented, and the segmented single particles and single pores are classified and counted according to the particle size to obtain the characteristic parameters of single particles with different particle sizes, and the local coordinates of the single particle surface point cloud are obtained; The extended hyperellipsoid equation is used to fit the local coordinates of all single-particle surface point clouds to obtain the parameters of the single-particle extended hyperellipsoid equation; Based on the particle flow numerical simulation technology, the initial particle numerical model is generated in the discrete element software using the single particle characteristic parameters and the single particle extended hyperellipsoid equation parameters; The basic physical property parameters of the initial particle numerical model are adjusted until the basic physical property parameters of the initial particle numerical model are consistent with the basic physical property parameters of the original loess, thereby obtaining the particle numerical model; S4, adjusting the inter-particle interaction parameters of the particle numerical model of the original loess to obtain the porosity ratio-pressure curve and the pressure-collapse coefficient curve of the particle numerical model under different pressure and water content conditions, so that the porosity ratio-pressure curve of the particle numerical model is the same as the porosity ratio-pressure curve of the original loess, and the pressure-collapse coefficient curve of the particle numerical model is the same as the pressure-collapse coefficient curve of the original loess; S5. Based on different depth strata, S1 to S4 are repeated to obtain the inter-particle interaction parameters of the original loess in different depth strata, and to construct a loess inter-particle interaction parameter library; The basic physical property parameters include natural density, moisture content, specific gravity and particle size grading curve.

2. The method for constructing a parameter library of loess particle interactions according to claim 1, characterized in that: The inter-particle interaction parameters of the particle numerical model include: normal contact stiffness, tangential contact stiffness, sliding friction coefficient, rolling friction coefficient, non-contact gravity constant, non-contact gravity critical coefficient, inter-particle bonding tensile strength, inter-particle bonding cohesion and inter-particle bonding internal friction angle.

3. The method for constructing a parameter library of loess particle interactions according to claim 1, characterized in that: The inter-particle interaction parameters of the particle numerical model of the original loess are adjusted to obtain the void ratio-pressure curve and the pressure-wetting coefficient curve of the particle numerical model under different pressure and water content conditions, as follows: Adjust the inter-particle interaction parameters of the particle numerical model of the original loess, and conduct a simulation test on the collapsibility test of the particle numerical model; During the simulation test, plane wall units are set at the upper and lower ends of the particle numerical model, and the specified pressure is applied through the plane wall units. Fixed annular wall units are set around the particle numerical model. In the numerical experiment of the collapsibility test, the speed of the plane walls at the upper and lower ends of the particle numerical model is adjusted through the servo control function, so that the load acting on the upper and lower boundaries of the particle numerical model is always equal to the set axial pressure. By changing the pressure and water content, the porosity ratio-pressure curve and pressure-wetting coefficient curve of the particle numerical model under different pressure and water content conditions are obtained.

4. The method for constructing a parameter library of loess particle interactions according to claim 1, characterized in that: In the step of adjusting the inter-particle interaction parameters of the particle numerical model of the original loess to obtain the porosity ratio-pressure curve and the pressure-wetting coefficient curve of the particle numerical model under different pressure and water content conditions, the pore size-volume fraction distribution curves of each single pore in the particle numerical model and the three-dimensional structural model of the original loess are respectively statistically analyzed, and the inter-particle interaction parameters between particles of different particle sizes are adjusted so that the particle numerical model and the three-dimensional structural model of the original loess under the same pressure and water content conditions have the same pore size distribution characteristics.

5. A method for predicting loess collapsibility, characterized in that: The following steps are involved: S10, obtaining basic physical property parameters and two-dimensional slice photos of the original loess to be tested; S20, reconstructing a three-dimensional structural model of the original loess to be tested according to the two-dimensional slice photos of the original loess to be tested; S30, constructing a particle numerical model of the original loess to be tested according to the basic physical property parameters and the three-dimensional structure model of the original loess to be tested; S40. According to the loess inter-particle interaction parameter library obtained by the method for constructing the loess inter-particle interaction parameter library described in any one of claims 1 to 4, the inter-particle interaction parameters of the particle numerical model of the original loess to be tested are set, a numerical simulation test of the loess collapsibility coefficient test is carried out, the pressure-collapse coefficient curve of the original loess to be tested is obtained, and the collapsibility coefficient of the original loess to be tested is predicted.

6. A system for constructing a parameter library of interactions between loess particles, characterized in that: include: Parameter acquisition module: used to obtain the basic physical parameters and porosity-pressure curve of the original loess, the pressure-wetting coefficient curve of the original loess under different pressures and the corresponding two-dimensional slice photos; The basic physical property parameters include natural density, moisture content, specific gravity and particle size distribution curve; 3D structural model building module: used to reconstruct the 3D structural model of the original loess under different pressures and water contents based on 2D slice photos; Particle numerical model construction module: It is used to construct the particle numerical model of the original loess according to the basic physical properties of the original loess and the three-dimensional structure model of the original loess, as follows: The single particles and single pores in the three-dimensional structural model of the original loess are segmented, and the segmented single particles and single pores are classified and counted according to the particle size to obtain the characteristic parameters of single particles with different particle sizes, and the local coordinates of the single particle surface point cloud are obtained; The extended hyperellipsoid equation is used to fit the local coordinates of all single-particle surface point clouds to obtain the parameters of the single-particle extended hyperellipsoid equation; Based on the particle flow numerical simulation technology, the initial particle numerical model is generated in the discrete element software using the single particle characteristic parameters and the single particle extended hyperellipsoid equation parameters; The basic physical property parameters of the initial particle numerical model are adjusted until the basic physical property parameters of the initial particle numerical model are consistent with the basic physical property parameters of the original loess, thereby obtaining the particle numerical model; Action parameter acquisition module: used to adjust the inter-particle action parameters of the particle numerical model of the original loess, and obtain the porosity ratio-pressure curve and pressure-wetting coefficient curve of the particle numerical model under different pressure and water content conditions, so that the porosity ratio-pressure curve of the particle numerical model is the same as the porosity ratio-pressure curve of the original loess, and the pressure-wetting coefficient curve of the particle numerical model is the same as the pressure-wetting coefficient curve of the original loess; Database construction module: It is used to obtain the inter-particle interaction parameters of the original loess in different depths of strata based on the above modules, and to construct a loess inter-particle interaction parameter library.

7. Loess collapsibility prediction system, characterized by: include: Loess parameter acquisition module to be tested: used to obtain basic physical parameters and two-dimensional slice photos of the original loess to be tested; The loess model construction module to be tested is used to reconstruct the three-dimensional structural model of the original loess to be tested according to the two-dimensional slice photos of the original loess to be tested; The module for constructing the numerical model of the loess particles to be tested is used to construct the numerical model of the particles of the original loess to be tested according to the basic physical property parameters and the three-dimensional structure model of the original loess to be tested; Collapsibility coefficient prediction module: used to obtain the loess particle interaction parameter library according to the construction method of the loess particle interaction parameter library described in any one of claims 1 to 4, set the particle interaction parameters of the particle numerical model of the original loess to be tested, carry out numerical simulation experiments on the loess collapsibility coefficient test, obtain the pressure-collapse coefficient curve of the original loess to be tested, and predict the collapsibility coefficient of the original loess to be tested.

Images