Ultra-compact asphalt mixture design method

Abstract

The invention relates to a design method of an ultra-compact asphalt mixture. The design method comprises the following steps: S1, selecting the grading type of the asphalt mixture, the number and particle size specification of composition specifications of coarse aggregate, and the types and specifications of fine aggregate and mineral powder filler; s2, acquiring physical property parameters of the selected aggregates; s3A: generating a plurality of groups of mineral aggregate proportions, establishing a main skeleton discrete element model of each mineral aggregate proportion, and calculating a mineral aggregate void rate and a main skeleton tight loading density of a coarse aggregate main skeleton; s3B, calculating the designed void ratio according to the grading type and the coarse aggregate composition specification; s4, respectively calculating the asphalt dosage of each group of mineral aggregate proportion; and S5, calculating the powder-binder ratio and the oil film thickness of each group of mineral aggregate proportion, and selecting the mineral aggregate proportion of which the powder-binder ratio and the oil film thickness accord with a preset range as specific parameters of the asphalt mixture. According to the design method of the super-compact asphalt mixture, the super-compact asphalt mixture can be designed.

Description

Technical Field

[0001] This invention relates to the field of road engineering bridge deck paving technology, and in particular to a design method for ultra-dense asphalt mixtures. Background Technology

[0002] A steel bridge deck pavement structure typically includes steel plates, a pavement layer, and the interface between the two. The steel plates form the bridge deck; the interface is a waterproof adhesive layer laid on the surface of the steel plates; the pavement layer is usually formed by compacting epoxy asphalt mixture and laid on top of the waterproof adhesive layer to further prevent water from seeping into the steel plates.

[0003] Asphalt mixtures are typically formed by mixing asphalt with granular aggregates of different sizes in a certain proportion and then compacting the mixture. Since the asphalt cannot completely fill the spaces between the aggregate particles, voids are formed between them. The presence of voids is usually expressed by the porosity, which is the percentage of void volume to the total volume of the specimen.

[0004] To achieve better waterproofing performance, the porosity of epoxy asphalt mixtures is generally designed to be less than 3%. At this point, the permeability coefficient of the pavement surface is 0 mL / min. Therefore, theoretically, rainwater is unlikely to seep into the pavement layer, let alone reach the steel plate. This effectively prevents rainwater from penetrating the steel bridge surface and corroding the steel plate.

[0005] However, in practice, it has been found that steel bridge decks with a pavement layer with a porosity of less than 3% still frequently suffer from defects such as micro-cracks, pumping, exposed aggregate, and potholes caused by water seepage into the pavement layer within 2-3 years after opening to traffic, and cannot be effectively protected. Summary of the Invention

[0006] Based on this, the purpose of this invention is to design asphalt mixtures with better waterproof performance and to provide an asphalt mixture design method.

[0007] A design method for ultra-dense asphalt mixtures includes the following steps: S1: Select the gradation type of the asphalt mixture, and select the composition specifications of coarse aggregate, the type and specifications of fine aggregate and mineral powder filler according to the gradation type; S2: Obtain the physical property parameters of each selected ore; S3A: Set the mass percentage range of coarse aggregates of each particle size according to the composition relationship between the minerals, and generate at least two sets of mineral proportions; at the same time, establish a discrete element model of the main skeleton to calculate the mineral gap ratio and main skeleton compaction density of the coarse aggregate main skeleton under each mineral proportion. S3B: Calculate the design void ratio threshold based on the gradation type and coarse aggregate composition specifications; specifically including: S3B1: Select asphalt mixture samples of the same gradation type as the selected gradation type; S3B2: X-ray CT scans were performed on each of the aforementioned asphalt mixture samples to obtain multiple sets of tomographic images; S3B3: Based on the fault images of each group, detect the porosity and connectivity porosity of each asphalt mixture sample; the statistical formula for the porosity is: , in, The porosity of each specimen, For the first The area of ​​the gaps in a CT scan tomographic image. For the first The total area of ​​the CT scan tomographic images. The number of CT scan tomographic images; S3B4: Using a nonlinear curve fitting method, a correspondence between porosity and interconnected porosity is established based on the porosity and interconnected porosity of each asphalt mixture sample; the relationship between interconnected porosity and porosity satisfies an exponential function as follows: , in, The interconnected porosity of asphalt mixtures. The porosity of asphalt mixture. , and The parameters are for fitting the function; S3B5: Establish a design void ratio threshold function for asphalt mixtures based on the aforementioned correspondence; the design void ratio threshold is the void ratio when the connected void ratio is zero; the design void ratio threshold function is: , in, To design the porosity threshold, , and The parameters are for fitting the function, and , ; S3B6: Calculate the standard deviation and estimated value of each of the function fitting parameters according to their confidence intervals, and calculate the variance and standard deviation of the design porosity threshold in combination with the function of the design porosity threshold; S3B7: Calculate the design porosity threshold based on the variance and standard deviation of the design porosity threshold; S3B8: The target porosity is calculated based on the measured porosity error distribution and the designed porosity threshold. S4: Based on the relationship between each mineral aggregate, the target void ratio, the mineral aggregate void ratio of the coarse aggregate main skeleton, and the asphalt content, calculate the asphalt content for each group of mineral aggregates. S5: Calculate the powder-to-binder ratio and oil film thickness of each group of aggregate proportions, and select aggregate proportions with powder-to-binder ratio and oil film thickness within a preset range as specific parameters of asphalt mixture.

[0008] Compared with existing technologies, the ultra-dense asphalt mixture design method of the present invention can design asphalt mixtures with zero interconnected void ratio, achieve an ultra-dense state with zero diffusion coefficient of asphalt mixture, and effectively prevent gaseous water from diffusing in asphalt mixture.

[0009] Further, in step S3B3, the specific steps for detecting the connectivity porosity are as follows: S3B301: Convert all the fault images into binary image data, and select the fault image set of one asphalt mixture sample; S3B302: Take the top-level fault image from the fault image set; S3B303: Check the grayscale value of the current pixel. If the grayscale value of the current pixel is 0, proceed to step S3B04. If the grayscale value of the current pixel is not 0, check the grayscale value of the next pixel. After traversing all pixels, proceed to step S3B05. S3B304: Check the connectivity of the 8-neighborhood of the current pixel. If there are pixels with a gray value of 0 in the 8-neighborhood, mark these pixels with a gray value of 0 as the same gap object. S3B305: Determine whether the pixels of the current tomographic image have been traversed. If the current pixel is the last pixel, proceed to step S3B306; if the current pixel is not the last pixel, repeat steps S3B303 to S3B304. S3B306: Determine whether the current tomographic image is the bottom layer tomographic image. If yes, proceed to step S3B307; otherwise, take the tomographic image located in the next layer and repeat steps S3B303 to S3B305. S3B307: Acquire the tomographic image located at the top layer; S3B308: Retrieves a gap object in the current tomographic image; S3B309: Check if there are any pixels with a gray value of 0 in the 8-neighborhood of the gap object in the next layer of tomographic image. If so, mark them as vertically connected objects. S3B310: Determine whether the void objects in the current fault image have been traversed. If not, take the next void object and repeat step S3B309; if the traversal is complete, execute step S3B311. S3B311: Determine whether the current fault image is the bottom fault image. If yes, proceed to step S3B312. If not, take the fault image located in the next layer and repeat steps S3B309 to S3B310. S3B312: Obtain the tomographic image located at the bottom layer; S3B313: Retrieves a gap object in the current tomographic image; S3B314: Check if there are any pixels with a gray value of 0 in the 8-neighborhood of the gap object in the previous layer tomographic image. If they exist, mark them as connected; otherwise, mark them as closed. S3B315: Determine whether the void objects in the current fault image have been traversed. If not, take the next void object and repeat step S3B313; if the traversal is completed, execute step S3B316. S3B316: Determine whether the current fault image is the top fault image. If yes, proceed to step S3B317; otherwise, take the fault image located in the previous layer and repeat steps S3B313 to S3B312. S3B317: Check whether there is a connection mark in the gaps of each top fault image. If there is, mark the gap with the gap mark as a connected gap. If there is no connection mark, mark it as a closed gap. S3B318: Calculate the connectivity porosity based on the connectivity gaps.

[0010] Furthermore, the relationship between the various aggregates, the target porosity, the aggregate void ratio of the coarse aggregate main skeleton, and the asphalt content is expressed as follows: , in, This represents the mass percentage of coarse aggregate. Here, n represents the mass percentage of each particle size, and n is the number of coarse aggregate components. This represents the mass percentage of fine aggregate. This represents the mass percentage of mineral powder. This represents the percentage of asphalt used. For the target porosity, The main frame is tightly packed with high density. The apparent density of fine aggregate, The apparent density of the mineral powder is... The density of the asphalt binder. The aggregate void ratio is the main skeleton of coarse aggregate; For different sieve apertures, the throughput is... This refers to the sieve aperture diameter; only aggregates with particle sizes smaller than this sieve aperture diameter can pass through the corresponding sieve aperture. sieve aperture diameter This represents the lower limit of the sieve aperture throughput. sieve aperture diameter This represents the upper limit of the sieve aperture throughput.

[0011] Furthermore, step S3A specifically includes: S3A1: Establish corresponding aggregate particle models based on the particle size specifications of each coarse aggregate; S3A2: Set the mass percentage range of fine aggregate, mineral powder and coarse aggregate of different particle sizes according to the composition relationship between each mineral material, and generate at least two sets of coarse aggregate ratios; S3A3: Using the aggregate particle models mentioned above, discrete element models of the asphalt coarse aggregate main skeleton structure with each coarse aggregate ratio are established respectively. S3A4: The mesoscopic contact stiffness of each aggregate particle model in the model space is set using a linear contact model; S3A5: Calculate the total volume of all coarse aggregate particles under each coarse aggregate ratio; S3A6: Calculate the aggregate void ratio and aggregate compaction density of the asphalt coarse aggregate main skeleton structure based on the discrete element model of each coarse aggregate ratio; the formula for calculating the aggregate void ratio of the coarse aggregate main skeleton is: , in, The aggregate void ratio is the main skeleton of coarse aggregate. The volume of the model space. This represents the total volume of the aggregate particle model for coarse aggregates of various particle sizes. , in, The aggregate void ratio is the main skeleton of coarse aggregate. The main frame is tightly packed with high density. The apparent density of the composite material after mixing coarse aggregates of various specifications according to the coarse aggregate ratio; The formula for the synthetic apparent density is: , in, The apparent density of the composite material is the result of mixing coarse aggregates of various specifications according to the coarse aggregate ratio. This represents the mass percentage of coarse aggregate. Let n be the mass percentage of each particle size, and n be the number of coarse aggregate components. The formula for the mass percentage of the coarse aggregate is: , in, This represents the mass percentage of coarse aggregate. denoted as the mass percentage of each particle size specification, and n is the number of component specifications of the coarse aggregate. S2A7: Based on the compositional relationship between the aggregates and the mass percentage range of fine aggregates and mineral powder, at least two aggregate ratios are generated for each group of coarse aggregate ratios; the compositional relationship between the coarse aggregates, fine aggregates, mineral powder, asphalt, and aggregates is expressed as follows: , , in, This represents the mass percentage of coarse aggregate. This represents the mass percentage of fine aggregate. This represents the mass percentage of mineral powder. Here, n represents the mass percentage of each particle size, and n is the number of coarse aggregate components. sieve aperture diameter This represents the lower limit of the sieve aperture throughput. sieve aperture diameter This represents the upper limit of the sieve aperture throughput.

[0012] Furthermore, the aggregate particle model is established as follows: S3A101: Select coarse aggregate particles according to the selected particle size specification; S3A102: Perform X-ray CT scans on each of the coarse aggregate particles with different particle sizes to obtain a set of tomographic images of each coarse aggregate particle; S3A103: Obtain the corresponding equivalent volume sphere diameter based on the tomographic image set of each coarse aggregate particle; S3A104: Generate a sphere according to the equivalent volume sphere diameter; S3A105: Generate multiple random points on the surface of the sphere and connect them to form a three-dimensional convex hull; S3A106: Voxelize the surface and interior of the three-dimensional convex hull; S3A107: Extract the skeleton points of each 3D convex hull S3A108: Generate the maximum inscribed sphere covering each skeleton point of each three-dimensional convex hull to obtain the aggregate particle model of each coarse aggregate particle; S3A109: Assign the bulk density to each aggregate particle model corresponding to its particle size specification.

[0013] Furthermore, the operation of surface voxelization of the three-dimensional convex hull is specifically as follows: a voxel space composed of orthogonal meshes is established, then the spatial relationship between each voxel and the closed surface of the three-dimensional convex hull is determined, and the voxels that intersect with the closed surface are retained, thus realizing the surface voxelization of the three-dimensional convex hull. The specific operation of voxelizing the interior of the three-dimensional convex hull is as follows: filling the interior voxels by scanning or ray casting to achieve voxelization of the interior of the three-dimensional convex hull. The orthogonal mesh is a cube, and its function expression is: , Among them voxels Centered on The cube, where h is the resolution and the side length of the cube is... The spatial coordinates of the voxel are .

[0014] The functional expression of the three-dimensional convex hull after surface voxelization is as follows: , The functional expression of the three-dimensional convex hull after internal voxelization is:

[0015] S3A108: Extract the skeleton points of each 3D convex hull.

[0016] Further, in step S3A109, when generating each covering maximum inscribed sphere, the skeleton points of the three-dimensional convex hull are arranged in descending order of radius; then, the current maximum radius sphere is placed sequentially to cover the space not occupied by the previous sphere; when the sphere filling volume loss rate reaches a certain threshold, space filling is stopped. The filling volume loss rate is the ratio of the volume of the sphere filling to the volume of the three-dimensional convex hull, and its calculation formula is as follows: , in, For the filling volume loss rate, Let the volume of each sphere be , The volume is the 3D convex hull volume. In this embodiment, the filling volume loss rate is 5%. In this embodiment, the clump (cluster) block generation technology in PFC3D is used to cover the maximum inscribed sphere at each skeleton point; The threshold for the cumulative volume is set according to a threshold function, which is: , in, The volume of the model space. For cumulative volume, It represents the volume of the smallest aggregate particle model within a given particle size specification.

[0017] Furthermore, according to the design method for ultra-dense asphalt mixtures according to claim 8, the characteristic is that a reduction factor is set. The size of the aggregate particle model is linearly reduced to Subsequently, the surface normal stiffness of the aggregate particle model was temporarily set to... To ensure that particles bounce off each other upon contact, surface friction and damping are temporarily set to zero, allowing particles to slide freely and rebound almost elastically upon contact, thus randomizing the initial arrangement and orientation of each aggregate particle model. Subsequently, the Cycle command is used to progressively expand the aggregate particle model to its original volume, causing random rotation and bounce-off of overlapping areas within the model space. This progressive expansion is based on a step-by-step magnification factor, which is: , This indicates that expansion to the original volume size is achieved through N-level amplification.

[0018] Further, in step S5, based on a multiple linear regression model relating the mass percentage of each aggregate, the interstitial ratio of the coarse aggregate main skeleton, and the passing rate of each sieve size, the asphalt content for each aggregate proportion is calculated; the expression for the multiple linear regression model is: , in, This represents the percentage of asphalt used. The aggregate void ratio is the main skeleton of coarse aggregate. This refers to the mass percentage of coarse aggregate with a particle size of 5-10 mm. This represents the mass percentage of fine aggregate. The throughput of a sieve with a aperture size of 9.5 mm. The throughput of a sieve with a aperture size of 4.75 mm. The throughput of a sieve with a aperture size of 2.36 mm. The throughput of a sieve with a aperture size of 1.18 mm. The throughput of a sieve with a aperture size of 0.6 mm. This represents the throughput of a sieve with a aperture size of 0.3 mm.

[0019] Based on the same inventive concept, the present invention also includes an electronic device comprising a processor; a memory for storing a computer program executed by the processor; wherein, when the processor executes the computer program, it implements the super-dense asphalt mixture design method described above.

[0020] To better understand and implement this invention, the following detailed description is provided in conjunction with the accompanying drawings. Attached Figure Description

[0021] Figure 1 A graph showing the functional relationship between diffusion flux and connectivity porosity; Figure 2 This is a flowchart of the ultra-dense asphalt mixture design method of the present invention; Figure 3 A flowchart illustrating the method for establishing the aggregate particle model in step S3A1; Figure 4 A comparative schematic diagram of the three-dimensional convex hull and aggregate particle model with different particle sizes; Figure 5 This is a schematic diagram of the discrete element model of the main skeleton structure; Figure 6 This is a distribution characteristic diagram of the void ratio of the mineral aggregate; Figure 7 This is a tomographic image of an asphalt mixture; Figure 8 A flowchart for detecting connectivity porosity; Figure 9 A diagram showing the porosity error distribution for different structural layers in different projects; Figure 10 The graph shows the relationship between the measured porosity error and the probability density. Figure 11 This is a mathematical distribution diagram of asphalt usage. Figure 12 A graph showing the relationship between asphalt content and the aggregate void ratio of the coarse aggregate main skeleton; Figure 13 The results of the multivariate correlation analysis are shown in the figure. Figure 14 This is a graph showing the relationship between water vapor diffusion flux and porosity. Detailed Implementation

[0022] The inventors carefully studied the current situation where asphalt mixtures cannot effectively prevent the steel plates of steel bridge decks from being corroded. They found that the current asphalt mixtures can only prevent liquid water penetration, but in reality, gaseous water in the environment can also diffuse into the asphalt mixture pavement layer and reach the interface of the steel bridge deck. Under the influence of changes in environmental conditions, it condenses into liquid water and eventually covers the surface of the steel plate, causing an electrochemical reaction that leads to rusting of the steel plate.

[0023] Further research by the inventors revealed that most diffusion phenomena in nature are non-steady-state diffusions, meaning that the diffusion phenomenon changes with both time and space. However, asphalt mixtures continuously undergo water vapor diffusion during long-term service, especially for asphalt mixtures used in steel bridge deck pavements in cross-river and cross-sea environments, where the mixtures are essentially saturated. The gaseous water diffusion in the asphalt mixture of steel bridge deck pavements is a steady-state penetrating diffusion, thus revealing the change in water concentration over time: , Where C is the concentration of water.

[0024] Simultaneously, combining Fick's first and second laws, as well as the diffusion boundary conditions of asphalt mixtures, the diffusion model of gaseous water in asphalt mixtures is obtained as follows: , Where D is the water diffusion coefficient of the asphalt mixture, C is the water concentration, and x, y, and z are the three directions of the asphalt mixture.

[0025] At the same time, combining the two formulas about diffusion flux in Fick's first law: , and , in, For diffusion flux.

[0026] Solving the gaseous water diffusion model yields the water diffusion coefficient of the asphalt mixture as follows: , Where D is the water diffusion coefficient of the asphalt mixture, C is the water concentration, and x, y, and z represent the three directions of the asphalt mixture. For diffusion flux.

[0027] According to the equivalence principle, the diffusion coefficient of a homogeneous and isotropic asphalt mixture can be expressed as: , According to fundamental thermodynamic theory, at a constant temperature, the concentration of water can be expressed as: , Where D is the water diffusion coefficient of the asphalt mixture, C is the water concentration, and x, y, and z represent the three directions of the asphalt mixture. For diffusion flux, The vapor pressure of water. Atmospheric pressure. The molar density of air, This represents the molecular mass of gaseous water. The vapor pressure of water and the relative humidity of water vapor are also considered. satisfy , The saturated vapor pressure of gaseous water at 25ºC is 3169 Pa; The molecular weight of gaseous water is 18.015. .

[0028] From the above formulas, the concentration difference of gaseous water humidity at one end and the other end of the infinitesimal element can be obtained as follows: , At the same time, according to Fick's second law, the formula for the mass change of matter within a infinitesimal element is: , Based on the relationship between changes in mass and concentration, the initial mass within the infinitesimal element is: After time Afterwards, the concentration of matter within the infinitesimal element becomes The formula for the change in mass of matter within a infinitesimal element is obtained as follows: , The combined results are: , By incorporating the concentration difference, the water diffusivity and diffusion flux of the asphalt mixture under the simplified homogeneous isotropic condition can be determined. The expression for the water diffusivity is: , The expression for the diffusion flux of the asphalt mixture is: , Where D is the water diffusion coefficient of the asphalt mixture, in m² / s; R is the universal gas constant, 8.314. T is the Kelvin temperature of 298.15 K. The distance of water vapor diffusion from the specimen is the specimen thickness, in meters (m); A is the self-measured diffusion area, in square meters (m²). This represents the rate of change of the device's mass loss over time, as measured in the water vapor diffusion experiment. The relative humidity difference at both ends of the asphalt mixture specimen is used to drive water vapor diffusion. This represents the molecular mass of gaseous water.

[0029] Subsequently, the interconnected porosity variable of asphalt mixture is introduced into the model of water vapor diffusion coefficient and diffusion flux, thereby quantifying the relationship between the diffusion coefficient and the interconnected porosity structure inside the asphalt mixture. This yields a model for the water vapor diffusion coefficient and diffusion flux of asphalt mixture after correction for the influence of the internal porosity structure. The expressions for this water vapor diffusion coefficient and diffusion flux are as follows: , and , in, Let be the interconnected void ratio of the asphalt mixture, and m and k be the model coefficients, respectively.

[0030] Please see Figure 1 , Figure 1 This is a graph showing the functional relationship between the diffusion flux and the connectivity porosity. The inventors then studied the water vapor diffusion performance of 100 EA-10 Marshall specimens of epoxy asphalt mixture obtained from on-site core samples of a typical steel bridge deck pavement, based on the aforementioned model of the water vapor diffusion coefficient and diffusion flux of the asphalt mixture. A nonlinear curve fitting method was used to construct the diffusion flux of the connectivity porosity of the asphalt mixture, obtaining the functional relationship between the diffusion flux and the connectivity porosity. The expression for this functional relationship is: , Where a and b are both function fitting parameters, The connected void ratio of the asphalt mixture is given in Table 1 below: Table 1

[0031] get , The solution is obtained based on the relationship between the diffusion flux and the water diffusion coefficient: , The water diffusion coefficient is calculated using Fick's first and second laws as follows: , Where D is the water diffusion coefficient model of asphalt mixture. It represents the interconnected void ratio of asphalt mixtures.

[0032] The inventors discovered that while dense-graded asphalt mixtures have a low porosity, they still contain randomly distributed voids that are partially or completely interconnected. This allows moisture from the natural environment to diffuse from top to bottom across the steel bridge deck pavement, preventing the existing pavement from achieving the expected waterproofing performance. Based on a water diffusion model, the inventors proposed an ultra-dense asphalt mixture. By achieving a zero interconnected porosity, the water diffusion coefficient of the asphalt mixture is reduced to zero, placing it in an ultra-dense state and effectively preventing the diffusion of gaseous water within the asphalt mixture.

[0033] Based on the formation conditions and properties of ultra-dense asphalt mixtures, the inventors further designed a design method for ultra-dense asphalt mixtures. This method is based on the relationship between interconnected void ratio and void ratio, as well as digital simulation of aggregate mixing state. Combined with the CAVF volume design method of asphalt mixtures, the mineral aggregate ratio of asphalt mixtures under ultra-dense state is designed to effectively prevent gaseous water diffusion.

[0034] The following description is provided through various embodiments.

[0035] Example 1 Based on this, the present invention provides a design method for ultra-dense asphalt mixtures, which is used to design asphalt mixtures with zero water diffusion rate, thereby improving the service life of steel bridge deck plates.

[0036] Please see Figure 2 , Figure 2 This is a flowchart of the ultra-dense asphalt mixture design method of the present invention. The ultra-dense asphalt mixture design method of the present invention includes the following steps: S1: Select the gradation type of the asphalt mixture, and select the coarse aggregate composition specifications, fine aggregate, and mineral powder types and specifications according to the gradation type. The gradation type refers to the particle size range of various aggregates in the asphalt mixture; different gradation types have corresponding particle size ranges. The number of coarse aggregate composition specifications refers to the number of coarse aggregate particle sizes in the asphalt mixture, and the composition specifications are the particle size specifications of each selected coarse aggregate particle. In this embodiment, the gradation type is EA-10, and the number of coarse aggregate composition specifications is two types, selecting basalt with particle sizes of 3~5mm and 5~10mm; at the same time, manufactured sand with a particle size specification of 0-3mm is selected as fine aggregate, limestone mineral powder is selected as mineral powder filler, and A-70 base asphalt and epoxy resin are selected as binders to design the asphalt mixture.

[0037] S2: Obtain the physical property parameters of each selected aggregate. These physical property parameters include the bulk density and apparent density of the coarse aggregate, fine aggregate, and mineral powder, as well as the density of the asphalt binder. In this embodiment, the bulk density of the two selected coarse aggregate particle sizes is shown in Table 2 below: Table 2

[0038] S3A: Based on the compositional relationship between the mineral materials, set the mass percentage range of coarse aggregates of each particle size and generate at least two sets of mineral material ratios; at the same time, establish a discrete element model of the main skeleton to calculate the mineral gap ratio and main skeleton compaction density of the coarse aggregate main skeleton under each mineral material ratio.

[0039] Step S3A specifically includes: S3A1: Establish corresponding aggregate particle models based on the particle size specifications of each coarse aggregate. The aggregate particle models adopt the three-dimensional ODEC algorithm based on the ODEC algorithm extension, namely the meta-overlapping particle family algorithm. Its core idea is to arrange a series of overlapping inscribed spheres along the skeleton structure to approximate the real outer contour of the particles, which can significantly improve the geometric accuracy and computational efficiency of three-dimensional modeling of irregular particles.

[0040] The aggregate particle model is established as follows: Please see Figure 3 , Figure 3 This is a flowchart illustrating the process of establishing the aggregate particle model in step S3A1. S3A101: Select coarse aggregate particles based on the chosen particle size. To improve randomness, multiple coarse aggregate particles of each particle size are selected. In this embodiment, 500 coarse aggregate particles of each particle size (3-5mm and 5-10mm) are selected.

[0041] S3A102: X-ray CT scans are performed on each of the coarse aggregate particles of different sizes to obtain a set of tomographic images for each coarse aggregate particle. During the X-ray CT scan imaging of each coarse aggregate particle, imaging is performed at regular depths from the top to the bottom of each particle. Each coarse aggregate particle yields a set of tomographic images from different depths, with each set including multiple tomographic images evenly spaced from the top to the bottom of the coarse aggregate. In the X-ray CT tomographic imaging, X-rays undergo photoelectric and Compton scattering and pair production at the molecular or atomic level, causing a significant attenuation of X-ray energy. Since the degree of X-ray attenuation is related to the properties of the medium, the grayscale distribution on the CT image reflects the distribution of the attenuation coefficient of the material within the object under test at that energy level.

[0042] S3A103: Obtain the corresponding equivalent volume sphere diameter based on the tomographic image set of each coarse aggregate particle.

[0043] S3A104: Generate a sphere based on the diameter of the equivalent volume sphere. In this embodiment, the sphere is generated and subsequent modeling steps are performed using Matlab.

[0044] S3A105: Generate multiple random points on the surface of the sphere and connect them to form a three-dimensional convex hull. The expression for the Cartesian spatial coordinates of the random points generated on the surface of the sphere is: , in, , Let be the angles between the lines connecting the random point and the center of the sphere and the z-axis and x-axis, respectively. and It follows a uniform distribution.

[0045] S3A106: Voxelize the surface and interior of the three-dimensional convex hull. A voxel is a basic unit in three-dimensional space, which can be regarded as an extension of two-dimensional pixels in three dimensions, represented as a cube distributed on an orthogonal grid. Voxelization can discretize the three-dimensional convex hull obtained from coarse aggregate particles into a set of voxels, which facilitates subsequent digital processing.

[0046] The specific operation of surface voxelization of the three-dimensional convex hull is as follows: establish a voxel space composed of orthogonal meshes, then determine the spatial relationship between each voxel and the closed surface of the three-dimensional convex hull, and retain the voxels that intersect with the closed surface, thus realizing the surface voxelization of the three-dimensional convex hull.

[0047] The specific operation of internal voxelization of the three-dimensional convex hull is as follows: the internal voxels are filled by scanning or ray casting to achieve internal voxelization of the three-dimensional convex hull.

[0048] The orthogonal mesh is a cube, and its function expression is: , Among them voxels Centered on The cube, where h is the resolution. Let be the side length of the cube, and let the spatial coordinates of the voxel be... .

[0049] The functional expression of the three-dimensional convex hull after surface voxelization is: , The functional expression of the three-dimensional convex hull after internal voxelization is: .

[0050] S3A107: Extract the skeleton points of each 3D convex hull. Specifically, the skeleton points are extracted by calculating the Euclidean distance from each voxel in the 3D convex hull to the boundary, and extracting voxels whose Euclidean distance satisfies the condition of being the point with the maximum distance from the boundary among the local 26 neighborhood voxels. These voxels are then used as the skeleton points of each 3D convex hull. The formula for calculating the Euclidean distance is: , in, For Euclidean distance, For each element, Let the coordinates be the center coordinates of the voxel. Let P be the boundary and P be the surface. The point on top.

[0051] The set of skeleton points for each 3D convex hull is: , in, For 26 domain voxels, Let the coordinates be the center coordinates of the voxel. For the internal voxels of the 3D convex hull, The distance is the Euclidean distance. Where, for each skeleton point... The radius of the associated skeleton point is r(s) = d(v).

[0052] Please see Figure 4 , Figure 4This is a comparative schematic diagram of three-dimensional convex hulls and aggregate particle models with different particle sizes. S3A108: Generate the maximum inscribed sphere covering each skeleton point of the three-dimensional convex hull to obtain the aggregate particle model of each coarse aggregate particle. The maximum inscribed sphere covers the outer contour of each inscribed sphere to form the shape of the convex hull. The outer envelope of each inscribed sphere together approximates the geometric contour of the aggregate particle. In generating each maximum inscribed sphere, to reduce the number of spheres required, this invention further adopts a space filling optimization strategy. This space filling optimization strategy arranges the skeleton points of the three-dimensional convex hull in descending order of radius; then, the spheres with the current maximum radius are placed sequentially to cover the space not occupied by the previous spheres; when the sphere filling volume loss rate reaches a certain threshold, space filling is stopped, thus obtaining the aggregate particle model. The filling volume loss rate is the ratio of the volume filled by the spheres to the volume of the three-dimensional convex hull, and its calculation formula is: , in, For the filling volume loss rate, Let the volume of each sphere be , The volume is the 3D convex hull volume. In this embodiment, the filling volume loss rate is 5%. In this embodiment, the clump (cluster) block generation technology in PFC3D is used to cover the maximum inscribed sphere at each skeleton point.

[0053] S3A109: Assign bulk density to each aggregate particle model corresponding to its particle size. In this embodiment, according to Table 2, the bulk density of coarse aggregate particles with a particle size of 3~5mm is 2.876g / cm3, and the bulk density of coarse aggregate particles with a particle size of 5~10mm is 2.888g / cm3.

[0054] S3A2: Based on the compositional relationships between various mineral materials, set the mass percentage ranges for fine aggregates, mineral powder, and coarse aggregates of different particle sizes, and generate at least two sets of coarse aggregate ratios. Each set of coarse aggregate ratios is a collection of the mass percentages of coarse aggregates of each particle size generated according to their respective mass percentage ranges. Under each set of coarse aggregate ratios, coarse aggregates of each particle size have different mass percentages.

[0055] In this embodiment, the mass percentage range of coarse aggregate with a particle size of 5-10 mm is set to 10%-60%, the mass percentage range of coarse aggregate with a particle size of 3-5 mm is set to 5%-30%, the mass percentage range of fine aggregate with a particle size of 0-3 mm is set to 5%-75%, and the mass percentage range of mineral powder is set to 2%-15%.

[0056] Please see Figure 5 , Figure 5This is a schematic diagram of the discrete element model of the main skeleton structure. S3A3: Using the aggregate particle models, discrete element models of the asphalt coarse aggregate main skeleton structure with each coarse aggregate ratio are established respectively. When establishing the discrete element model of the asphalt coarse aggregate main skeleton structure with each coarse aggregate ratio, a model space is first established. Then, according to the coarse aggregate ratio, the aggregate particle model of the corresponding particle size is called and randomly placed in the model space until the cumulative volume of each particle size meets a certain threshold. Each aggregate particle model is randomly called from all aggregate particle models under the specified particle size. The threshold of the cumulative volume is set according to a threshold function, which is: , in, The volume of the model space. For cumulative volume, It represents the volume of the smallest aggregate particle model within a given particle size specification.

[0057] In this embodiment, the Bubble Pack filling algorithm of the PFC5.0 platform is used to fill the internal space of the aggregate aggregate shape, so as to achieve a more effective characterization of the angularity and convex and concave surface information of the aggregate particles.

[0058] Furthermore, when generating the discrete element model of the main skeleton structure in the model space, the aggregate particle model suffers from problems such as a fixed centroidal axis direction rather than random distribution and geometric overlap during stacking. To address this issue, the present invention also employs a reduction-expansion method, setting a reduction factor. The size of the aggregate particle model is linearly reduced to Subsequently, the surface normal stiffness of the aggregate particle model was temporarily set to... To ensure that particles bounce off each other upon contact, surface friction and damping are temporarily set to zero, allowing particles to slide freely and rebound almost elastically upon contact, thus randomizing the initial arrangement and orientation of each aggregate particle model. Subsequently, the Cycle command is used to progressively expand the aggregate particle model to its original volume, causing random rotation and bounce-off of overlapping areas within the model space. This progressive expansion is based on a step-by-step magnification factor, which is: , This indicates that expansion to the original volume is achieved through N-level scaling. In this embodiment, Take 200, N is 9, and finally get It is 1.80.

[0059] S3A4: A linear contact model is used to set the microscopic contact stiffness of each aggregate particle model within the model space. A top plate is provided at the top of the model space, and an initial top plate pressure is set on the top plate. In this embodiment, the initial top plate pressure is 7 MPa; simultaneously, based on the elastic modulus and Poisson's ratio of basalt aggregate being taken as 56 GPa and 0.3 respectively, the habitual contact stiffness of each aggregate particle model is obtained as follows: , in, For the mesoscopic normal contact stiffness of a single particle element, For the mesoscopic tangential contact stiffness of a single particle unit, For elastic modulus, The length of the elastic beam is the sum of the radii of the two contacting particle elements. Therefore, the length of a single coarse aggregate particle model in the model space of this embodiment can be obtained. It is 1.29e9 N / m. The value is 4.95e8 N / m, and the coefficient of friction is taken as 0.3.

[0060] S3A5: Calculate the total volume of all coarse aggregate particles for each coarse aggregate ratio. The formula for calculating the total volume is: , in, This represents the total volume of the coarse aggregate particle model for each particle size specification. This represents the cumulative volume of a coarse aggregate particle model with one of the particle sizes in the model space.

[0061] S3A6: Based on the discrete element model of the asphalt coarse aggregate main skeleton structure under each coarse aggregate ratio, calculate the aggregate void ratio and main skeleton compaction density of the coarse aggregate main skeleton respectively. The formula for calculating the aggregate void ratio of the coarse aggregate main skeleton is: , in, The aggregate void ratio is the main skeleton of coarse aggregate. The volume of the model space. This represents the total volume of the aggregate particle model for coarse aggregates of various particle sizes.

[0062] The main skeleton compaction density is obtained based on the relationship function between the aggregate void ratio and the main skeleton compaction density of the coarse aggregate main skeleton. The relationship function between the aggregate void ratio and the main skeleton compaction density of the coarse aggregate main skeleton is expressed as follows: , in, The aggregate void ratio is the main skeleton of coarse aggregate. The main frame is tightly packed with high density. The apparent density is the composite density of coarse aggregates of various specifications mixed according to the specified coarse aggregate ratio. The formula for the composite apparent density is: , in, The apparent density of the composite material is the result of mixing coarse aggregates of various specifications according to the coarse aggregate ratio. This represents the mass percentage of coarse aggregate. The mass percentage of each particle size specification; the formula for the mass percentage of the coarse aggregate is: , in, This represents the mass percentage of coarse aggregate. Let n represent the mass percentage of each particle size, and n be the number of constituent particle sizes of the coarse aggregate. The mass percentage of the coarse aggregate is the mass percentage of the coarse aggregate relative to the fine aggregate and mineral powder, i.e., the sum of the mass percentages of the coarse aggregate, fine aggregate, and mineral powder is 100%. The sum of the mass percentages of each particle size is 100%.

[0063] S2A7: Based on the compositional relationship between the mineral materials and the mass percentage range of fine aggregate and mineral powder, at least two sets of mineral material ratios are generated for each set of coarse aggregate ratios. The mineral material ratios are expressed as different mass percentages of coarse aggregate, fine aggregate, and mineral powder for each particle size, and the sum of the mass percentages of coarse aggregate, fine aggregate, and mineral powder for each particle size is 100%.

[0064] The compositional relationship between the coarse aggregate, fine aggregate, mineral powder, asphalt, and mineral aggregate is expressed as follows: , , in, This represents the mass percentage of coarse aggregate. This represents the mass percentage of fine aggregate. This represents the mass percentage of mineral powder. Here, n represents the mass percentage of each particle size, and n is the number of coarse aggregate components. sieve aperture diameter This represents the lower limit of the sieve aperture throughput. sieve aperture diameter This represents the upper limit of the sieve aperture throughput.

[0065] In this embodiment, based on a set mass percentage range, 1000 sets of ore proportions with different mass percentage ratios are generated. The 1000 sets of ore proportions generated in this embodiment are shown in Table 3 below: Table 3

[0066] The gradation distribution of each mineral component is shown in Table 4 below: Table 4

[0067] Please see Figure 6 , Figure 6 This is a distribution characteristic diagram of the aggregate void ratio. In this embodiment, simulations were performed on 1000 different aggregate proportions to obtain the aggregate void ratio of the main coarse aggregate skeleton for each coarse aggregate proportion. Its distribution characteristics are shown below. Figure 6 As shown, Figure 6 Furthermore, mathematical characteristic tests were conducted on the distribution characteristics using methods including normal distribution, log-normal distribution, Gamma distribution, and Weibull distribution. Based on the distribution curve characteristics and the result that the KS statistical test value is greater than 0.1, it can be concluded that the mineral gap ratio of the main skeleton of the randomly synthesized coarse aggregate approximately follows a Weibull distribution.

[0068] Furthermore, this invention employs 10 different sets of ore proportion parameters and places the corresponding ore into a Marshall-style mold with the same dimensions as the model space to test the ore void ratio. During the test, 100 Marshall compactions are performed, and the ore void ratio is measured. Simultaneously, the void ratio is compared with that obtained by simulating the ore proportion using this invention. The comparison results are shown in Table 5 below: Table 5

[0069] As can be seen from the table, the measured void ratio of the ore is lower than that of the ore obtained by simulation using this method, but the Pearson correlation coefficient between the two reaches 0.91.

[0070] Therefore, a linear relationship is used to correct the pressure on the top plate in the coarse aggregate discrete element model. The linear relationship correction function is: , in, To correct the pressure on the top plate, The initial top plate pressure, In this embodiment, the pressure value corresponding to the preset initial top plate pressure is corrected from 7MPa to 7.6MPa.

[0071] S3B: Calculate the design void ratio based on the gradation type and coarse aggregate composition specifications. This step S3B specifically includes: S3B1: Select asphalt mixture samples of the same gradation type as chosen. In this embodiment, 100 EA-10 Marshall specimens of epoxy asphalt mixture for steel bridge decks, each 100 mm in diameter and 35 mm thick, are used.

[0072] Please see Figure 7 , Figure 7 S3B2: X-ray CT scans were performed on each of the aforementioned asphalt mixture samples to obtain multiple sets of tomographic images. During the tomographic imaging process, a combination of ring segmentation and the Great Law algorithm was used. This method divides the complete image into continuous but non-overlapping ring sub-images. The Great Law algorithm is then used to segment each ring region. Finally, the images of each ring region form a complete tomographic image. Each asphalt mixture sample generates a set of multiple tomographic images with uniform depth intervals from the top to the bottom of the coarse aggregate. For example... Figure 7 As shown, Figure 7 (a) is a schematic diagram of the ring-shaped blocks. Figure 7 (b) is a schematic diagram of the DaLaw algorithm for image segmentation. Figure 7 (c) is a tomographic image of the improved large law algorithm of the circumferential segmentation method of the present invention.

[0073] The basic idea of ​​the large-law algorithm is to maximize the inter-class variance. When detecting asphalt mixtures, the multi-phase classification problem of asphalt mixtures is transformed into multiple one-to-two classifications, sequentially segmenting the background, aggregates, asphalt mortar, and voids in the cut-off layer image to achieve threshold segmentation of the asphalt mixture phases. However, asphalt mixtures often exhibit uneven brightness in the radial direction in CT scan images, which can easily lead to black dead pixels in the CT image, affecting the phase segmentation quality. The annular block method divides the CT image into annular regions of different brightness levels. Therefore, this application combines the large-law algorithm and the annular block method to effectively improve the imaging quality of asphalt mixtures in X-ray CT scans.

[0074] S3B3: Based on the tomographic images of each group, the porosity and connectivity porosity of each asphalt mixture sample are detected. The porosity is obtained by statistically analyzing the ratio of the porosity area to the total area in a single CT tomographic image, and further by calculating the mean porosity of all scanned tomographic images. The expression is as follows: , in, The porosity of each specimen, For the first The area of ​​the gaps in a CT scan tomographic image. For the first The total area of ​​the CT scan tomographic images. This represents the number of CT scan tomographic images.

[0075] Please see Figure 8 , Figure 8 This is a flowchart for detecting the connectivity porosity. The steps for detecting the connectivity porosity are as follows: S3B301: Convert all the fault images into binary image data and select the fault image set of one of the asphalt mixture samples.

[0076] S3B302: Take the top-level fault image from the fault image set.

[0077] S3B303: Check the grayscale value of the current pixel. If the grayscale value of the current pixel is 0, proceed to step S3B04. If the grayscale value of the current pixel is not 0, check the grayscale value of the next pixel. After traversing all pixels, proceed to step S3B05.

[0078] S3B304: Check the connectivity of the 8-neighborhood of the current pixel. If there are pixels with a gray value of 0 in the 8-neighborhood, mark these pixels with a gray value of 0 as the same gap object.

[0079] S3B305: Determine whether the pixels of the current tomographic image have been traversed. If the current pixel is the last pixel, proceed to step S3B306; if the current pixel is not the last pixel, repeat steps S3B303 to S3B304.

[0080] S3B306: Determine whether the current tomographic image is the bottom layer of the tomographic image. If yes, proceed to step S3B307; otherwise, select the tomographic image located in the next layer and repeat steps S3B303 to S3B305.

[0081] S3B307: Obtain the tomographic image located at the top layer.

[0082] S3B308: Get a gap object in the current tomographic image.

[0083] S3B309: Check if there are any pixels with a gray value of 0 in the 8-neighborhood of the gap object in the next layer of tomographic image. If so, mark them as vertically connected objects.

[0084] S3B310: Determine whether the void objects in the current fault image have been traversed. If not, take the next void object and repeat step S3B309; if the traversal is completed, execute step S3B311.

[0085] S3B311: Determine whether the current fault image is the bottom fault image. If so, proceed to step S3B312. If not, take the fault image located in the next layer and repeat steps S3B309 to S3B310.

[0086] S3B312: Obtain the tomographic image located at the bottom layer.

[0087] S3B313: Get a gap object in the current tomographic image.

[0088] S3B314: Check if there are any pixels with a gray value of 0 in the 8-neighborhood of the gap object in the previous layer tomographic image. If they exist, mark them as connected; otherwise, mark them as closed.

[0089] S3B315: Determine whether the void objects in the current fault image have been traversed. If not, take the next void object and repeat step S3B313; if the traversal is complete, execute step S3B316.

[0090] S3B316: Determine whether the current fault image is the top-level fault image. If yes, proceed to step S3B317; otherwise, take the fault image located on the previous level and repeat steps S3B313 to S3B312.

[0091] S3B317: Check whether there is a connection mark in the gaps of each top fault image. If there is, mark the gap with the gap mark as a connected gap. If there is no connection mark, mark it as a closed gap.

[0092] S3B318: Calculate the connectivity porosity based on the connectivity gaps.

[0093] S3B4: A nonlinear curve fitting method is used to establish a correspondence between the porosity and the connected porosity of each asphalt mixture sample. The designed porosity threshold function is a functional relationship between the connected porosity (when it is zero) and the porosity. This correspondence includes multiple function fitting parameters. The relationship between the connected porosity and the porosity satisfies an exponential function as follows: , in, The interconnected porosity of asphalt mixtures. The porosity of asphalt mixture. , and These are the parameters for fitting the function.

[0094] The function fitting parameters , and The range of values ​​for is shown in Table 6 below: Table 6

[0095] S3B5: Establish a design void ratio threshold function for asphalt mixtures based on the aforementioned correspondence, and establish the design void ratio threshold function. The formula for the design void ratio threshold is: , in, To design the porosity threshold, , and The parameters are for fitting the function, and , .

[0096] S3B6: Calculate the standard deviation and estimated value of each of the function fitting parameters according to their confidence intervals, and calculate the variance and standard deviation of the design porosity threshold in combination with the function of the design porosity threshold.

[0097] The design void ratio threshold is the maximum allowable void ratio when the connected void ratio is 0; that is, when the measured void ratio of the asphalt mixture is lower than the design void ratio threshold, the connected void ratio is 0. Different confidence intervals represent different guarantee rates. Refer to the function fitting parameters in Table 6. , and The prediction range is defined as the range within which the estimated value fluctuates by 1.96 standard deviations. In this embodiment, the standard deviations and estimated values ​​of each function fitting parameter are calculated based on the 95% confidence interval. The function fitting parameters are then used as the basis for the prediction. For example, its prediction range is expressed as Therefore, its standard deviation is obtained by dividing the tail value of the lower limit of the 95% confidence interval by 1.96. The estimated value is Similarly, calculate the standard deviation of the fitting parameters for the other two functions to obtain... and The estimated values ​​are respectively and Therefore, the estimated value of the design porosity threshold is calculated as follows: , in, , and These are the fitting parameters. , and The estimated value is 1.178.

[0098] The variance of the design porosity threshold is further calculated using the Delta method, and the variance formula is as follows: , in, , , , , and The parameters are respectively the function fitting parameters. , and The standard deviation within the 95% confidence interval; based on the estimated values ​​of each fitted parameter, the following is obtained: , and Therefore, the variance of the design porosity threshold is calculated to be 0.03326, and the standard deviation of the design porosity threshold is further obtained. .

[0099] S3B7: Calculate the design void ratio threshold based on the variance and standard deviation of the design void ratio threshold. In this embodiment, based on the estimated value, variance, and standard deviation of the design void ratio threshold, the predicted range of the design void ratio threshold that guarantees a 95% connectivity void ratio of 0 in the asphalt mixture is obtained as follows: , in, To estimate the porosity threshold, The standard deviation is used to design the porosity threshold.

[0100] The predicted range for the design porosity threshold with a 95% guarantee rate is [0.821%, 1.535%]. Based on the relationship between the connected porosity and the porosity, the design porosity threshold is set at 1.535%. This means that when the measured porosity of the asphalt mixture is less than 1.535%, there is a 95% probability that the connected porosity of the asphalt mixture will be zero, allowing the asphalt mixture to reach an ultra-dense state and effectively preventing water vapor from diffusing within the asphalt mixture.

[0101] S3B8: The target porosity is calculated based on the measured porosity error distribution and the designed porosity threshold. The measured porosity error is the random deviation between the actual porosity of the asphalt mixture and the designed porosity due to construction variability. This random deviation follows a normal distribution. The designed porosity threshold is a correction based on the measured porosity error to ensure that the porosity of the asphalt mixture after random deviations during actual construction still reaches the designed porosity threshold.

[0102] In this embodiment, a nucleus-free density meter was used to measure the measured void ratio of epoxy asphalt mixtures at different construction stages of different structural layers in different projects. The sampling method for the measuring points was as follows: the spacing between measuring points perpendicular to the station number was set to 1m, and the spacing between measuring points along the station number was also set to 1m. The number of measuring points for each structural layer was no less than 400. The measured void ratio error was calculated by combining the designed void ratio of each epoxy asphalt mixture measuring point with the measured void ratio. Please refer to [link to relevant documentation]. Figure 9 , Figure 9 This is a distribution map of porosity errors for different structural layers in different projects. Figure 9 (a) is the upper paving layer of Project A. Figure 9 (b) is the lower layer of paving for Project A. Figure 9 (c) is the upper layer of paving for Project B. Figure 9 (d) is the lower layer of paving for Project B. Figure 9 (e) is the upper layer of paving for Project C. Figure 9 (f) is the lower layer of paving for Project C. Figure 9 (g) is the upper layer of paving for Project D. Figure 9 (h) is the lower layer of paving for Project D. (By...) Figure 9 It can be seen that the histograms of the measured porosity errors of different structures in different projects are similar and bell-shaped, showing a normal distribution. At the same time, the mathematical distribution of the measured porosity error data of different structural layers in different projects was tested using the Shapiro-Wilk method. The mathematical distribution is shown in Table 7 below: Table 7

[0103] As shown in Table 7, the P-values ​​in the mathematical distribution are all greater than 0.05, indicating that the measured void ratio errors of epoxy asphalt mixtures with different structural layers in different projects strictly follow a normal distribution. Therefore, the measured void ratio error follows a normal distribution.

[0104] Please see Figure 10 , Figure 10 This is a graph showing the relationship between the measured porosity error and the probability density. Based on the fundamental principles of the normal distribution, in this embodiment, the measured porosity error fluctuation range is selected as [range missing]. At this point, the probability density reaches 99.73%, that is: , This leads to the conclusion that the asphalt mixture after construction achieves a target porosity of 99.73%. It should be less than the design porosity threshold of 1.535% determined according to the water vapor diffusion theory of epoxy asphalt mixtures, that is, the target porosity satisfies: , And further, we obtained: , in, For the target porosity, To design the porosity threshold.

[0105] The measured void ratio error distribution and corresponding target void ratio of epoxy asphalt mixtures with different structural layers in different projects were calculated. The measured error distribution is shown in Table 8 below: Table 8

[0106] As shown in Table 8, the target porosity of the upper layer epoxy asphalt mixture FAC-10 is 1.6%, which is the design target threshold. The range is approximately [0.62%, 0.85%], with a mean of approximately 0.70% and a standard deviation of approximately 0.10%; while the target void ratio for the epoxy asphalt mixture EA-10 in the lower layer of the pavement is 1.5%. The range is approximately [0.59%, 0.77%], with a mean of approximately 0.68% and a standard deviation of approximately 0.07%, slightly lower than the upper layer of the pavement. This is consistent with the functional positioning of the upper layer epoxy asphalt mixture as a surface anti-skid layer.

[0107] Therefore, in this embodiment, the designed void ratio threshold is 1.535%. The target void ratio needs to be between 0 and 0.85% to ensure that the asphalt mixture is in an ultra-dense state with a 99.73% guarantee rate. So the target void ratio is 0.85%. That is, when an asphalt formula with a measured void ratio of less than 0.85% is used for paving, there is a 99.73% guarantee rate that the measured void ratio of the paved asphalt mixture is less than the designed void ratio threshold, thus achieving an ultra-dense state with a connected void ratio of 0.

[0108] S4: Based on the relationship between each aggregate, the target void ratio, and the aggregate void ratio of the coarse aggregate main skeleton, and the asphalt content, calculate the asphalt content for each aggregate group. The Coarse Aggregate Void Design (CAVF) method, the mainstream design method for asphalt mixtures, can fully guarantee the interlocking structure of the coarse aggregate skeleton and achieve complete filling of the gaps in the main skeleton by the adhesive, which can significantly improve the comprehensive performance of asphalt mixtures.

[0109] The relationship between the various aggregates, the target porosity, the aggregate void ratio of the coarse aggregate main skeleton, and the amount of asphalt is expressed as follows: , in, This represents the mass percentage of coarse aggregate. Here, n represents the mass percentage of each particle size, and n is the number of coarse aggregate components. This represents the mass percentage of fine aggregate. This represents the mass percentage of mineral powder. This represents the percentage of asphalt used. For the target porosity, The main frame is tightly packed with high density. The apparent density of fine aggregate, The apparent density of the mineral powder is... The density of the asphalt binder. The aggregate void ratio is the main skeleton of coarse aggregate; For different sieve apertures, the throughput is... This refers to the sieve aperture diameter; only aggregates with particle sizes smaller than this sieve aperture diameter can pass through the corresponding sieve aperture. sieve aperture diameter This represents the lower limit of the sieve aperture throughput. sieve aperture diameter This represents the upper limit of the sieve aperture throughput.

[0110] In this embodiment, the mathematical distribution of asphalt content in ultra-dense asphalt mixtures under the EA-10 gradation type and its relationship with the aggregate void ratio of the coarse aggregate main skeleton are analyzed. Please refer to [link to relevant documentation]. Figures 11 to 12 , Figure 11 This is a mathematical distribution diagram of asphalt usage. Figure 12 This is a graph showing the relationship between asphalt content and the void ratio of the coarse aggregate main skeleton. Figure 11 Analysis shows that using the KS method to test for normality, the p-value is greater than 0.05, indicating that it approximately conforms to a normal mathematical distribution; Figure 12 Analysis shows that the relationship between asphalt content and the void ratio of the coarse aggregate main skeleton is not a linear mapping relationship, but rather presents as follows: , in, This represents the percentage of asphalt used. The interstitial ratio of the coarse aggregate main skeleton.

[0111] Therefore, it can be seen that the greater the coarse aggregate void ratio of asphalt mixture, the greater the asphalt content tends to be.

[0112] S5: Calculate the powder-to-binder ratio and oil film thickness for each group of aggregate proportions, and select aggregate proportions with powder-to-binder ratios and oil film thicknesses within a preset range as the specific parameters of the asphalt mixture. The formula for calculating the powder-to-binder ratio is: , Wherein, FB represents the powder-to-resin ratio. This represents the passing rate of a sieve with a mesh size of 0.075 mm. The expression for the effective asphalt content is: , in, This represents the percentage of asphalt used. This represents the proportion of asphalt binder absorbed by the aggregate in the asphalt mixture. This represents the sum of the percentages of various mineral aggregates in the total mass of the asphalt mixture. The relationship between the sum of these percentages and the amount of asphalt used is as follows: , The expression for the proportion of asphalt binder absorbed by the aggregate in the asphalt mixture is: , in, The effective relative density of the mineral. The total bulk relative density of all mineral materials. This represents the relative density of asphalt.

[0113] The formula for calculating the oil film thickness is: , in, DA represents the effective asphalt content in the asphalt mixture, and DA represents the asphalt film thickness. Let SA be the relative density of asphalt, and SA be the specific surface area of ​​all aggregates. The unit of the specific surface area of ​​all aggregates is m² / kg, and its expression is: , in, The aggregate surface area coefficient for each particle size range The formula for the relative density of asphalt is: , in, The relative density of asphalt. This represents the density of the asphalt binder.

[0114] In this embodiment, the powder-to-binder ratio is set to a range of 1.5-2.0, and the minimum oil film thickness is 6-9. .

[0115] Example 2 The design method for ultra-dense asphalt mixture in Example 2 is basically the same as that in Example 1, except that in step S5, the asphalt content for each aggregate proportion is calculated based on a multiple linear regression model relating the mass percentage of each aggregate, the interstitial ratio of the coarse aggregate main skeleton, and the passing rate of each sieve size. The expression for the multiple linear regression model is as follows: , in, This represents the percentage of asphalt used. The aggregate void ratio is the main skeleton of coarse aggregate. This refers to the mass percentage of coarse aggregate with a particle size of 5-10 mm. This represents the mass percentage of fine aggregate. The throughput of a sieve with a aperture size of 9.5 mm. The throughput of a sieve with a aperture size of 4.75 mm. The throughput of a sieve with a aperture size of 2.36 mm. The throughput of a sieve with a aperture size of 1.18 mm. The throughput of a sieve with a aperture size of 0.6 mm. This represents the throughput of a sieve with a aperture size of 0.3 mm.

[0116] Please see Figure 13 , Figure 13 The image shows the results of the multivariate correlation analysis. In this embodiment, asphalt content is used as the dependent variable, and the mass percentage of each aggregate, the passing rate of different sieve sizes, and the interstitial ratio of the coarse aggregate main skeleton are used as independent variables. Origin was used to perform multivariate correlation analysis and construct a multiple linear regression model. Figure 13 The results of the multivariate correlation analysis, the regression model parameters of the multiple linear regression model, and the analysis of variance are shown in Table 9 below: Table 9

[0117] according to Figure 13 Analysis shows that the asphalt content of ultra-dense asphalt mixtures is significantly affected not only by the void ratio of the coarse aggregate main skeleton, but also by the mass percentage of each aggregate and the passing rate of different sieve sizes. According to Table 9, the multiple linear regression model relationship between the asphalt content and various parameters of ultra-dense asphalt mixtures under the EA-10 gradation type is expressed as follows:

[0118] in, This represents the percentage of asphalt used. The aggregate void ratio is the main skeleton of coarse aggregate. This refers to the mass percentage of coarse aggregate with a particle size of 5-10 mm. This represents the mass percentage of fine aggregate. The throughput of a sieve with a aperture size of 9.5 mm. The throughput of a sieve with a aperture size of 4.75 mm. The throughput of a sieve with a aperture size of 2.36 mm. The throughput of a sieve with a aperture size of 1.18 mm. The throughput of a sieve with a aperture size of 0.6 mm. This represents the throughput of a sieve with a aperture size of 0.3 mm.

[0119] The amount of asphalt used under different aggregate ratios was calculated using a multiple linear regression model.

[0120] Furthermore, based on the tomographic image set of each coarse aggregate particle, the three-view particle roundness of each coarse aggregate particle is obtained, and the weighted roundness of each coarse aggregate particle is calculated. By comparing the weighted roundness of the aggregate particle model with the particle roundness of the corresponding aggregate particles, the modeling effect of the aggregate particles can be compared. The comparison results are shown in Table 10 below: Table 10

[0121] The formula for calculating the weighted roundness is:

[0122] in, , and These represent the particle roundness of each coarse aggregate particle in the two-dimensional figures of its three views in the x, y, and z directions. This represents the particle area of ​​each coarse aggregate particle in the two-dimensional graph shown in the x, y, and z views. These are designations for different coarse aggregate particles, representing different types of coarse aggregate particles. The particle roundness is the ratio of the area of ​​a circle with the same circumference as the actual particle to the actual particle area.

[0123] As shown in Table 10, the aggregate particle model established using the construction method of this invention has a better simulation effect than the actual aggregate particles.

[0124] Furthermore, the actual method of this application was used to design ultra-dense asphalt mixtures and to conduct performance tests.

[0125] Based on the EA-10 ingredient grade, the specific design parameters are shown in Table 11 below: Table 11

[0126] The volumetric parameters of the ultra-dense epoxy asphalt mixture Marshall specimens obtained from the design are shown in Table 12 below: Table 12

[0127] Next, the water vapor diffusion, high-temperature stability, low-temperature stability, and impact toughness of the ultra-dense epoxy asphalt mixture Marshall specimens were tested one by one: (1) A water vapor diffusion test was conducted on the obtained ultra-dense epoxy asphalt mixture Marshall specimen: a plastic cup with a height of 85 mm, an opening diameter of 55 mm and a volume of 230 ml was used, and the height of each 1 cm was marked to ensure that the water volume was consistent; water with controlled relative humidity was poured into the plastic cup to 4 cm from the inner scale line, and silicone oil was applied to the edge of the plastic cup. A clean annular silicone sheet with an inner ring diameter of 54 mm was used to tightly cover the edge of the plastic cup; then, paraffin wax was melted and sealed on the side wall and joint of the epoxy asphalt mixture specimen, and the solidified wax on the surface of the specimen was repaired. The entire device was cooled and dried to ensure that the paraffin wax was completely solidified, and the water vapor diffusion test device was assembled.

[0128] Multiple groups of Marshall specimens with target porosity ranging from 0.8% to 5.6% were selected and placed directly on the water vapor diffusion test device as experimental groups. Another Marshall specimen was sealed with water-based epoxy resin to make the effective porosity equivalent to 0% as a control group and placed on another water vapor diffusion test device.

[0129] Subsequently, a vacuum drying oven with an internal gas pressure of 31 kPa and phosphorus pentoxide were used to control the external relative humidity. The test parameters for the water vapor diffusion experiment are shown in Table 13 below: Table 13

[0130] The relationship between water vapor diffusion flux and porosity in the experimental and control groups was obtained through testing. Please refer to [the relevant documentation / reference]. Figure 14 , Figure 14 This is a graph showing the relationship between water vapor diffusion flux and porosity. From... Figure 14 As can be seen, the water vapor diffusion flux of the control group is 0. The maximum water vapor diffusion flux for the experimental group with a target porosity of less than 1.5% was 0.45. The minimum value is 0 The average value is approximately 0.21. Approaching 0 When the target porosity increased to 1.5%–3.0%, the maximum water vapor diffusion flux for the corresponding experimental group was 3.53. The minimum value is 1.44. The mean increased to approximately 2.41. Compared to the experimental group with a target porosity of less than 1.5%, the water vapor diffusion flux increased by 11.48 times; when the target porosity exceeded 3%, the maximum water vapor diffusion flux of the corresponding experimental group was 14.87. The minimum value is 6.42. The average value is approximately 9.95. .

[0131] Based on the experimental results, the water vapor diffusion flux of the asphalt mixture with a design void ratio of 0.85% designed using the design method of this invention in this test was approximately 0.089. The value is close to 0, indicating an ultra-dense asphalt mixture.

[0132] (2) High-temperature stability test of the obtained ultra-dense epoxy asphalt mixture Marshall specimen: According to the test procedure of highway engineering asphalt and asphalt mixture T0719-2011 (JTG E20-2011) rutting test method, since the steel bridge deck has high requirements for rutting resistance index, the dynamic stability test is carried out under the summer hot zone test at 70℃.

[0133] According to the test, the dynamic stability of the ultra-dense asphalt mixture designed using the method of this invention is greater than 10,000 times / mm, which meets the requirement of dynamic stability of not less than 6,000 times / mm required by the "Technical Specification for Design and Construction of Highway Steel Bridge Deck" (JTG / T3364-02), and has good high-temperature stability.

[0134] (3) Low-temperature stability test of the obtained ultra-dense epoxy asphalt mixture Marshall specimen: Low-temperature bending test was carried out according to the asphalt mixture bending test method of the Highway Engineering Asphalt and Asphalt Mixture Test Procedure T0719-2011 (JTG E20-2011). The test conditions were -10℃ and 50mm / min.

[0135] According to the test, the flexural limit strain of the ultra-dense asphalt mixture designed using the method of this invention is 4.17×10-3, which meets the requirement of the "Technical Specification for Design and Construction of Highway Steel Bridge Pavement" (JTG / T3364-02) that the ultimate failure is not less than 3×10-3, and it has good low-temperature stability.

[0136] (4) Impact toughness test of the obtained ultra-dense epoxy asphalt mixture Marshall specimen: Surface texture depth test was carried out according to the surface texture depth test method of the Highway Engineering Asphalt and Asphalt Mixture Test Procedure T0731-2011 (JTG E20-2011).

[0137] According to the test, the surface texture depth of the ultra-dense asphalt mixture designed using the method of this invention is 0.63mm, which meets the requirement of the "Technical Specification for Design and Construction of Highway Steel Bridge Pavement" (JTG / T3364-02) that the ultimate failure is not less than 3×10-3, and it has good low-temperature stability.

[0138] In summary, the design method of this invention can be used to design ultra-dense asphalt mixtures that simultaneously possess high-temperature stability, low-temperature stability, and impact toughness.

[0139] Compared with existing technologies, the ultra-dense asphalt mixture design method of the present invention can design asphalt mixtures with zero interconnected void ratio, achieve an ultra-dense state with zero diffusion coefficient of asphalt mixture, and effectively prevent gaseous water from diffusing in asphalt mixture.

[0140] This application may take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. Computer storage media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information may be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to: phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0141] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the embodiments of this application. The singular forms “a,” “the,” and “the” used in the embodiments and claims of this application are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that, unless otherwise stated, “a plurality” means two or more; the terms “first,” “second,” “third,” etc., are used only to distinguish and not to describe a particular order or sequence, nor should they be construed as indicating or implying relative importance. The term “and / or” as used herein refers to and includes any or all possible combinations of one or more associated listed items. When the above description relates to drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. In the description of this application, those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0142] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and the present invention also intends to include these modifications and variations.

Claims

1. A design method for ultra-dense asphalt mixtures, characterized in that, Includes the following steps: S1: Select the gradation type of the asphalt mixture, and select the composition specifications of coarse aggregate, the type and specifications of fine aggregate and mineral powder filler according to the gradation type; S2: Obtain the physical property parameters of each selected ore; S3A: Set the mass percentage range of coarse aggregates of each particle size according to the composition relationship between the minerals, and generate at least two sets of mineral proportions; at the same time, establish a discrete element model of the main skeleton to calculate the mineral gap ratio and main skeleton compaction density of the coarse aggregate main skeleton under each mineral proportion. S3B: Calculate the design void ratio threshold based on the gradation type and coarse aggregate composition specifications; Specifically, it includes: S3B1: Select asphalt mixture samples of the same gradation type as the selected gradation type; S3B2: X-ray CT scans were performed on each of the aforementioned asphalt mixture samples to obtain multiple sets of tomographic images; S3B3: Based on the fault images of each group, the porosity and connectivity porosity of each asphalt mixture sample are detected respectively; the statistical formula for the porosity is: ， in, The porosity of each specimen, For the first The area of ​​the gaps in a CT scan tomographic image. For the first The total area of ​​the CT scan tomographic images. The number of CT scan tomographic images; S3B4: Using a nonlinear curve fitting method, a correspondence between porosity and interconnected porosity is established based on the porosity and interconnected porosity of each asphalt mixture sample; the relationship between interconnected porosity and porosity satisfies an exponential function as follows: ， in, The interconnected porosity of asphalt mixtures. The porosity of asphalt mixture. , and The parameters are for fitting the function; S3B5: Establish a design void ratio threshold function for asphalt mixtures based on the aforementioned correspondence; the design void ratio threshold is the void ratio when the connected void ratio is zero; the design void ratio threshold function is: ， in, To design the porosity threshold, , and The parameters are for fitting the function, and , ; S3B6: Calculate the standard deviation and estimated value of each of the function fitting parameters according to their confidence intervals, and calculate the variance and standard deviation of the design porosity threshold in combination with the function of the design porosity threshold; S3B7: Calculate the design porosity threshold based on the variance and standard deviation of the design porosity threshold; S3B8: The target porosity is calculated based on the measured porosity error distribution and the designed porosity threshold. S4: Based on the relationship between each mineral aggregate, the target void ratio, the mineral aggregate void ratio of the coarse aggregate main skeleton, and the asphalt content, calculate the asphalt content for each group of mineral aggregates. S5: Calculate the powder-to-binder ratio and oil film thickness of each group of aggregate proportions, and select aggregate proportions with powder-to-binder ratio and oil film thickness within a preset range as specific parameters of asphalt mixture.

2. The design method for ultra-dense asphalt mixtures according to claim 1, characterized in that, In step S3B3, the specific steps for detecting the connectivity porosity are as follows: S3B301: Convert all the fault images into binary image data, and select the fault image set of one asphalt mixture sample; S3B302: Take the top-level fault image from the fault image set; S3B303: Check the grayscale value of the current pixel. If the grayscale value of the current pixel is 0, proceed to step S3B04. If the grayscale value of the current pixel is not 0, check the grayscale value of the next pixel. After traversing all pixels, proceed to step S3B05. S3B304: Check the connectivity of the 8-neighborhood of the current pixel. If there are pixels with a gray value of 0 in the 8-neighborhood, mark these pixels with a gray value of 0 as the same gap object. S3B305: Determine whether the pixels of the current tomographic image have been traversed. If the current pixel is the last pixel, proceed to step S3B306; if the current pixel is not the last pixel, repeat steps S3B303 to S3B304. S3B306: Determine whether the current tomographic image is the bottom layer tomographic image. If yes, proceed to step S3B307; otherwise, take the tomographic image located in the next layer and repeat steps S3B303 to S3B305. S3B307: Acquire the tomographic image located at the top layer; S3B308: Retrieves a gap object in the current tomographic image; S3B309: Check if there are any pixels with a gray value of 0 in the 8-neighborhood of the gap object in the next layer of tomographic image. If so, mark them as vertically connected objects. S3B310: Determine whether the void objects in the current fault image have been traversed. If not, take the next void object and repeat step S3B309. If the traversal is complete, proceed to step S3B311; S3B311: Determine whether the current fault image is the bottom fault image. If yes, proceed to step S3B312. If not, take the fault image located in the next layer and repeat steps S3B309 to S3B310. S3B312: Obtain the tomographic image located at the bottom layer; S3B313: Retrieves a gap object in the current tomographic image; S3B314: Check if there are any pixels with a gray value of 0 in the 8-neighborhood of the gap object in the previous layer tomographic image. If they exist, mark them as connected; otherwise, mark them as closed. S3B315: Determine whether the void objects in the current fault image have been traversed. If not, take the next void object and repeat step S3B313; if the traversal is completed, execute step S3B316. S3B316: Determine whether the current fault image is the top fault image. If yes, proceed to step S3B317; otherwise, take the fault image located in the previous layer and repeat steps S3B313 to S3B312. S3B317: Check whether there is a connection mark in the gaps of each top fault image. If there is, mark the gap with the gap mark as a connected gap. If there is no connection mark, mark it as a closed gap. S3B318: Calculate the connectivity porosity based on the connectivity gaps.

3. The design method for ultra-dense asphalt mixtures according to claim 2, characterized in that, The relationship between the various aggregates, the target porosity, the aggregate void ratio of the coarse aggregate main skeleton, and the asphalt content is expressed as follows: ， in, This represents the mass percentage of coarse aggregate. Here, n represents the mass percentage of each particle size, and n is the number of coarse aggregate components. This represents the mass percentage of fine aggregate. This represents the mass percentage of mineral powder. This represents the percentage of asphalt used. For the target porosity, The main frame is tightly packed with high density. The apparent density of fine aggregate, The apparent density of the mineral powder is... The density of the asphalt binder. The aggregate void ratio is the main skeleton of coarse aggregate; For different sieve apertures, the throughput is... This refers to the sieve aperture diameter, which allows only aggregates with particle sizes smaller than this sieve aperture diameter to pass through the corresponding sieve aperture. sieve aperture diameter This represents the lower limit of the sieve aperture throughput. sieve aperture diameter This represents the upper limit of the sieve aperture throughput.

4. The design method for ultra-dense asphalt mixtures according to claim 1, characterized in that, Step S3A specifically includes: S3A1: Establish corresponding aggregate particle models based on the particle size specifications of each coarse aggregate; S3A2: Set the mass percentage range of fine aggregate, mineral powder and coarse aggregate of different particle sizes according to the composition relationship between each mineral material, and generate at least two sets of coarse aggregate ratios; S3A3: Using the aggregate particle models described above, discrete element models of the asphalt coarse aggregate main skeleton structure with each coarse aggregate ratio are established respectively. S3A4: The mesoscopic contact stiffness of each aggregate particle model in the model space is set using a linear contact model; S3A5: Calculate the total volume of all coarse aggregate particles under each coarse aggregate ratio; S3A6: Calculate the aggregate void ratio and aggregate compaction density of the asphalt coarse aggregate main skeleton structure based on the discrete element model of each coarse aggregate ratio; the formula for calculating the aggregate void ratio of the coarse aggregate main skeleton is: ， in, The aggregate void ratio is the main skeleton of coarse aggregate. The volume of the model space. This represents the total volume of the aggregate particle model for coarse aggregates of various particle sizes. ， in, The aggregate void ratio is the main skeleton of coarse aggregate. The main frame is tightly packed with high density. The apparent density of the composite material after mixing coarse aggregates of various specifications according to the coarse aggregate ratio; The formula for the synthetic apparent density is: ， in, The apparent density of the composite material is the result of mixing coarse aggregates of various specifications according to the coarse aggregate ratio. This represents the mass percentage of coarse aggregate. Let n be the mass percentage of each particle size, and n be the number of coarse aggregate components. The formula for the mass percentage of the coarse aggregate is: ， in, This represents the mass percentage of coarse aggregate. denoted as the mass percentage of each particle size specification, and n is the number of component specifications of the coarse aggregate. S2A7: Based on the compositional relationship between the aggregates and the mass percentage range of fine aggregates and mineral powder, at least two aggregate ratios are generated for each group of coarse aggregate ratios; the compositional relationship between the coarse aggregates, fine aggregates, mineral powder, asphalt, and aggregates is expressed as follows: ， ， in, This represents the mass percentage of coarse aggregate. This represents the mass percentage of fine aggregate. This represents the mass percentage of mineral powder. Here, n represents the mass percentage of each particle size, and n is the number of coarse aggregate components. sieve aperture diameter This represents the lower limit of the sieve aperture throughput. sieve aperture diameter This represents the upper limit of the sieve aperture throughput.

5. The design method for ultra-dense asphalt mixtures according to claim 4, characterized in that, The aggregate particle model is established as follows: S3A101: Select coarse aggregate particles according to the selected particle size specification; S3A102: Perform X-ray CT scans on each of the coarse aggregate particles with different particle sizes to obtain a set of tomographic images of each coarse aggregate particle; S3A103: Obtain the corresponding equivalent volume sphere diameter based on the tomographic image set of each coarse aggregate particle; S3A104: Generate a sphere according to the equivalent volume sphere diameter; S3A105: Generate multiple random points on the surface of the sphere and connect them to form a three-dimensional convex hull; S3A106: Voxelize the surface and interior of the three-dimensional convex hull; S3A107: Extract the skeleton points of each 3D convex hull; S3A108: Generate the maximum inscribed sphere covering each skeleton point of each three-dimensional convex hull to obtain the aggregate particle model of each coarse aggregate particle; S3A109: Assign the bulk density to each aggregate particle model corresponding to its particle size specification.

6. The design method for ultra-dense asphalt mixtures according to claim 5, characterized in that, The specific operation of surface voxelization of the three-dimensional convex hull is as follows: establish a voxel space composed of orthogonal meshes, then determine the spatial relationship between each voxel and the closed surface of the three-dimensional convex hull, and retain the voxels that intersect with the closed surface, thus realizing the surface voxelization of the three-dimensional convex hull. The specific operation of voxelizing the interior of the three-dimensional convex hull is as follows: filling the interior voxels by scanning or ray casting to achieve voxelization of the interior of the three-dimensional convex hull. The orthogonal mesh is a cube, and its function expression is: ， Among them, voxels Centered on The cube, where h is the resolution and the side length of the cube is... The spatial coordinates of the voxel are . The functional expression of the three-dimensional convex hull after surface voxelization is as follows: ， The functional expression of the three-dimensional convex hull after internal voxelization is: S3A108: Extract the skeleton points of each 3D convex hull.

7. The design method for ultra-dense asphalt mixtures according to claim 6, characterized in that, In step S3A109, when generating each sphere with the largest inscribed radius, the skeleton points of the three-dimensional convex hull are arranged in descending order of radius; then, the spheres with the largest radius are placed sequentially to cover the space not occupied by the previous spheres; when the sphere filling volume loss rate reaches a certain threshold, the space filling is stopped. The filling volume loss rate is the ratio of the volume of the sphere filling to the volume of the three-dimensional convex hull, and its calculation formula is as follows: ， in, For the filling volume loss rate, Let the volume of each sphere be , The volume is the 3D convex hull volume. In this embodiment, the filling volume loss rate is 5%. In this embodiment, the clump (cluster) block generation technology in PFC3D is used to cover the maximum inscribed sphere at each skeleton point; The threshold for the cumulative volume is set according to a threshold function, which is: ， in, The volume of the model space. For cumulative volume, It represents the volume of the smallest aggregate particle model within a given particle size specification.

8. The design method for ultra-dense asphalt mixtures according to claim 7, characterized in that, Set reduction factor The size of the aggregate particle model is linearly reduced to Subsequently, the surface normal stiffness of the aggregate particle model was temporarily set to... To ensure that particles bounce off each other upon contact, surface friction and damping are temporarily set to zero, allowing particles to slide freely and rebound almost elastically upon contact, thus randomizing the initial arrangement and orientation of each aggregate particle model. Subsequently, the Cycle command is used to progressively expand the aggregate particle model to its original volume, causing random rotation and bounce-off of overlapping areas within the model space. This progressive expansion is based on a step-by-step magnification factor, which is: ， This indicates that expansion to the original volume size is achieved through N-level amplification.

9. The design method for ultra-dense asphalt mixtures according to claim 1, characterized in that, Step S4 calculates the asphalt content for each aggregate proportion based on a multiple linear regression model relating the mass percentage of each aggregate, the interstitial ratio of the coarse aggregate main skeleton, and the passing rate of each sieve size. The expression for the multiple linear regression model is as follows: ， in, This represents the percentage of asphalt used. The aggregate void ratio is the main skeleton of coarse aggregate. This refers to the mass percentage of coarse aggregate with a particle size of 5-10 mm. This represents the mass percentage of fine aggregate. The throughput of a sieve with a aperture size of 9.5 mm. The throughput of a sieve with a aperture size of 4.75 mm. The throughput of a sieve with a aperture size of 2.36 mm. The throughput of a sieve with a aperture size of 1.18 mm. The throughput of a sieve with a aperture size of 0.6 mm. This represents the throughput of a sieve with a aperture size of 0.3 mm.

10. An electronic device, characterized in that, processor; Memory for storing computer programs executed by the processor; When the processor executes the computer program, it implements the ultra-dense asphalt mixture design method according to any one of claims 1-9.

Images