A method for determining the "elastic-fabric-void" structure of a pavement mixture

Abstract

The application provides a method for determining the "elastic-skeleton-void" structure of pavement mixture, and belongs to the technical field of pavement materials.The application creatively proposes an "elastic-skeleton-void" structure form aiming at the special material composition of the super-elastic porous mixture, and the structure is the fundamental reason for enabling the super-elastic porous pavement to have the composite functions of noise reduction, drainage, self-stress deicing and the like.Meanwhile, in combination with macroscopic physical and mechanical tests, mesoscopic discrete element simulation and digital image processing methods, and based on the uniaxial compression dynamic modulus of the mixture, the void ratio, the contact number of rubber particles and coarse aggregates, the internal force chain structure composition form and the contact force of the mixture, a method for determining the "elastic-skeleton-void" structure is provided, which is used for determining the skeleton composition structure of the mixture.

Description

Technical Field

[0001] This invention relates to the field of road material technology, and in particular to an "elastic-skeleton-void" structure for road mixtures. Background Technology

[0002] With the continuous promotion of the concept of green highway construction, the requirements for road functionality and comfort are constantly increasing. Hyperelastic porous mixture is a heterogeneous multiphase composite pavement material with high-molecular polymers as binders, coarse aggregates as the main component, high rubber particle content, and large porosity. Due to its unique material composition and structural characteristics, hyperelastic porous pavement possesses multiple functions, including noise reduction, drainage, self-stressing de-icing, energy conservation and carbon reduction, and solid waste utilization. The skeleton structure of the hyperelastic porous mixture is the main body for bearing and transmitting external loads on the pavement. The skeleton structure directly affects the mechanical properties of the hyperelastic porous pavement and is the fundamental reason why it possesses special composite functions such as noise reduction, drainage, and self-stressing de-icing.

[0003] Current research on the skeleton structure of pavement mixtures mainly focuses on asphalt mixtures. Generally, the composition and structure of pavement mixtures are categorized into dense-suspended structure, skeleton-void structure, and dense-skeleton structure. A dense-suspended structure means that the asphalt mixture contains excessive fine aggregates, fillers, and asphalt mastic, with their sum exceeding the theoretical usable porosity between coarse aggregates. This necessitates spreading out the skeleton formed by the coarse aggregates, resulting in minimal or no contact between them. A skeleton-void structure means that the sum of the volume fractions of fine aggregates, fillers, and asphalt mastic is less than the theoretical usable porosity between coarse aggregates. A dense-skeleton structure is where the sum of the volume fractions of fine aggregates, fillers, and asphalt mastic equals the theoretical usable porosity between coarse aggregates. The determination of the skeleton structure form is primarily based on the tightness of the skeleton contact, typically using methods such as the coarse aggregate void filling method (CAVF method), the dry compaction porosity of the coarse aggregate, and the loose density of the coarse aggregate. When using the above methods to determine the compositional structure of the superelastic porous mixture (measured porosity of approximately 20%~30%) involved in this invention, it was found that the skeletal structure formed by coarse aggregates was propped open by rubber particles, fine aggregates, and polymer adhesives, with the coarse aggregate particles not in contact with each other. Therefore, defining this mixture as a dense-suspended structure is clearly inappropriate. Thus, existing methods for determining the compositional structure of mixtures cannot satisfy the structural determination of superelastic porous mixtures. Summary of the Invention

[0004] To address the gaps in the aforementioned fields, the present invention aims to propose a novel "elastic-skeleton-void" composition structure for road mixtures.

[0005] A pavement mixture with an "elastic-skeleton-void" structure is composed of graded crushed stone aggregate, rubber particles, mineral powder, and high-molecular polymers. The graded crushed stone aggregate in the pavement mixture has a particle size distribution of 0–16 mm, with the 1.18–4.75 mm graded crushed stone aggregate partially or completely replaced by rubber particles of the same particle size, with the rubber particle content not less than 10%. The proportion of coarse aggregate with a particle size greater than 4.75 mm is not less than 70%. The porosity of the mixture is 20%–30%. Under a loading frequency of 10 Hz and 20°C, the uniaxial compressive dynamic modulus of the mixture is not higher than 1000. MPa; the number of contacts between rubber particles and coarse aggregate is not less than 30% of the total number of contacts between particles; under external load, there are three types of contact force chains inside: rubber particle-coarse aggregate particle, rubber particle-rubber particle, and coarse aggregate particle-coarse aggregate particle, and all three types of internal contact forces can account for more than 50% of the maximum internal contact force; then the mixture is considered to have an "elastic-skeleton-void" structure.

[0006] A method for identifying the "elastic-skeleton-void" structure of road mixtures includes the following sequential steps:

[0007] (1) Prepare graded crushed stone aggregate, rubber particles, mineral powder and polymer adhesive raw materials, wherein the particle size distribution of the graded crushed stone aggregate is 0~16mm, and the graded crushed stone aggregate of 1.18~4.75mm is partially or completely replaced by rubber particles of the same particle size, the rubber particle content is not less than 10%, and the proportion of coarse aggregate with a particle size greater than 4.75mm is not less than 70%; make Marshall specimens, cylindrical specimens and loose mixtures of superelastic porous mixture, determine and calculate the porosity of the mixture as 20%~30%, and the uniaxial compression dynamic modulus at 20℃ and 10Hz loading frequency is not higher than 1000 MPa;

[0008] (2) The cross-sectional images of the mixture are processed using digital image processing technology to confirm the contact characteristics of rubber particles, coarse aggregate, pores, and polymer mortar, thereby confirming that there is interlocking contact between the coarse aggregate particles and the rubber particles; the number of contacts between the rubber particles and the coarse aggregate is not less than 30% of the total number of contacts between particles;

[0009] (3) Based on discrete element numerical simulation, from the perspective of transmitting external loads, if there are contact force chains formed by rubber particles-coarse aggregate particles, rubber particles-rubber particles, and coarse aggregate particles-coarse aggregate particles, and the three types of internal contact forces account for more than 50% of the maximum internal contact force, then it is considered that the rubber particles and coarse aggregate particles together play the role of force transmission path and force skeleton, forming the "skeleton" of the mixture; then the mixture has an "elastic-skeleton-void" structure.

[0010] The method of step (1) is to first prepare raw materials to prepare Marshall specimens, cylindrical specimens and loose mixtures of superelastic porous mixture. The raw materials include graded crushed stone, rubber particles, mineral powder and polymer adhesive. The graded crushed stone aggregate has a particle size distribution of 0~16mm. The graded crushed stone aggregate of 1.18~4.75mm is partially or completely replaced by rubber particles of the same particle size. The rubber particle content is not less than 10%. The proportion of coarse aggregate with a particle size greater than 4.75mm is not less than 70%. The volume of the mixture is measured by vacuum sealing method and the mass of the specimen is measured. The theoretical maximum relative density of the mixture is measured and calculated by vacuum method. And calculate the bulk relative density of the specimen. The porosity V of the actual specimen was calculated according to formula (1), and the porosity was within the range of 20% to 30%. Through uniaxial compression dynamic modulus test, it was found that the uniaxial compression dynamic modulus of the mixture at 20℃ and 10Hz loading frequency was not higher than 1000 MPa.

[0011] Formula (1).

[0012] The digital image processing technology in step (2) mainly includes the following steps:

[0013] Step 1: Use a cutting machine to cut the Marshall specimen to obtain a flat cross-section; place the cross-section specimen on a flat white background plate, and use a digital camera to take a vertical picture of the cross-section to obtain an initial cross-section image;

[0014] Step 2: Use Image-pro plus software to extract the cross-section of the specimen from the image in Step 1, and perform preprocessing such as size calibration, noise reduction, sharpening, and enhancement to remove noise from the image and make the color contrast of each component in the image more obvious;

[0015] Step 3: Perform threshold segmentation on the image processed in Step 2 to extract the rubber particles. Use the "Eyedropper" function in "Histogram Based" to compare with the rubber particles in the actual sample, select a portion of the image containing rubber particles, and label it with a specific color A to obtain an approximate threshold range; then roughly adjust the threshold with a difference of ±5; when the A color area is close to the rubber particle area in the actual sample, adjust the threshold with a difference of ±1; finally, when the A color area and the rubber particle area almost overlap, the threshold at this point is the threshold Y of the rubber particles. XJ The area of ​​region A (color A) is statistically analyzed to obtain the area of ​​the rubber granules.

[0016] Step 4: Based on the images from Step 3, perform threshold segmentation to extract the graded crushed stone particles. Referring to the rubber threshold selection process in Step 3, continuously adjust the threshold range of the graded crushed stone by comparing it with the actual sample. Mark the graded crushed stone with a specific color B based on the images from Step 3. Finally, when the area of ​​color B almost overlaps with the area of ​​graded crushed stone particles, this threshold is the threshold Y of the graded crushed stone particles. SL The total area, particle size, and area of ​​each particle in color B were statistically analyzed, and particles with a diameter greater than 4.75 mm were considered coarse aggregate. The area of ​​independent particles with a diameter greater than 4.75 mm in color B was then statistically analyzed, which is the area of ​​coarse aggregate. Based on color B, coarse aggregate was labeled as color C.

[0017] Step 5: Based on the image from Step 4, threshold segmentation is performed to extract the pore particles. The selection of pore particles is divided into two parts: 1) Since the rubber color in the image is dark, some pores that are far from the cross-section are also dark; 2) Due to the influence of the underlying particles, some pores will be darker than the graded crushed stone color and brighter than the rubber color when identifying their color. Therefore, pore color identification is divided into two parts: the darkest color and the brightest color. Referring to the threshold selection process for rubber particles in Step 3, the threshold range is continuously adjusted by comparing the pores in the actual sample to extract the pore particles. Based on the image from Step 4, in both steps 1) and 2) of Step 5, each pore particle is labeled with a specific color D, and the area of ​​the color D region is statistically analyzed to obtain the pore area.

[0018] Step Six: The polymer adhesive is composed of polymer adhesive and mineral powder, which uniformly coats the surface of rubber particles and graded crushed stone. It is only a thin layer, and it is difficult to extract it by threshold segmentation. In this invention, the outer edges of each rubber particle and graded crushed stone are identified and used as polymer films.

[0019] Step 7: Calculate the ratios of the area of ​​rubber granules, the area of ​​coarse aggregate, and the total cross-sectional area of ​​the specimen. If the error between the calculated results and the volume ratio of each added material to the total sample does not exceed ±2%, and the error between the ratio of pore area to the total cross-sectional area and the measured porosity does not exceed ±2%, then the threshold selection is considered reliable. Otherwise, the threshold selection for each component is not accurate enough. Repeat steps 3 to 6 until the calculation results are within the error range.

[0020] Step 8: The final image generated in Step 6 allows for a direct observation of the contact characteristics of the rubber particles, coarse aggregate, pores, and polymer mortar inside the superelastic porous mixture. If the rubber is not simply filling the pores but is directly interlocked with the coarse aggregate, the number of contacts between the rubber particles and the coarse aggregate particles (X) and the total number of contacts between particles (Y) are counted. The ratio of X to Y is calculated to be no less than 30%.

[0021] The discrete element numerical simulation method in step (3) mainly includes the following steps:

[0022] Step 1: Determine the gradation of the superelastic porous mixture, the volume fraction of each aggregate, and the size and shape of the discrete element specimen of the mixture;

[0023] Step 2: Using PFC discrete element software, the spheres are divided into three categories: rubber spheres, coarse aggregate spheres, and polymer mortar spheres. A discrete element virtual specimen with the same gradation as the superelastic porous mixture is reconstructed. At the same time, the walls around the virtual specimen are generated, and the physical parameters such as density and friction coefficient of different types of spheres in the virtual specimen are assigned values. Finally, a discrete element virtual specimen with a uniform distribution of spheres of different particle sizes and properties is obtained.

[0024] Step 3: By applying an initial motion velocity to the walls around the virtual specimen, the particles of the discrete element virtual specimen are compacted until all particles reach a stress equilibrium state. The porosity of the specimen is measured by measuring the circle method, and the value is in the range of 20% to 30%.

[0025] Step 4: Set up the internal contact constitutive model of the particles, and use Burger's model and parallel bonding model to characterize the viscoelastic characteristics of the polymer adhesive. Linear models are used for the interaction between rubber particles, between coarse aggregate particles, and between rubber particles and coarse aggregate particles. For the interaction between polymer adhesive particles and other particles, a contact bonding model is used. Run the program until the mixture reaches equilibrium.

[0026] Step 5: Set the displacement of all spherical particles to zero, and only set the moving speed of the upper wall in the direction of the specimen inward, i.e., the negative Y-axis direction. Set the program to stop when the strain in the Y-direction of the virtual specimen is not less than 0.2%.

[0027] Step 6: By analyzing the contact force vector diagram and force chain network diagram inside the virtual specimen, if there are contact force chains formed by rubber particles-coarse aggregate particles, rubber particles-rubber particles, and coarse aggregate particles-coarse aggregate particles, then it is considered that both rubber particles and coarse aggregate particles play the role of force transmission paths. If the contact force between the three types of particles reaches more than 50% of the maximum contact force between particles inside the virtual specimen, then both coarse aggregate particles and rubber particles play the role of force-bearing skeleton and load-bearing when subjected to external loads. At this time, it is considered that the rubber particles and coarse aggregate particles together form the "skeleton" of the mixture, and the mixture has an "elastic-skeleton-void" structure.

[0028] Therefore, this invention argues that the super-elastic porous mixture does not belong to the skeleton-void structure, the dense-skeleton structure, or the dense-suspension structure. Based on its special material composition and structural characteristics, a novel "elastic-skeleton-void" structural composition form for road mixtures is proposed.

[0029] Combining macroscopic physical and mechanical experiments, microscopic discrete element simulation, and digital image processing, this paper proposes a discrimination method for the "elastic-skeleton-void" structure based on indicators such as the uniaxial compression dynamic modulus of the mixture, porosity, number of contacts between rubber particles and coarse aggregates, internal force chain structure composition, and contact force. This method is used to determine the skeleton composition structure of the mixture. Attached Figure Description

[0030] Figure 1 For the discrete element model of hyperelastic porous mixture,

[0031] Wherein: black represents rubber granules, light gray represents coarse aggregate granules, dark gray represents polymer granules, and the lines around the specimen represent virtual walls.

[0032] Figure 2 To measure the porosity of the model,

[0033] Figure 3 Force distribution diagram inside the specimen after assigning contact force parameters.

[0034] Figure 4 This describes the stress state of the rubber particles.

[0035] Figure 5 This is a stress diagram of the mixture.

[0036] Figure 6 This is a contact force diagram of the mixture.

[0037] Figure 7 A conceptual diagram of force transfer in a mixture;

[0038] Figure 8 This is a cross-sectional view of the Marshall specimen.

[0039] Figure 9 Image showing the extraction of rubber particles.

[0040] Figure 10 This is a diagram of coarse aggregate extraction.

[0041] Figure 11 This is a structural diagram of the superelastic porous mixture after processing.

[0042] Figure 12 for Figure 4 A magnified view of the central area.

[0043] Among them: 1, 2, 3, 4 and 5 are all coarse aggregates; 1 and 2 and 3 are in direct interlocking contact; 1 and 4 and 5 are not in direct contact, but a large number of rubber particles are distributed between 1 and 4 and between 1 and 5. Detailed Implementation

[0044] The present invention will be further described in detail below with reference to the embodiments.

[0045] Example 1

[0046] 1. Confirm porosity

[0047] Marshall specimens of the hyperelastic porous mixture were prepared. The particle size range of the raw materials was 0-16 mm, and the graded crushed stone aggregate is shown in Table 1 below. Among them, coarse aggregate, i.e., crushed stone with a particle size greater than 4.75 mm, accounted for 70% of the total content (by volume). The aggregates with particle sizes of 1.18-2.36 mm and 2.36-4.75 mm were completely replaced by rubber granules made from waste tires, with a total rubber content of 20% (by volume). The mass ratio of polymer to solid aggregate was 4.5%. Referring to standard JTG E20-2011, the theoretical maximum relative density of the hyperelastic porous mixture was measured and calculated using the vacuum method, and the result was 2.38 g / cm³. 3 Marshall specimens of a superelastic porous mixture were prepared by compaction. The volume of the mixture was measured to be 516.23 cm³ using the vacuum sealing method. 3 The porosity was measured and calculated to be approximately 24%; the uniaxial compression dynamic modulus was 658 MPa at 20℃ and a loading frequency of 10 Hz.

[0048] Table 1. Particle size of graded crushed stone aggregate

[0049] Sieve aperture size (mm) 16 13.2 9.5 4.75 2.36 1.18 0.6 0.3 0.15 0.075 Pass rate (%) 100 91 68 30 18 10 8 6 4 3

[0050] 2. Confirm the contact characteristics of rubber particles, coarse aggregates, voids, and polymer adhesive.

[0051] (1) The specimen was cut using a cutting machine to obtain a flat cross-section. After drying at room temperature, the cross-section of the specimen was photographed perpendicularly using a digital camera. See Figure 8 .

[0052] (2) Import the image into Image-pro plus software. First, the image size is calibrated. Then, the tomographic image is denoised to remove noise. Next, the image is sharpened and enhanced to make the color contrast of each component in the image more obvious, which prepares for the subsequent identification of each component.

[0053] (3) Rubber particles were extracted by comparing with the actual Marshall cross section. The "eyedropper" tool in Histogram Based was used to first collect the threshold values ​​of some rubber particles. Then, by continuously adjusting the threshold range, a relatively complete threshold range of 28-76 was finally obtained. The rubber particle portion was filled with dark gray, and the image was saved. See [link to image]. Figure 9 Simultaneously, the area of ​​the rubber granules is calculated to be 1518 mm². 2 .

[0054] (4) Compare the actual Marshall cross section and extract the graded crushed stone portion from the processed image. Repeat the method in (3) above to extract the threshold of the graded crushed stone particles, ranging from 103 to 255. Fill the graded crushed stone particle portion with gray and save the image; at the same time, calculate the area of ​​the coarse aggregate particles, which is 4464 mm². 2 The area of ​​coarse aggregate particles with a diameter greater than 4.75 mm was calculated to be 4354 mm². 2 It is marked as light gray; see Figure 10 .

[0055] (5) Compare the actual Marshall cross-section and extract the pores from the processed image. First, extract the threshold values ​​of the darker pores in the image, ranging from 0 to 28; then extract the threshold values ​​of the lighter pores, ranging from 76 to 103. Fill the pores with black and save the image. See [link to image]. Figure 11 Meanwhile, the areas of the two pore sections were calculated, and they were 434 mm². 2 and 1537mm 2 .

[0056] (6) The calculated ratio of rubber area to total area is 19%, the ratio of coarse aggregate area to total area is 73%, and the ratio of total pore area to total area is 25%. The results are close to the actual test material dosage and porosity, with an error of no more than ±2%, indicating that the operation method and threshold selection results are accurate.

[0057] (7) Figure 11 For the processed image, Figure 11 Enlarged portion of the area Figure 12Among them, 1, 2, 3, 4, and 5 are all coarse aggregates. 1 and 2, and 3 are in direct interlocking contact; 1 and 4, and 5 are not in direct contact, but rather a large number of rubber particles are distributed between 1 and 4, and between 1 and 5. This shows that the rubber is not simply filling the pores, but rather interlocking and contacting with the coarse aggregates.

[0058] (8) The number of contacts between rubber particles and coarse aggregate particles is 482, and the total number of contacts between particles is 1274, that is, the number of contacts between rubber particles and coarse aggregate particles accounts for 38% of the total number of contacts between particles.

[0059] 3. Confirm the "elastic-skeleton-void" structure

[0060] This section provides a "elastic-skeleton-void" discrimination method from the perspective of external load transmission using discrete element numerical simulation, taking the process of a hyperelastic porous mixture specimen subjected to external load as an example.

[0061] (1) Determine the gradation of the virtual specimen and the volume fraction of each aggregate grade. Set the shape of the virtual specimen of the mixture to 600 mm in length and 100 mm in width. See Table 1. The aggregate grades with particle sizes of 1.18~2.36 mm and 2.36~4.75 mm are completely replaced by rubber granules made from waste tires.

[0062] (2) Using PFC discrete element method software, the spheres were divided into three "groups": polymer mortar, rubber spheres, and coarse aggregate spheres. Among them, 4.75~16mm was considered coarse aggregate spheres; 1.18~2.36mm and 2.36~4.75mm were rubber granule spheres; and particles smaller than 1.18mm were considered polymer mortar spheres. Considering the program running time, the minimum particle size in the specimen was set to 1.13mm. Using the method of generating according to particle size distribution, a discrete element virtual specimen with the same gradation and volume fraction as the superelastic porous mixture was reconstructed. At the same time, walls were generated around the virtual specimen, and the upper wall was numbered 3. The physical parameters such as the density of the spheres in different "groups" of the virtual specimen and the friction coefficient of the wall surface were assigned values. After running the program using the "solve" command, a discrete element virtual specimen with uniformly distributed spheres of different sizes and properties is obtained. Black represents rubber particles, light gray represents coarse aggregate particles, and dark gray represents polymer particles. The lines around the specimen represent virtual walls. (See...) Figure 1 .

[0063] (3) Assign values ​​to parameters such as density, damping ratio, and friction coefficient for the three types of spheres. The densities of coarse aggregate, rubber, and polymer mortar are 2700, 1150, and 1000 kg / m³, respectively. 3The damping ratios were 0.7, 0.7, and 0.35, and the friction coefficients were 1, 0.6, and 0.25, respectively. During this stage, the friction coefficient between particles was assigned a value of 0.25, and a linear connection model was used between all particles. The elastic modulus of the mixture was set to 300e6 Pa, and the Poisson's ratio was 1.5. The wall stiffness was assigned a value of 1e8 N / m. The program was run until the specimen reached equilibrium, and the porosity of the virtual specimen was measured to be 27% using a measuring circle.

[0064] (3) By applying a movement speed to walls 1, 2, 3, and 4, i.e., applying a pressure of 10 kPa to the virtual specimen, the particles are displaced and squeezed, thereby achieving the purpose of compacting the mixture. Finally, the "solve" command is used to run the program until all particles reach a stress equilibrium state. By defining a measuring sphere, the porosity is measured to be approximately 24%, see [link to relevant documentation]. Figure 2 .

[0065] (4) Set up a contact constitutive model for the particles, and use Burger's model and parallel bonding model to characterize the viscoelastic characteristics of the polymer adhesive particles. Linear models are used for the interaction between rubber particles, between coarse aggregate particles, and between rubber particles and coarse aggregate particles. Contact bonding model is used for the interaction between polymer adhesive particles and other particles. The elastic moduli of coarse aggregate, rubber, and polymer adhesive particles are set to 55e9, 78e5, and 25e8 Pa, respectively. The Poisson's ratios are set to 0.25, 0.47, and 0.42, respectively. The normal parallel bonding stiffnesses are set to 1e10, 1e5, and 1e6 N / m, respectively. The parameters such as the connection elastic modulus and damping ratio between different types of particles are selected to be the same as the parameters of the polymer adhesive particles. The friction angle of the mixture is set to 30° and the cohesion is set to 1e9 Pa. Finally, the "solve" command is used to run the program until the mixture system reaches equilibrium. No pressure was applied in this part. If the internal force chains appear thickened and disordered in the final contact forces within the specimen, it indicates that the parameter assignment was successful. See Figure 3 .

[0066] (5) Set the displacement of all spherical particles to zero, and set a downward moving speed only for wall No. 3, i.e., the upper wall. Set the program to stop when the strain of the virtual specimen is greater than 0.2%. Figure 4 The diagram only shows the distribution of force chains and rubber particles in a partial area. At point 4a, multiple rubber particles can be seen forming a relatively long force chain, which plays a role in transmitting force; at point 4b, a single rubber particle is sandwiched between coarse aggregates, sharing the load with the coarse aggregates; at point 4c, the rubber particles only serve a filling function. Figure 5 This shows the particle distribution in a localized area of ​​the mixture. Figure 6This is a force chain distribution map for this region. (Compare) Figure 5 and Figure 6 In region a, the force chain consists of multiple coarse aggregate particles; in region b, the force is transmitted between rubber particles and coarse aggregate particles. Therefore, the force can be transmitted to the surrounding particles through the rubber particles and coarse aggregate particles. After the virtual mixture is subjected to force, there are three force chain structures inside the virtual specimen: rubber particle-coarse aggregate particle, rubber particle-rubber particle, and coarse aggregate particle-coarse aggregate particle. Both the rubber spheres and the coarse aggregate spheres act as force transmission paths. Figure 7 ).

[0067] (6) When the strain in the Y-axis direction of the virtual specimen reaches 0.2%, the overlying load reaches 530 kPa. Among the contact forces inside the specimen, the maximum contact force between coarse aggregate particles and rubber particles is approximately 59 N; the maximum contact force between coarse aggregate particles is approximately 47 N; and the maximum contact force between rubber particles is approximately 40 N. Comparing the three force chains, it is found that the maximum contact force between coarse aggregate particles and rubber particles is the largest; the maximum contact force between coarse aggregate particles and coarse aggregate particles is the second largest; and the maximum contact force between rubber particles and rubber particles is the smallest, accounting for 100%, 80%, and 68% of the maximum contact force inside the specimen, respectively. This indicates that the rubber particles and coarse aggregate particles together act as a load-bearing skeleton.

[0068] (7) Combining steps 1 and 2, it is determined that the superelastic porous mixture is an "elastic-skeleton-void" structure.

Claims

1. A method for identifying the "elastic-skeleton-void" structure of a pavement mixture, comprising the following sequential steps: (1) Preparing graded aggregate of macadam, rubber particles, mineral powder and raw material of high polymer polymerization adhesive, wherein, The graded crushed stone aggregate has a particle size distribution of 0~16mm, of which the graded crushed stone aggregate of 1.18~4.75mm is partially or completely replaced by rubber particles of the same particle size, with the rubber particle content not less than 10%, and the proportion of coarse aggregate with a particle size greater than 4.75mm is not less than 70%; Marshall specimens, cylindrical specimens and loose mixtures of superelastic porous mixtures are prepared, and the porosity of the mixture is measured and calculated to be 20%~30%, and the uniaxial compression dynamic modulus is not higher than 1000 MPa at 20℃ and 10Hz loading frequency; (2) The cross-sectional images of the mixture are processed using digital image processing technology to confirm the contact characteristics of rubber particles, coarse aggregate, pores, and polymer mortar, thereby confirming that there is interlocking contact between the coarse aggregate particles and the rubber particles; the number of contacts between the rubber particles and the coarse aggregate is not less than 30% of the total number of contacts between particles; (3) Based on discrete element numerical simulation, from the perspective of transmitting external loads, if there are contact force chains formed by rubber particles-coarse aggregate particles, rubber particles-rubber particles, and coarse aggregate particles-coarse aggregate particles, and the three types of internal contact forces account for more than 50% of the maximum internal contact force, then it is considered that the rubber particles and coarse aggregate particles together play the role of force transmission path and force skeleton, forming the "skeleton" of the mixture; then the mixture has an "elastic-skeleton-void" structure.

2. The method of claim 1, wherein the method of step (1) is to prepare a super-elastic porous mixture Marshall specimen, a cylinder specimen, and a loose mixture by preliminarily preparing raw materials including a graded aggregate, rubber particles, a mineral powder, and a polymer adhesive, and wherein, The graded crushed stone aggregate has a particle size distribution of 0~16mm. Graded crushed stone aggregate of 1.18~4.75mm is partially or completely replaced by rubber particles of the same particle size, with the rubber particle content not less than 10%. Coarse aggregate with a particle size greater than 4.75mm accounts for not less than 70%. The volume of the mixture is measured using a vacuum sealing method, and the mass of the specimen is also measured. The theoretical maximum relative density of the mixture is measured and calculated using a vacuum method. And calculate the bulk relative density of the specimen. The porosity V of the actual specimen was calculated according to formula (1), and the porosity was within the range of 20% to 30%. The uniaxial compression dynamic modulus was measured to be no higher than 1000 MPa at 20℃ and 10Hz loading frequency through a uniaxial compression dynamic modulus test. Formula (1).

3. The method according to claim 1, wherein step (2) further includes quantitatively confirming the porosity of the mixture, with an error of ±2% of the porosity of the mixture measured and calculated in step (1).

4. According to claim 1, the digital image processing technology in step (2) mainly includes the following steps: Step 1: Use a cutting machine to cut the Marshall specimen to obtain a flat cross-section; place the cross-section specimen on a flat white background plate, and use a digital camera to take a vertical picture of the cross-section to obtain an initial cross-section image; Step 2: Use Image-pro plus software to extract the cross-section of the specimen from the image in Step 1, and perform preprocessing such as size calibration, noise reduction, sharpening, and enhancement to remove noise from the image and make the color contrast of each component in the image more obvious; Step 3: Perform threshold segmentation on the image processed in Step 2 to extract the rubber particles. Use the "Eyedropper" function in "Histogram Based" to compare with the actual rubber particles in the sample, select a portion of the image containing the rubber particles, and label it with a specific color A to obtain the threshold range. Roughly adjust the threshold with a difference of ±5. When the A color region is close to the rubber particle region in the actual sample, adjust the threshold with a difference of ±1. Finally, when the A color region almost overlaps with the rubber particle region, the threshold at this point is the threshold Y of the rubber particles. XJ The area of ​​region A (color A) is statistically analyzed to obtain the area of ​​the rubber granules. Step 4: Based on the images from Step 3, perform threshold segmentation to extract the graded crushed stone particles. Referring to the rubber threshold selection process in Step 3, continuously adjust the threshold range of the graded crushed stone by comparing it with the actual sample. Mark the graded crushed stone with a specific color B based on the images from Step 3. Finally, when the area of ​​color B almost overlaps with the area of ​​graded crushed stone particles, this threshold is the threshold Y of the graded crushed stone particles. SL The total area, particle size, and area parameters of each particle in color B were statistically analyzed, and particles with a diameter greater than 4.75 mm were considered as coarse aggregate. The area of ​​independent particles with a diameter greater than 4.75 mm in color B was then statistically analyzed, which is the area of ​​coarse aggregate. Based on color B, coarse aggregate was labeled as color C. Step 5: Based on the image from Step 4, threshold segmentation is performed to extract the pore particles. The selection of pore particles is divided into two parts: 1) Since the rubber color in the image is dark, some pores that are far from the cross-section are also dark; 2) Due to the influence of the underlying particles, some pores will be darker than the graded crushed stone color and brighter than the rubber color when identifying their color. Therefore, pore color identification is divided into two parts: the darkest color and the brightest color. Referring to the threshold selection process for rubber particles in Step 3, the threshold range is continuously adjusted by comparing the pores in the actual sample to extract the pore particles. Based on the image from Step 4, in both steps 1) and 2) of Step 5, each pore particle is labeled with a specific color D, and the area of ​​the color D region is statistically analyzed to obtain the pore area. Step Six: The polymer adhesive is composed of polymer adhesive and mineral powder, which uniformly coats the surface of rubber particles and graded crushed stone. It is only a thin layer, and it is difficult to extract it by threshold segmentation. In this invention, the outer edges of each rubber particle and graded crushed stone are identified and used as polymer films. Step 7: Calculate the ratios of the area of ​​rubber granules, the area of ​​coarse aggregate, and the total cross-sectional area of ​​the specimen. If the error between the calculated results and the volume ratio of each added material to the total sample does not exceed ±2%, and the error between the ratio of pore area to the total cross-sectional area and the measured porosity does not exceed ±2%, then the threshold selection is considered reliable. Otherwise, the threshold selection for each component is not accurate enough. Repeat steps 3 to 6 until the calculation results are within the error range. Step 8: The final image generated in Step 6 allows for a direct observation of the contact characteristics of the rubber particles, coarse aggregate, pores, and polymer mortar inside the superelastic porous mixture. If the rubber is not simply filling the pores but is directly interlocked with the coarse aggregate, the number of contacts between the rubber particles and the coarse aggregate particles (X) and the total number of contacts between particles (Y) are counted. The ratio of X to Y is calculated to be no less than 30%.

5. According to claim 2, the discrete element numerical simulation in step (3) mainly includes the following steps: Step 1: Determine the gradation of the superelastic porous mixture, the volume fraction of each aggregate, and the size and shape of the discrete element specimen of the mixture; Step 2: Using PFC discrete element software, the spheres are divided into three categories: rubber spheres, coarse aggregate spheres, and polymer mortar spheres. A discrete element virtual specimen with the same gradation as the superelastic porous mixture is reconstructed. At the same time, the walls around the virtual specimen are generated, and the physical parameters of density and friction coefficient of different types of spheres in the virtual specimen are assigned values. Finally, a discrete element virtual specimen with a uniform distribution of spheres of different particle sizes and properties is obtained. Step 3: By applying an initial motion velocity to the walls around the virtual specimen, the particles of the discrete element virtual specimen are compacted until all particles reach a stress equilibrium state. The porosity of the specimen is measured by measuring the circle method, and the value is in the range of 20% to 30%. Step 4: Set up the internal contact constitutive model of the particles, and use Burger's model and parallel bonding model to characterize the viscoelastic characteristics of the polymer adhesive. Linear models are used for the interaction between rubber particles, between coarse aggregate particles, and between rubber particles and coarse aggregate particles. For the interaction between polymer adhesive particles and other particles, a contact bonding model is used. Run the program until the mixture reaches equilibrium. Step 5: Set the displacement of all spherical particles to zero, and only set the moving speed of the upper wall in the direction of the specimen inward, i.e., the negative Y-axis direction. Set the program to stop when the strain in the Y-direction of the virtual specimen is not less than 0.2%. Step 6: By analyzing the contact force vector diagram and force chain network diagram inside the virtual specimen, if there are contact force chains formed by rubber particles-coarse aggregate particles, rubber particles-rubber particles, and coarse aggregate particles-coarse aggregate particles, then it is considered that both rubber particles and coarse aggregate particles play the role of force transmission paths; if the contact force between the three types of particles reaches more than 50% of the maximum contact force between particles inside the virtual specimen, then both coarse aggregate particles and rubber particles play the role of force-bearing skeleton and load-bearing when subjected to external loads; at this time, it is considered that rubber particles and coarse aggregate particles together form the "skeleton" of the mixture, and the mixture has an "elastic-skeleton-void" structure.

Images