A method for preparing a coarse and fine particle inclusion photoelastic sample and analyzing a contact force chain network
By using the discrete element method and photoelastic image processing, photoelastic samples with consistent initial density were prepared and a contact force network was constructed. This solved the problem of insufficient initial state control and analysis in the preparation of photoelastic samples with coarse and fine inclusions, and achieved efficient sample preparation and quantitative analysis of micromechanical characteristics.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHENZHEN UNIV
- Filing Date
- 2026-04-13
- Publication Date
- 2026-06-26
Smart Images
Figure CN122016663B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of mechanical testing of particulate materials, and more specifically to a method for preparing photoelastic samples with inclusions of coarse and fine particles and for analyzing contact force chain networks. Background Technology
[0002] Particulate materials are widely used in many engineering fields such as geotechnical engineering, geological disaster prevention and control, and powder metallurgy. The evolution characteristics of their internal force chain network directly determine the overall mechanical response behavior of the material. As a non-contact measurement method that can intuitively display the distribution of contact forces between particles, photoelasticity experiments have been widely used in the micromechanical research of two-dimensional particulate systems. In order to simulate the gradation characteristics commonly found in actual engineering materials, coarse and fine mixed photoelastic samples are often used to carry out relevant experimental research.
[0003] However, in the existing preparation process of coarse and fine mixed photoelastic samples, it is difficult to effectively control the initial compaction state of samples with different coarse particle contents. Since the change in coarse particle content will significantly affect the packing characteristics of the particle system, when preparing samples with different coarse particle contents using conventional sample laying or layered sample loading methods, each sample often has different initial void ratios and relative compaction. This difference in initial state will directly interfere with the formation of contact relationships between particles, thereby affecting the distribution characteristics of the influence chain network. As a result, the test results are difficult to truly reflect the influence of changes in coarse particle content on force transmission characteristics, reducing the comparability and repeatability of test results between different samples.
[0004] On the other hand, existing methods for preparing photoelastic samples generally suffer from cumbersome operation and low efficiency. For different experimental requirements with different coarse particle contents, multiple independent sample laying or loading operations are usually required. Each sample preparation requires repeating the steps of particle screening, spreading, and leveling, which not only consumes a lot of time, but also makes it difficult to ensure the consistency of particle arrangement in different samples during the preparation process. This inefficient sample preparation method limits the development of multiple sets of comparative experiments with different coarse particle contents and restricts the in-depth study of the mechanical behavior of coarse and fine mixed particle systems.
[0005] Furthermore, existing contact force analysis methods for photoelastic images are mostly focused on single-size or nearly homogeneous particle systems, with insufficient attention paid to the unique microstructure in systems with coarse and fine inclusions. Conventional methods mainly extract interparticle contact forces and construct contact force networks based on photoelastic images, but lack systematic analysis of the topology of strong contact force networks. It is difficult to quantify the connectivity, directionality, and critical conditions for force chains under different coarse particle contents. In particular, for the load-bearing and force-transferring structures formed by fine particles between adjacent coarse particles, existing technologies have not yet established effective identification and quantification methods, resulting in an insufficient understanding of the mechanical contributions of fine particles in coarse and fine inclusion systems.
[0006] Therefore, how to design a method for preparing photoelastic samples with inclusions of coarse and fine particles and analyzing the contact force chain network, which can effectively control the consistency of the initial state of different samples, improve the sample preparation efficiency, and perform systematic quantitative analysis of the particle contact force network, so as to achieve a reliable comparative study of force transmission characteristics under different coarse particle contents, is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention
[0007] In view of this, the present invention provides a method for preparing photoelastic samples with coarse and fine particle inclusions and analyzing contact force chain networks. It aims to solve the problems in the prior art, such as the difficulty in maintaining consistent initial compactness of samples with different coarse particle contents, low sample preparation efficiency, and lack of systematic quantitative analysis of contact force networks and microstructures in coarse and fine particle inclusion systems. This enables a reliable comparative study of force transmission characteristics under different coarse particle contents.
[0008] To achieve the above objectives, the present invention adopts the following technical solution:
[0009] A method for preparing a photoelastic sample with inclusions of coarse and fine particles and for analyzing its contact force chain network includes the following steps:
[0010] S1. Calculate the target porosity for different coarse particle contents using the discrete element method;
[0011] S2. Generate corresponding numerical calculation samples based on the target porosity and preset coarse particle content, and derive relevant parameters to prepare coarse and fine particle-in-place photoelastic samples with the same initial density.
[0012] S3. Combining the unpolarized and polarized photoelastic images of the photoelastic sample, extract the interparticle contact force and construct a contact force network;
[0013] S4. Based on the contact force network, extract the strong contact force network and perform microscopic analysis to obtain the comparison results of force transmission characteristics under different coarse grain content conditions.
[0014] Preferably, S1 includes:
[0015] The maximum void ratio corresponding to the loosest packing state of the particle system with different coarse particle contents was obtained by discrete element numerical simulation. The minimum void ratio corresponding to the densest packing state ;
[0016] Based on the preset relative density Calculate the target porosity .
[0017] Preferably, S2 includes:
[0018] Based on the target porosity and preset coarse particle content, a corresponding numerical calculation sample is generated, and a two-dimensional photoelastic sample is prepared according to the number of coarse and fine particles in the sample.
[0019] A progressive sample preparation strategy is used to construct samples with different coarse particle contents. This includes first preparing a photoelastic sample with a high coarse particle content as the initial structure; when preparing a photoelastic sample with a low coarse particle content, some coarse particles are gradually removed from the initial structure, the remaining coarse particles are disturbed, and a corresponding number of fine particles are added to obtain the target coarse particle content sample.
[0020] Preferably, S3 includes:
[0021] S31. Identify the particle outlines in the unpolarized image and obtain the center coordinates and radius of each particle;
[0022] S32. Identify mutually contacting particle pairs according to preset contact judgment criteria and establish particle contact relationships;
[0023] S33. Extract the corresponding interparticle contact force based on polarized photoelastic images;
[0024] S34. Treat particles as network nodes, and the contact between particles as network edges. Use the extracted contact force magnitude as the edge weight to construct a contact force network.
[0025] Preferably, in step S32, the preset contact determination criterion is: for any particle With particles If the distance between the centers of the two particles satisfies Then Let them be contact pairs, where and Particles and radius, This is the amount of compensation for image recognition errors.
[0026] Preferably, S33 includes:
[0027] The polarized photoelastic image is converted to grayscale, the gradient magnitude of pixels within a single particle is calculated, and the photoelastic response of the particle is defined. It is the integral average of the squared internal gradients;
[0028] Establishing the photoelastic response of particles based on single-particle calibration experiments Empirical relationship between the average contact force F and the mean contact force F Calculate the average contact force borne by each particle. , where a and b are the fitting parameters;
[0029] Based on the relative magnitudes of the photoelastic response intensity at each contact interface, the average contact force borne by each particle is calculated. The initial estimated values of the contact forces at each contact point are obtained by assigning them to the corresponding particle contact locations. , i and j represent particle numbers;
[0030] A particle contact force inversion model was constructed, and the corresponding photoelastic response image was generated through the theoretical contact force distribution. The image was then compared with the polarized photoelastic image obtained from the experiment, and a residual function between the observed image and the theoretical image was established.
[0031] The contact force parameters are optimized using a nonlinear least squares algorithm to minimize the residual function, thereby obtaining the contact force distribution that best matches the experimental photoelastic image and satisfies the particle force equilibrium condition. .
[0032] Preferably, in step S4, the mesoscopic analysis includes topological analysis, comprising:
[0033] Identify the largest connected cluster in a strong contact force network, count the proportion of its number of nodes to the total number of nodes in the strong contact force network, and obtain the relative size of the largest connected cluster;
[0034] Based on the direction of the particle contact lines in the strong contact force network, the number of contacts in each directional interval is counted to obtain the distribution characteristics of the contact direction.
[0035] Preferably, in step S4, the microscopic analysis further includes percolation characteristic analysis, including:
[0036] Gradually adjust the contact force threshold Strong contact subnetworks were constructed under different threshold conditions, where This is a dimensionless parameter used to screen for contact forces greater than [a certain value]. Contact relationship with multiples of average contact force;
[0037] By statistically analyzing the relative sizes of the largest connected clusters at each threshold, the critical threshold for a strong contact force network to transition from a non-connected state to a connected state is determined, serving as the force chain percolation threshold. The connected state is when the particle set in the largest connected cluster is simultaneously in contact with both sides of the sample boundary.
[0038] Preferably, in step S4, the microscopic analysis further includes matrix bridge structure analysis, including:
[0039] A set of fine particles located between adjacent coarse particles, consisting of multiple fine particles, and whose contact relationship is strong, is defined as a matrix bridge structure.
[0040] The number, spatial distribution, or connectivity scale of the matrix bridge structures are statistically analyzed.
[0041] As can be seen from the above technical solution, compared with the prior art, the technical solution of the present invention has the following beneficial effects:
[0042] 1. This method uses the discrete element method to determine the target void ratio that satisfies the preset relative density under different coarse particle contents, and prepares photoelastic samples based on this. This ensures that samples with different coarse particle contents have the same initial relative density before the experiment, effectively avoiding random interference caused by differences in particle arrangement and pore structure, and improving the comparability and repeatability of test results with different coarse particle contents.
[0043] 2. When preparing samples with different coarse particle content, a high coarse particle content photoelastic sample is prepared first, and then the target coarse particle content sample is quickly obtained by gradually removing some coarse particles, disturbing the remaining coarse particles and supplementing the corresponding number of fine particles. This eliminates the need for repeated sample laying or overall sample reconstruction, simplifies the sample preparation process, and improves the preparation efficiency of multiple sets of samples with different coarse particle content.
[0044] 3. Based on photoelastic images, interparticle contact forces are extracted and a contact force network is constructed. Then, strong contact force networks are extracted and microscopic analysis is performed, including topological structure analysis, percolation characteristic analysis, and matrix bridge structure analysis. It can quantify the connectivity, directionality, and critical conditions of force chain structures under different coarse particle contents, and can identify the matrix bridge structure formed by fine particles between coarse particle skeletons, providing a systematic analytical means to reveal the force transmission mechanism in coarse and fine mixed particle systems. Attached Figure Description
[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.
[0046] Figure 1 A flowchart illustrating a method for preparing a photoelastic sample with inclusions of coarse and fine particles and analyzing the contact force chain network, provided in an embodiment of the present invention.
[0047] Figure 2 This is a schematic diagram showing the maximum, minimum, and target porosity of samples at various coarse particle contents provided in the embodiments of the present invention.
[0048] Figure 3 This is a schematic diagram of the discrete element sample and the photoelastic sample provided in the embodiments of the present invention;
[0049] Figure 4 A schematic diagram of strong contact network particles provided in an embodiment of the present invention;
[0050] Figure 5This is a schematic diagram illustrating the variation of the maximum cluster relative size with a threshold under different coarse grain contents, provided by an embodiment of the present invention.
[0051] Figure 6 This is a schematic diagram showing the distribution of the number of contacts in each direction under different thresholds of 49.1% coarse particle content, provided in an embodiment of the present invention.
[0052] Figure 7 This is a schematic diagram illustrating the change of percolation threshold with coarse particle content, provided in an embodiment of the present invention. Detailed Implementation
[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0054] like Figure 1 As shown, this embodiment provides a method for preparing a photoelastic sample with inclusions of coarse and fine particles and for analyzing the contact force chain network, including the following steps:
[0055] S1. Calculate the target porosity under different coarse particle contents using the discrete element method;
[0056] S2. Generate a corresponding numerical calculation sample based on the target porosity and preset coarse particle content, and derive relevant parameters to prepare a photoelastic sample with coarse and fine particles of the same initial density.
[0057] S3. Combining the unpolarized and polarized photoelastic images of the photoelastic sample, extract the interparticle contact force and construct a contact force network;
[0058] S4. Based on the contact force network, extract the strong contact force network and perform microscopic analysis to obtain the comparison results of force transmission characteristics under different coarse grain content conditions.
[0059] This method achieves consistent control of the initial relative density of samples with different coarse-grained contents through the discrete element method. It combines photoelastic image processing to extract interparticle contact forces and construct a contact force network. Based on the topological structure analysis, permeability analysis, and matrix bridge structure identification of the strong contact force network, it can quantify the connectivity, directionality, and critical conditions of the force chain network under different coarse-grained contents. This reveals the influence of changes in coarse-grained content on the internal force transmission path and the evolution of the load-bearing skeleton of the coarse-fine particle mixed system, providing experimental evidence at the micromechanical level for strength prediction and deformation analysis of granular materials such as coarse-grained soil and riprap in geotechnical engineering.
[0060] The following provides a further explanation of each step and related features in the above method;
[0061] In this embodiment S1, the target porosity under different coarse particle contents is calculated by the discrete element method;
[0062] In this step, the discrete element model is used to perform numerical simulations for multiple preset coarse particle contents. By changing the particle friction coefficient during the sample preparation process, the loosest packing state is simulated to obtain the maximum porosity, and the densest packing state is simulated to obtain the minimum porosity.
[0063] like Figure 2 The figure shows the calibration results of the porosity of samples under different coarse particle contents. The figure presents the maximum porosity, minimum porosity, and target porosity calculated based on a preset relative density for each coarse particle content. Calculate the target porosity The target porosity is the porosity control target that needs to be achieved during the subsequent physical sample preparation, ensuring that samples with different coarse particle contents have the same initial relative density before the experiment begins.
[0064] In this embodiment S2, a numerical calculation sample is generated based on the target porosity and preset coarse particle content, and relevant parameters are derived to prepare a photoelastic sample with coarse and fine particle inclusions of the same initial density; including:
[0065] Numerical calculation samples are generated based on the target porosity and preset coarse particle content. Two-dimensional photoelastic samples are then prepared according to the number of coarse and fine particles in the samples. Figure 3 As shown, Figure 3 In the image, 'a' represents the generated discrete element sample diagram, which clearly shows the spatial arrangement of the particles. Figure 3 In the diagram, b represents the corresponding photoelastic sample that has been prepared. Both samples are consistent in terms of particle number and porosity, which verifies the accurate guiding role of the numerical model in the preparation of physical samples.
[0066] Further preparation of photoelastic samples with inclusions of coarse and fine particles includes:
[0067] First, a photoelastic sample with high coarse particle content (49.1%) was prepared. When preparing photoelastic samples with low coarse particle content (9.4%, 20.3%, 30.7%, 40.6%), some coarse particles were gradually removed from the high coarse particle content photoelastic sample, the remaining coarse particles were disturbed, and a corresponding number of fine particles were added to obtain the target coarse particle content sample.
[0068] This step, through a progressive sample preparation strategy from high to low, combined with precise control of the target porosity, achieves uniformity in the initial relative density of samples with different coarse particle contents, providing high-quality comparative samples for subsequent loading tests and improving the repeatability and comparability of multiple control tests.
[0069] In this embodiment S3, combining the unpolarized and polarized photoelastic images of the photoelastic sample, the interparticle contact force is extracted and a contact force network is constructed; including:
[0070] S31. Identify the particle outlines in the unpolarized image and obtain the center coordinates and radius of each particle;
[0071] S32. Identify mutually contacting particle pairs according to preset contact judgment criteria and establish particle contact relationships;
[0072] S33. Extract the corresponding interparticle contact force based on polarized photoelastic images;
[0073] S34. Treat particles as network nodes, and the contact between particles as network edges. Use the extracted contact force magnitude as the edge weight to construct a contact force network.
[0074] The construction of a strong contact force network includes: classifying the contact relationships in the contact force network into strong and weak categories based on the average contact force as a threshold; determining contact with a contact force greater than the average contact force as a strong contact; and constructing a strong contact force network by all strong contacts and their corresponding particle nodes.
[0075] Furthermore, the preset contact determination criterion in S32 is: for any particle With particles If the distance between the centers of the two particles satisfies Then Let them be contact pairs, where and Particles and radius, This is the amount of compensation for image recognition errors.
[0076] Furthermore, S33 includes:
[0077] The polarized photoelastic image is converted to grayscale, the gradient magnitude of pixels within a single particle is calculated, and the photoelastic response of the particle is defined. It is the integral average of the squared internal gradients;
[0078] Establishing the photoelastic response of particles based on single-particle calibration experiments Empirical relationship between the average contact force F and the mean contact force F Calculate the average contact force borne by each particle. , where a and b are the fitting parameters;
[0079] Based on the relative magnitudes of the photoelastic response intensity at each contact interface, the average contact force borne by each particle is calculated. The initial estimated values of the contact forces at each contact point are obtained by assigning them to the corresponding particle contact locations. , i and j represent particle numbers;
[0080] The initial estimate here Represented as:
[0081]
[0082] in, This represents the initial estimate of the contact force between particle i and particle j, where i is the number of the particle currently under consideration, and j is the number of a particle in contact with particle i. The average contact force borne by particle i. Let be the local photoelastic response intensity at the interface between particle i and particle j, s be the interface, and k be the number of particles in contact with particle i.
[0083] A particle contact force inversion model was constructed, and the corresponding photoelastic response image was generated through the theoretical contact force distribution. The image was then compared with the polarized photoelastic image obtained from the experiment, and a residual function between the observed image and the theoretical image was established.
[0084] The contact force parameters are optimized using a nonlinear least squares algorithm to minimize the residual function, thereby obtaining the contact force distribution that best matches the experimental photoelastic image and satisfies the particle force equilibrium condition. .
[0085] In step S33, the residual function can be defined as the sum of squares of the differences in the corresponding pixel grayscale values between the theoretically generated photoelastic response image and the experimentally obtained polarized photoelastic image. By iteratively optimizing the contact force parameters, the theoretical image gradually approaches the experimental image, thereby achieving accurate inversion of the contact force. The optimization problem is solved by combining a nonlinear least squares algorithm. In each iteration, the contact force estimate is updated according to the gradient of the residual function until it converges to the contact force distribution that minimizes the residual and satisfies the particle force equilibrium condition.
[0086] By constructing a photoelastic image inversion model and combining it with nonlinear least squares optimization, it can accurately extract the contact force distribution between particles from experimental images. This overcomes the limitations of traditional methods that rely solely on qualitative analysis of photoelastic fringes. The obtained contact force data not only satisfies the force equilibrium condition of particles, but also provides a reliable quantitative basis for subsequent construction of contact force networks and force chain characteristic analysis.
[0087] In this embodiment S4, based on the contact force network, a strong contact force network is extracted and microscopic analysis is performed to obtain a comparison of force transmission characteristics under different coarse particle content conditions;
[0088] like Figure 4As shown, the average contact force of the entire contact force network is used as the threshold. Contact relationships with edge weights greater than the threshold are identified as strong contacts. All strong contacts and their connected nodes are extracted to construct a strong contact force network. The dark particles and their connecting lines in the figure are the strong contact force network, which is used to characterize the main load transfer paths in the particle system.
[0089] The mesoscopic analysis includes topological analysis, which includes:
[0090] Identify the largest connected cluster in a strong contact force network, count the proportion of its number of nodes to the total number of nodes in the strong contact force network, and obtain the relative size of the largest connected cluster;
[0091] like Figure 5 The figures show different coarse particle contents (0%, 9.4%, 20.3%, 30.7%, 40.6%, 49.1%).
[0092] The curve of the relative size of the maximum connected cluster as a function of the contact force threshold is used to identify the maximum connected cluster in the strong contact force network. The proportion of its node number to the total number of nodes in the strong contact force network is counted to obtain the relative size of the maximum connected cluster. By comparing the variation law of the relative size of the maximum connected cluster under different coarse grain content conditions, the difference in the connectivity of the force chain structure is quantified.
[0093] It also includes counting the number of contacts in each directional interval based on the direction of the particle contact connection lines in the strong contact force network, and obtaining the distribution characteristics of the contact direction;
[0094] like Figure 6 As shown, the distribution rose diagrams of strong contact directions under different thresholds (ξ=0, ξ=0.5, ξ=1.0) with a coarse particle content of 49.1% are presented. Based on the direction of the particle contact lines in the strong contact force network, the number of contacts in each directional interval is counted, and the contact direction rose diagram is drawn. This diagram can intuitively reflect the degree of concentration of the force chain in different directions and its anisotropic characteristics, revealing the influence of changes in coarse particle content on the directionality of force transmission.
[0095] Further microscopic analysis also includes percolation characteristic analysis, including:
[0096] Gradually adjust the contact force threshold Strong contact subnetworks were constructed under different threshold conditions, where This is a dimensionless parameter used to screen for contact forces greater than [a certain value]. Contact relationship with multiples of average contact force;
[0097] By statistically analyzing the relative sizes of the largest connected clusters at each threshold, the critical threshold for a strong contact force network to transition from a non-connected state to a connected state is determined, serving as the force chain percolation threshold. The described interconnected state is when the particle collection in the largest interconnected cluster is simultaneously in contact with both sides of the sample boundary.
[0098] like Figure 7 As shown, the force chain percolation threshold is the value at different coarse particle contents. The trend graph shows that, in the percolation characteristic analysis, the contact force threshold ξ (a dimensionless parameter defined as the ratio of contact force to average contact force) used to define strong contact is gradually adjusted. Strong contact subnetworks under different threshold conditions are constructed, and the relative size of the maximum connected clusters under each threshold is statistically analyzed. The changes in the force chain are used to identify the critical threshold at which a strong contact force network transitions from a non-connected state to a connected state, serving as the force chain percolation threshold. By comparing the force chain percolation thresholds of samples with different coarse particle contents under the same loading conditions, the influence of coarse particle content on the connectivity of the force chain network was quantified.
[0099] Further microscopic analysis also includes matrix bridging structure analysis, including:
[0100] In a strong contact force network, the matrix bridge structure refers to a collection of fine particles consisting of one or more fine particles located between two adjacent coarse particles, and the contact relationship between these fine particles and the adjacent coarse particles or between the fine particles is a strong contact.
[0101] The analysis of this structure mainly includes: counting the number of matrix bridges in the sample, evaluating their spatial distribution characteristics, and measuring their connectivity scale in the loading direction; through the above statistics and measurements, the ability of fine particles to transfer loads between coarse particle skeletons can be quantified, revealing the mechanical contribution of fine particles in the coarse-fine mixed particle system.
[0102] This step, through topological analysis, percolation characteristic analysis, and matrix bridge structure identification, quantifies the evolution of force chain networks under different coarse particle contents from multiple dimensions such as connectivity, directionality, critical conditions for penetration, and the bearing effect of fine particles. It provides systematic micromechanical evidence for a deeper understanding of the internal force transmission mechanism of coarse and fine particle mixed systems.
[0103] This embodiment provides a method for preparing photoelastic samples with coarse and fine particle inclusions and analyzing contact force chain networks. This method can accurately control the initial compaction state of samples with different coarse particle contents, efficiently prepare multiple sets of comparative samples, and systematically reveal the influence of changes in coarse particle content on the force chain structure through photoelastic image processing and contact force network analysis. This method can be widely applied to the study of mechanical properties of coarse-grained soil, riprap, ballast materials, etc. in geotechnical engineering, providing reliable micromechanical experimental basis for gradation optimization, strength prediction, and deformation analysis in engineering design. It can also be extended to the field of force chain research of other particulate materials.
[0104] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.
[0105] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for preparing photoelastic samples with inclusions of coarse and fine particles and for analyzing contact force chain networks, characterized in that, Includes the following steps: S1. Calculate the target void ratio for different coarse particle contents using the discrete element method; including: The maximum void ratio corresponding to the loosest packing state of the particle system with different coarse particle contents was obtained by discrete element numerical simulation. The minimum void ratio corresponding to the densest packing state According to the preset relative density Calculate the target porosity ; S2. Generate a corresponding numerical calculation sample based on the target porosity and preset coarse particle content, and derive relevant parameters to prepare a photoelastic sample with coarse and fine particle inclusions of the same initial density; including: Numerical calculation samples are generated based on the target porosity and preset coarse particle content. Two-dimensional photoelastic samples are prepared according to the number of coarse and fine particles in the sample. Samples with different coarse particle contents are constructed through a progressive sample preparation strategy, including first preparing a high coarse particle content photoelastic sample as the initial structure. When preparing a low coarse particle content photoelastic sample, some coarse particles are gradually removed from the initial structure, the remaining coarse particles are disturbed, and a corresponding number of fine particles are added to obtain the target coarse particle content sample. S3. Combining the unpolarized and polarized photoelastic images of the photoelastic sample, extract the interparticle contact force and construct a contact force network; S4. Based on the aforementioned contact force network, a strong contact force network is extracted and subjected to mesoscopic analysis to obtain comparative results of force transmission characteristics under different coarse-grained content conditions; wherein, the mesoscopic analysis includes topological structure analysis, including: Identify the largest connected cluster in the strong contact force network, count the proportion of its node count to the total number of nodes in the strong contact force network, and obtain the relative size of the largest connected cluster; based on the direction of the particle contact connection lines in the strong contact force network, count the number of contacts in each directional interval to obtain the distribution characteristics of the contact direction. Microscopic analysis also includes percolation characteristic analysis, which also includes: Gradually adjust the contact force threshold Strong contact subnetworks were constructed under different threshold conditions, where This is a dimensionless parameter used to screen for contact forces greater than [a certain value]. The contact relationship of the average contact force was analyzed; the relative size of the largest connected clusters under each threshold was statistically analyzed to determine the critical threshold for the strong contact force network to transform from a non-connected state to a connected state, which was then used as the force chain percolation threshold. The connected state is when the particle set in the largest connected cluster is simultaneously in contact with both sides of the sample boundary.
2. The method for preparing a photoelastic sample with inclusions of coarse and fine particles and analyzing the contact force chain network according to claim 1, characterized in that, S3 includes: S31. Identify the particle outlines in the unpolarized image and obtain the center coordinates and radius of each particle; S32. Identify mutually contacting particle pairs according to preset contact judgment criteria and establish particle contact relationships; S33. Extract the corresponding interparticle contact force based on polarized photoelastic images; S34. Treat particles as network nodes, and the contact between particles as network edges. Use the extracted contact force magnitude as the edge weight to construct a contact force network.
3. The method for preparing a photoelastic sample with inclusions of coarse and fine particles and analyzing the contact force chain network according to claim 2, characterized in that, In step S32, the preset contact determination criterion is: for any particle With particles If the distance between the centers of the two particles satisfies Then Let them be contact pairs, where and Particles and radius, This is the amount of compensation for image recognition errors.
4. The method for preparing a photoelastic sample with inclusions of coarse and fine particles and analyzing the contact force chain network according to claim 2, characterized in that, S33 includes: The polarized photoelastic image is converted to grayscale, the gradient magnitude of pixels within a single particle is calculated, and the photoelastic response of the particle is defined. It is the integral average of the squared internal gradients; Establishing the photoelastic response of particles based on single-particle calibration experiments Empirical relationship between the average contact force F and the mean contact force F Calculate the average contact force borne by each particle. , where a and b are the fitting parameters; Based on the relative magnitudes of the photoelastic response intensity at each contact interface, the average contact force borne by each particle is calculated. The initial estimated values of the contact forces at each contact point are obtained by assigning them to the corresponding particle contact locations. , i and j represent particle numbers; A particle contact force inversion model was constructed, and the corresponding photoelastic response image was generated through the theoretical contact force distribution. The image was then compared with the polarized photoelastic image obtained from the experiment, and a residual function between the observed image and the theoretical image was established. The contact force parameters are optimized using a nonlinear least squares algorithm to minimize the residual function, thereby obtaining the contact force distribution that best matches the experimental photoelastic image and satisfies the particle force equilibrium condition. .
5. The method for preparing a photoelastic sample with inclusions of coarse and fine particles and analyzing the contact force chain network according to claim 1, characterized in that, In S4, the mesoscopic analysis also includes matrix bridge structure analysis, including: A set of fine particles located between adjacent coarse particles, consisting of multiple fine particles, and whose contact relationship is strong, is defined as a matrix bridge structure. The number, spatial distribution, or connectivity scale of the matrix bridge structures are statistically analyzed.