Discrete element-based two-dimensional asphalt mixture model construction method

Abstract

The invention belongs to the technical field of road materials, and relates to a discrete element-based two-dimensional asphalt mixture model construction method, which comprises the following steps of S1, manufacturing an asphalt mixture test piece; s2, establishing an asphalt mixture form model; s3, respectively endowing contact constitutive models to the asphalt mixture form models by adopting a discrete element technology; s4, performing preliminary assignment on the contact parameters corresponding to the endowed contact constitutive model; combining the actual stress-strain curve and the actual creep curve, and respectively adopting an empirical method, fitting and reverse iteration to obtain mesoscopic parameters of the contact constitutive model; and S5, completing the construction of the two-dimensional asphalt mixture model. According to the method, the dimensional conversion of the two-dimensional aggregate is carried out by utilizing stereology and probabilitics, then the mesoscopic parameters are endowed and determined through the contact constitutive model, and the mesoscopic parameters are compared and corrected, so that the morphological and rheological characteristics of the asphalt mixture can be reflected, and the constructed two-dimensional asphalt mixture model is good in accuracy and high in reliability.

Description

Technical Field

[0001] The invention belongs to the technical field of road materials and relates to a method for constructing a two-dimensional asphalt mixture model based on discrete elements. Background Art

[0002] Asphalt pavements are widely used in my country due to their short construction period, comfortable driving, and easy maintenance. However, they also face pavement defects such as rutting, cracking, and oil spillage. Currently, phenomenology and macroscopic testing are used to study the mechanical properties of asphalt pavements and, in turn, the occurrence of asphalt pavement defects. However, because the mechanical properties of asphalt pavements are not solely reflected at the macroscopic scale, phenomenology and macroscopic testing have made it difficult to analyze the complex rheological behavior of asphalt pavement defects. Therefore, researchers have gradually shifted their research focus from the macroscopic scale of asphalt mixture materials and structures to the smaller microscopic scale. Numerical simulation methods such as discrete element methods have provided the conditions for exploring the spatiotemporal evolution of asphalt mixtures from a microscopic perspective.

[0003] Currently, discrete element modeling of asphalt mixtures is divided into two-dimensional and three-dimensional discrete element modeling based on dimensionality. Although three-dimensional discrete element modeling offers slightly higher data accuracy than two-dimensional discrete element modeling, its computational complexity makes it far less efficient than two-dimensional discrete element modeling. However, existing two-dimensional discrete element asphalt mixture modeling methods are limited to morphological simulation and fail to consider the impact of dimensionality reduction on the mechanical properties of asphalt mixtures, resulting in models that fail to accurately and truly reflect changes in the mechanical properties of asphalt mixtures. Furthermore, existing two-dimensional discrete element modeling methods lack the internal contact constitutive model of asphalt mixtures and the computational transformation between macroscopic and microscopic intrinsic parameters. This results in a significant gap between the constructed models and actual conditions, leading to poor accuracy and reliability.

[0004] Therefore, how to develop an accurate and reliable two-dimensional asphalt mixture model construction method is a practical problem that needs to be solved urgently. Summary of the invention

[0005] In view of the technical problems that the existing two-dimensional asphalt mixture model construction has poor model accuracy and reliability, the present invention provides a two-dimensional asphalt mixture model construction method based on discrete elements.

[0006] The present invention uses stereology and probabilistics to perform dimensional transformation of two-dimensional aggregates to establish an asphalt mixture morphological model. Then, the model is used to assign and determine microscopic parameters, and the microscopic parameters are compared and corrected. This can reflect the morphology and rheological characteristics of the asphalt mixture. The constructed two-dimensional asphalt mixture model has good accuracy and high reliability.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is:

[0008] A method for constructing a two-dimensional asphalt mixture model based on the discrete element method includes the following steps:

[0009] S1. Fabricate asphalt mixture specimens

[0010] Fabricate asphalt mixture specimens with asphalt mortar and aggregates, and obtain the actual stress-strain curve and actual creep curve of the asphalt mixture specimens through indoor uniaxial compression tests and uniaxial creep tests respectively;

[0011] S2. Establish the morphological model of the asphalt mixture

[0012] Utilize stereology, probability theory, and discrete element technology to establish the morphological model of the asphalt mixture;

[0013] S3. Assign the contact constitutive model

[0014] Use the discrete element method to assign the contact constitutive model to the morphological model of the asphalt mixture respectively; the contact constitutive model includes a linear contact model, a parallel bond model, and a Burgers model;

[0015] S4. Obtain the mesoscopic parameters

[0016] Preliminarily assign values to all the contact parameters corresponding to the contact constitutive model assigned in step S3; and in combination with the actual stress-strain curve and actual creep curve obtained in step S1, use the empirical method, fitting, and inverse iteration respectively to obtain the mesoscopic parameters of the contact constitutive model;

[0017] S5. Complete the construction of the two-dimensional asphalt mixture model

[0018] Assign values to the asphalt mixture model in step S2 according to the mesoscopic parameters in step S4; conduct virtual uniaxial compression tests to obtain the virtual stress-strain curve and virtual creep curve respectively; compare them with the actual stress-strain curve and actual creep curve in step S1 respectively, and correct the mesoscopic parameters in step S4 according to the comparison error to complete the construction of the two-dimensional asphalt mixture model; the two-dimensional asphalt mixture model includes the overall morphological model of the asphalt mixture, the contact constitutive model, and the mesoscopic parameters.

[0019] Further define that in step S1, the asphalt mixture specimen is a cylindrical structure with a diameter of 100 mm and a height of 150 mm; the asphalt mortar is formed by fine aggregates, mineral powder, and asphalt.

[0020] Further define that in step S2, the process of establishing the morphological model of the asphalt mixture is:

[0021] S2.1. Establish the basic model of the coarse aggregate

[0022] Use stereology and probability to transform the three-dimensional gradation of asphalt mixture, obtain the two-dimensional gradation of asphalt mixture adapted to discrete element method and use a disc for placement, and establish a basic model of coarse aggregate;

[0023] S2.2. Establish the morphological clump model of coarse aggregate

[0024] Based on the discrete element technology, use the random algorithm to construct polygons with different shapes in the basic model of coarse aggregate to obtain the morphological clump model of coarse aggregate;

[0025] S2.3. Establish the morphological model of asphalt mixture

[0026] Use regularly arranged small particle units to simulate asphalt mortar and replace the morphological clump model of coarse aggregate to establish the morphological model of asphalt mixture.

[0027] Further defined, in the step S2.1, the formula for transforming the three-dimensional gradation of asphalt mixture is:

[0028]

[0029] Where:

[0030] D is the particle size of the aggregate; D0 is the grid diameter; D max is the maximum aggregate particle size; P c is the passing rate; P k is the percentage of the total volume of the aggregate in the total volume of the asphalt mixture specimen, taking 75%;

[0031] In the step S2.1, before placement, the aggregate with a particle size less than 2.36 mm needs to be removed; use the balldistribute command to randomly place the corresponding gradation disc, and then make the coarse aggregates bounce off each other through servo balance to form a basic model of coarse aggregate.

[0032] Further defined, in the step S2.2, the construction of polygons with different shapes in the basic model of coarse aggregate using the random algorithm is carried out according to the following formula:

[0033] r ik =(1 - aλ)R i

[0034]

[0035] Where:

[0036] r ik represents the distance from the k-th vertex of the i-th coarse aggregate to the center of the coarse aggregate; a represents the reduction coefficient of the coarse aggregate radius, taking 0.24; λ represents a random number from 0 to 1; R iDenote the radius of the $i$-th coarse aggregate before transformation;

[0037] x ik Denote the abscissa of the $k$-th vertex of the $i$-th coarse aggregate; y ik Denote the ordinate of the $k$-th vertex of the $i$-th coarse aggregate; X i Denote the abscissa of the $i$-th coarse aggregate; Y i Denote the ordinate of the $i$-th coarse aggregate; N represents the number of sides of the irregular polygon, randomly selected from 4 to 10; k represents the serial number of the vertex of the $i$-th coarse aggregate.

[0038] Further defined, the specific process of using regularly arranged small particle units to simulate asphalt mortar and replace the clump model of the morphology of coarse aggregates to establish the morphology model of asphalt mixture in step S2.3 is as follows:

[0039] S2.3.1. Generate a uniformly distributed asphalt mortar model with small particle units in the simulated asphalt mortar area. The arrangement method is a square regular arrangement, and the radius of the small particle units is 0.4 mm;

[0040] S2.3.2. Characterize the voids in the two-dimensional asphalt mixture by randomly deleting 4% of the asphalt mortar;

[0041] S2.3.3. Use small particle units of the same size as the asphalt mortar model to replace at the positions where coarse aggregates are generated to obtain the morphology model of asphalt mixture.

[0042] Further defined, in step S3, the specific process of assigning the contact constitutive model is as follows:

[0043] Adopt the discrete element technology to assign a linear contact model to the contact between coarse aggregates; assign a parallel bond model to the contact between coarse aggregates and asphalt mortar; assign a Burgers model and a parallel bond model to the contact between asphalt mortars according to a ratio of 1:3.

[0044] Further defined, in step S4, the specific process of obtaining the mesoscopic parameters is as follows:

[0045] S4.1. Initially assign values to the contact parameters corresponding to each contact constitutive model in the asphalt mixture;

[0046] S4.2. Set the ratio of the normal stiffness to the tangential stiffness in the linear contact model to 1:1, and use the empirical method to adjust the initial assignment to obtain the basic contact parameters corresponding to the linear contact model;

[0047] S4.3. For the Burgers model, substitute the preliminary assignment into the creep compliance formula and continuously fit the actual creep curve obtained in step S1 until the macroscopic parameters of the Burgers model are finally obtained. Then, convert the obtained macroscopic parameters of the Burgers model into the corresponding microscopic parameters according to the macroscopic-microscopic conversion formula group of the Burgers model;

[0048] S4.4. For the parallel bond model, set the ratio of the normal stiffness to the shear stiffness in the parallel bond model to 1:1. Based on the results of the actual stress-strain curve in step S1, perform reverse iterative adjustment on the preliminary assignment to obtain the key parameters of the parallel bond model;

[0049] S4.5. The basic contact parameters, microscopic parameters, and key parameters obtained in steps S4.2 to S4.4 are the mesoscopic parameters.

[0050] Further defined, in step S4.3, the creep compliance formula is as follows:

[0051]

[0052] where: J(t) is the creep compliance, E1, E2, η1, and η2 are the parameters of the macroscopic Burgers model; t is the particle thickness, taking 1 mm;

[0053] The macroscopic-microscopic conversion formula group of the Burgers model is as follows:

[0054]

[0055] where: C mn 、K mn 、C kn and K kn are the parameters of the normal mesoscopic scale Burgers model; C ms 、K ms 、C ks and K ks are the parameters of the tangential mesoscopic scale Burgers model; v is the Poisson's ratio of the asphalt mortar, taking 0.5.

[0056] Further defined, in step S4.5, the key parameters include the bond tensile strength, contact bond strength, normal stiffness, and shear stiffness; the basic contact parameters include the friction coefficient, aggregate density, normal stiffness, shear stiffness, damping coefficient, and friction angle.

[0057] Compared with the prior art, the beneficial effects of the present invention are:

[0058] 1. The present invention utilizes stereology and probability to perform dimensional transformation of two-dimensional aggregates to establish a morphological model of asphalt mixture. Then, mesoscopic parameters are assigned and determined through the model, and the mesoscopic parameters are compared and corrected, which can reflect the morphological and rheological characteristics of asphalt mixture. The constructed two-dimensional asphalt mixture model has good accuracy and high reliability.

[0059] 2. The present invention proposes a method for converting three-dimensional gradation into two-dimensional gradation using stereology and probability. The mechanical properties of asphalt mixture specimens under different dimensions will be different, and directly converting the mass gradation into two-dimensional area gradation in practice will bring errors. The method of correcting the gradation through formulas to reduce the dimension of the model is scientific and reliable.

[0060] 3. The present invention uses a random algorithm to generate the morphological model of coarse aggregates, which can achieve rapid modeling. Compared with modeling methods such as CT scanning, it saves time and effort, and also saves experimental funds, facilitating the study of the performance of asphalt mixtures with a large number of different sizes and gradation types.

[0061] 4. The present invention uses asphalt mortar to regard fine aggregates, mineral powder, and asphalt binder as a whole, and generates them by simulation with particles of the same particle size. The generated specimens have a high degree of similarity, and the calculation efficiency in the discrete element software is further improved.

[0062] 5. The present invention endows the contact between asphalt mortar and asphalt mortar with the Burgers model and parallel bond model with a certain proportion distribution, which not only reflects the complex viscoelastic properties of asphalt mortar but also emphasizes the bonding relationship between asphalt mortar, making the contact situation inside the asphalt mixture model more in line with the actual situation.

[0063] 6. The present invention corrects the mesoscopic parameters through the key parameters determined by the empirical method and inversion method, so that the mesoscopic parameters of the asphalt mixture model are more accurate, facilitating further research on the properties of asphalt mixture. Description of the Drawings

[0064] Figure 1 is the flow chart and partial enlarged view of the generation of the asphalt mixture model in the example of the present invention;

[0065] Figure 2 is the display diagram of the contact constitutive model of the asphalt mixture in the example of the present invention;

[0066] Figure 3 is the comparison diagram between the virtual test and the laboratory test in the example of the present invention. Detailed Embodiments

[0067] The technical solutions of the present invention will be further described below in conjunction with the drawings and embodiments, but the present invention is not limited to the following implementation cases.

[0068] Unless otherwise defined, technical or scientific terms used in the present invention shall have the ordinary meanings as understood by those of ordinary skill in the art to which the present invention pertains.

[0069] Technologies, methods, and devices known to those of ordinary skill in the relevant fields may not be discussed in detail, but where appropriate, the said technologies, methods, and devices shall be regarded as part of the specification.

[0070] It should also be understood that the specific embodiments described above are only for explaining the present invention, and the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and all should be covered within the protection scope of the present invention.

[0071] See Figure 1 , a method for constructing a two-dimensional asphalt mixture model based on discrete element in the present invention, includes the following steps:

[0072] S1. Making asphalt mixture specimens

[0073] Use asphalt mortar and aggregates to make asphalt mixture specimens, and obtain the actual stress-strain curve and actual creep curve of the asphalt mixture specimens through indoor uniaxial compression test and uniaxial creep test respectively.

[0074] Specifically, the aggregates are formed by coarse aggregates and fine aggregates with different particle size ranges according to a certain gradation. The asphalt mortar is formed by fine aggregates, mineral powder, and asphalt.

[0075] In the present invention, the specific steps for making asphalt mixture specimens are as follows:

[0076] S1.1. Screening asphalt and aggregate raw materials that meet the specification requirements and conducting mixture gradation design

[0077] The asphalt uses asphalt binder with indexes such as penetration, softening point, and ductility meeting the specification; select aggregates with higher strength, rich surface texture, no cracks, and smaller color difference, and conduct relevant performance tests on the selected aggregates according to the relevant regulations in "Test Regulations for Aggregates in Highway Engineering (JTG E42 - 2005)".

[0078] S1.2. Making asphalt mixture specimens

[0079] According to the gradation of the required asphalt mixture, calculate the masses of each size of coarse aggregate and fine aggregate needed. Put the obtained aggregates, asphalt, and the forming mold into a drying oven, set it to a constant temperature of 120 °C, and heat for 4 hours. After heating is completed, take out the dried aggregates, put them into an automatic mixer for stirring to make the aggregates evenly mixed; add matrix asphalt according to the specified asphalt-aggregate ratio (the mass ratio of matrix asphalt to aggregates). After stirring is completed, put in the weighed mineral powder for stirring. To ensure the fluidity of the matrix asphalt during the stirring process, the temperature of the mixer is set to 120 °C. After stirring is completed, put it into the prepared mold; since the rotational compaction effect is better, closer to the actual asphalt pavement compaction effect, and can simultaneously control the height, porosity, and arrangement of the mixture, a rotational compactor is used to prepare asphalt mixture specimens. Place the mold in the rotational instrument to form a cylindrical specimen with a height of 165 mm and a diameter of 150 mm, take it out and let it stand naturally to dry; after the cylindrical specimen is dried and formed, use a core drill to take samples; then cut and polish the cylindrical specimen to form an asphalt mixture specimen with a diameter of 100 mm and a height of 150 mm.

[0080] S1.3. Conduct uniaxial compression and uniaxial creep tests

[0081] Both the uniaxial compression test and the uniaxial creep test use UTM test instruments.

[0082] First, put the asphalt mixture specimen into an oven, set a specific temperature (the temperature is preferably selected within 40 - 60 °C) for heating. After 4 hours, take it out and put it into the UTM test instrument, control the loading plate to contact the asphalt mixture specimen, and then place it in an environmental chamber set at the corresponding temperature. The uniaxial compression test uses a constant strain rate loading test with a speed of 7.5 mm / min. When the axial stress drops to 15% of the stress peak value, terminate the loading and record the actual stress-strain curve; the loading stress of the uniaxial creep test is controlled at 0.07 MPa, and the loading time is 1800 s, record the actual creep curve.

[0083] Furthermore, obtain the specified parameters of the asphalt mixture specimen through the actual stress-strain curve and the actual creep curve. Specifically, the specified parameters include the macroscopic parameters of the Burgers model and two key parameters of the parallel bond model.

[0084] The macroscopic parameters of the Burgers model are obtained by fitting the actual creep curve, and the two key parameters of the parallel bond model are obtained by inverse iteration according to the actual stress-strain curve.

[0085] S2. Establish an asphalt mixture morphology model

[0086] Utilize stereology, probability theory, and discrete element technology to establish an asphalt mixture morphology model.

[0087] S2.1. Establish the basic model of coarse aggregates

[0088] Use stereology and probability to transform the three-dimensional gradation of asphalt mixture into two dimensions, obtain the two-dimensional gradation of the asphalt mixture specimen suitable for discrete element method, and use discs for placement to establish the basic model of coarse aggregates

[0089] The three-dimensional gradation is corrected to two-dimensional gradation through the following formula

[0090]

[0091] Where

[0092] D is the particle size of the aggregate; D0 is the grid diameter; D max is the maximum aggregate particle size; P c is the passing rate; P k is the percentage of the total volume of the aggregate in the total volume of the asphalt mixture specimen, generally taking 75%

[0093] Considering the calculation rate of the discrete element software, the asphalt mixture is defined as coarse aggregates (aggregates with a particle size greater than 2.36 mm) and asphalt mortar (aggregates with a particle size less than 2.36 mm and binder), so as to remove the particles below 2.36 mm; then use the ball distribute command to randomly place the corresponding gradation discs; if the aggregate models overlap, make the coarse aggregates bounce off each other through servo balance to form the basic model of coarse aggregates

[0094] S2.2. Establish the morphological clump model of coarse aggregates

[0095] Based on the discrete element technology, use the random algorithm to construct polygons with different shapes in the basic model of coarse aggregates to obtain the morphological clump model of coarse aggregates

[0096] Use the discrete element custom function, adopt the loop command to find the coordinates of each particle, and use the following formula for generating random polygons to construct and position the morphological clump model of coarse aggregates

[0097] r ik =(1 - aλ)R i

[0098]

[0099] Where

[0100] r ik represents the distance from the k-th vertex of the i-th coarse aggregate to the center of the coarse aggregate; a represents the reduction coefficient of the coarse aggregate radius, taking 0.24; λ represents a random number from 0 to 1; R iDenote the radius of the \(i\)-th coarse aggregate before transformation;

[0101] x ik Denote the abscissa of the \(k\)-th vertex of the \(i\)-th coarse aggregate; y ik Denote the ordinate of the \(k\)-th vertex of the \(i\)-th coarse aggregate; X i Denote the abscissa of the \(i\)-th coarse aggregate; Y i Denote the ordinate of the \(i\)-th coarse aggregate; N represents the number of sides of the irregular polygon, randomly selected from 4 to 10; k represents the serial number of the vertex of the \(i\)-th coarse aggregate.

[0102] S2.3. Establish the morphological model of asphalt mixture

[0103] Use regularly arranged small particle units to simulate asphalt mortar and replace the clump model of the morphology of coarse aggregates to establish the morphological model of asphalt mixture.

[0104] Use small particle units to generate a uniformly distributed asphalt mixture model in the simulated asphalt mortar area. The arrangement method adopts a square regular arrangement. While arranging, group the polygons into coarse aggregates, and use the group command to assign a unique aggregate group name. The uncovered small units are the asphalt mortar group; Characterize the voids in the asphalt mixture by randomly deleting 4% of the asphalt mortar. The smaller the uniformly arranged particles, the finer the model. However, considering the calculation efficiency, the radius of the small particle units finally adopts 0.4 mm. The generated virtual simulation test of the asphalt mixture has the same morphological characteristics and distribution law of coarse aggregates as the real asphalt mixture, and a certain contact of coarse aggregates is achieved in two dimensions. Use small particle units of the same size as the asphalt mortar model to replace at the generation position of coarse aggregates.

[0105] S3. Assign the contact constitutive model

[0106] Use the cmat command in the discrete element method. In the refined clump model of the morphology of coarse aggregates, assign a linear contact model to the contact between coarse aggregates; assign a parallel bond model to the contact between coarse aggregates and asphalt mortar; Considering the special properties of asphalt mortar, use a random algorithm in the asphalt mortar unit to assign the Burgers model and the parallel bond model to the contact between asphalt mortar and asphalt mortar in a ratio of 1:3.

[0107] In this embodiment, asphalt mortar is formed by fine aggregates, mineral powder and asphalt.

[0108] In this embodiment, the linear contact model, the parallel bond model and the Burgers model are the contact constitutive models.

[0109] S4. Obtain mesoscopic parameters

[0110] S4.1. Initially assign values to the contact parameters corresponding to each contact constitutive model in the asphalt mixture.

[0111] According to the default parameters of the contact constitutive model in the PFC software and the "Rock Mechanics Parameter Handbook", initially assign values to the contact parameters of each contact constitutive model in the asphalt mixture.

[0112] S4.2. Set the ratio of the normal stiffness to the tangential stiffness in the linear contact model to 1:1, and use the empirical method to adjust the initial assignment to obtain the basic contact parameters corresponding to the linear contact model.

[0113] Specifically, the basic contact parameters include the friction coefficient, aggregate density, normal stiffness, tangential stiffness, damping coefficient, and friction angle.

[0114] S4.3. For the Burgers model, substitute the initial assignment into the creep compliance formula and continuously fit the actual creep curve obtained in step S1 to finally obtain the macroscopic parameters of the Burgers model; then convert the obtained macroscopic parameters of the Burgers model into the corresponding microscopic parameters according to the macroscopic-microscopic conversion formula group of the Burgers model.

[0115] For the temperature-sensitive Burgers model, conduct a uniaxial creep test to obtain the creep curve at a specified temperature, fit the macroscopic parameters E1, E2, η1, η2 of the Burgers model according to the creep compliance formula, and then convert the obtained macroscopic parameters of the Burgers model into the corresponding eight microscopic parameters C mn 、K mn 、C kn 、K kn 、C ms 、K ms 、C ks and K ks .

[0116] The creep compliance formula is as follows:

[0117]

[0118] Where:

[0119] Where: J(t) is the creep compliance, E1, E2, η1, and η2 are the parameters of the macroscopic Burgers model; t is the particle thickness, taking 1 mm.

[0120] Referring to the specifications of the "Rock Mechanics Parameter Handbook", the elastic modulus of the coarse aggregate is taken as 55 GPa, the Poisson's ratio is 0.25, and the friction coefficient is taken as 0.5.

[0121] The macroscopic-microscopic conversion formula group of the Burgers model is as follows:

[0122]

[0123] Wherein:

[0124] C mn 、K mn 、C kn and K kn are the parameters of the normal mesoscopic scale Burgers model; C ms 、K ms 、C ks and K ks are the parameters of the tangential mesoscopic scale Burgers model; v is the Poisson's ratio of asphalt mortar, taking 0.5; E1, E2, η1 and η2 are the parameters of the macroscopic Burgers model; t is the particle thickness, taking 1 mm.

[0125] S4.4. For the parallel bond model, set the bond strength ratio of the parallel bond model to 1.1, and based on the actual stress-strain curve in step S1, perform reverse iterative adjustment on the preliminary assignment to obtain the key parameters of the parallel bond model.

[0126] The key parameters include the bond tensile strength, contact bond strength, normal stiffness and tangential stiffness.

[0127] S4.5. The basic contact parameters, microscopic parameters and key parameters obtained in steps S4.2 to S4.4 are the mesoscopic parameters.

[0128] S5. Complete the construction of the two-dimensional asphalt mixture model

[0129] Assign values to the asphalt mixture model in step S2 according to the mesoscopic parameters in step S4, and perform virtual uniaxial compression tests to obtain the virtual stress-strain curve and virtual creep curve respectively; compare them with the actual stress-strain curve and actual creep curve in step S1 respectively, and correct the mesoscopic parameters in step S4 according to the comparison error to complete the construction of the two-dimensional asphalt mixture model.

[0130] Specifically, compare the virtual stress-strain curve with the actual stress-strain curve in step S1, and compare the virtual creep curve with the actual creep curve in step S1; when the comparison error is less than 5%, the construction of the two-dimensional asphalt mixture model is completed. When the comparison error is greater than 5%, repeat step S4 to correct the mesoscopic parameters until the error is less than 5%.

[0131] In this step, the two-dimensional asphalt mixture model includes the clump model of the morphology of coarse aggregates, the contact constitutive model and the mesoscopic parameters.

[0132] In this step, in the discrete element software, the same load condition (7.5 mm / min) as that in the indoor uniaxial compression test is applied to the asphalt mixture model. Similarly, when the axial stress drops to 15% of the stress peak value, the loading is terminated, the virtual stress-strain curve is recorded, and it is compared and corrected with the results of the indoor uniaxial compression test in step S1 to ensure the accuracy of the constructed two-dimensional asphalt mixture model.

[0133] The following will specifically illustrate the method for constructing a two-dimensional asphalt mixture model based on the discrete element of the present invention with specific embodiments.

[0134] Step 1. Fabrication of asphalt mixture specimens

[0135] A cylindrical specimen, namely the asphalt mixture, is fabricated with aggregate and asphalt. Specifically, the aggregate is formed by coarse aggregate and fine aggregate with different particle size ranges according to a certain gradation.

[0136] Preferably, the asphalt is Donghai 70# base asphalt produced by Sinopec, and the aggregate is limestone alkaline aggregate. After testing the selected materials, the results are shown in Table 1 - Table 4.

[0137] Table 1 Performance parameters of 70# base asphalt

[0138] Performance Index Unit Specification Requirements Test Results Penetration (25°C) 0.1 mm 60～80 65 Ductility cm Not less than 100 >150 Softening Point ℃ Not less than 45 48

[0139] Table 2 Performance indicators of coarse aggregate

[0140]

[0141] Table 3 Performance indicators of fine aggregate

[0142] Performance Index Unit Specification Requirements Test Results Apparent Relative Density <![CDATA[g / cm 3 > ≥2.50 2.7129 Angularity s ≥30 43 Sand Equivalent ％ ≥60 67 Soundness (≥0.6 mm) ％ ≤12 7.9 Clay Content ％ ≤3 0.7

[0143] Table 4 Performance indicators of mineral powder

[0144]

[0145]

[0146] In this example, the AC-13 type dense-graded asphalt mixture is used as the research object, and the mix proportion design is carried out by taking the median value of the AC-13 gradation. The specific gradation is shown in Table 5.

[0147] Table 5 Gradation range of asphalt mixture in different dimensions

[0148]

[0149] Step 2. Establishment of the asphalt mixture morphology model

[0150] Use stereology and probability to transform the three-dimensional gradation of asphalt mixture, obtain the two-dimensional gradation of asphalt mixture adapted to discrete element method, and use discs for placement to establish the basic model of coarse aggregate. During placement, randomly shaped polygons are placed within a specimen with a width of 100 mm and a height of 150 mm to represent the coarse aggregate, as Figure 1 shown in Figure b.

[0151] Use regularly arranged small particle units to simulate asphalt mortar, and simultaneously replace the randomly shaped polygons representing the coarse aggregate, as Figure 1 shown in Figure c. In the figure, the blue small particle units (small balls) are asphalt mortar, and the collection of small balls of various other colors is the coarse aggregate of cluster type; establish the morphological model of asphalt mixture.

[0152] Step 3: Assign contact constitutive models

[0153] Assign a linear contact model to the contact between coarse aggregates, a parallel bond model to the contact between coarse aggregate and asphalt mortar, and a Burgers model and a parallel bond model to the contact between asphalt mortars, with a ratio of 1:3, as Figure 2 shown. The green part in the figure is the Burgers contact model, the red part is the linear contact model, and the blue part is the parallel bond model.

[0154] Step 4: Obtain mesoscopic parameters

[0155] In this embodiment, the basic contact parameters of the coarse asphalt are shown in Table 6. Among them, the ratio of the normal stiffness to the tangential stiffness in the linear contact model is 1:1, and the magnitude is the default value of 10 8 N / m.

[0156] Table 6 Basic contact parameters

[0157] Parameter Value Size Friction Coefficient 0.5 Aggregate Density 2700 kg / m Normal Stiffness <![CDATA[10 8 Pa]]> Tangential Stiffness <![CDATA[10 8 Pa]]> Damping Coefficient 0.2 Friction Angle 45° Normal Stiffness (Parallel Bond Group) <![CDATA[2×10 6 Pa]]> Tangential Stiffness (Parallel Bond Group) <![CDATA[2×10 6 Pa]]>

[0158] For the parameters of the Burgers contact model, in this embodiment, based on the uniaxial creep tests at three temperatures of 40 °C, 50 °C, and 60 °C, by fitting the creep compliance curves, the macroscopic Burgers model parameters of the asphalt mortar are obtained, as shown in Table 7.

[0159] Table 7 Macroscopic Burgers model parameters of asphalt mortar

[0160] Temperature (°C) <![CDATA[E1 / MPa]]> <![CDATA[η1 / MPa·s]]> <![CDATA[E2 / MPa]]> <![CDATA[η2 / MPa·s]]> 40 7.47 14457.42 3.22 350.90 50 4.81 12672.09 2.02 155.69 60 3.15 9894.90 1.56 122.22

[0161] By substituting the macroscopic Burgers model parameters into the macroscopic-microscopic conversion formula group of the Burgers model, the microscopic Burgers model parameters are obtained, as shown in Table 8.

[0162] Table 8 Microscopic Burgers model parameters

[0163] Temperature (°C) Kmn / MPa Kms / MPa Kkn / MPa Kks / MPa 40 7.47 2.988 3.22 1.288 50 4.81 1.924 2.02 0.808 60 3.15 1.26 1.56 0.624 Temperature (°C) Cmn / MPa·s Cms / MPa·s Ckn / MPa·s Cks / MPa·s 40 14457.42 5782.968 350.90 140.36 50 12672.09 5068.836 155.69 62.276 60 9894.90 3957.96 122.22 48.888

[0164] For the parallel bond model, set the ratio of the normal stiffness to the shear stiffness in the parallel bond model to 1:1. Based on the actual stress-strain curve in step S1, perform reverse iterative adjustment on the preliminary assignment to obtain the key parameters of the parallel bond model. As shown in Table 9. The normal stiffness and shear stiffness of the parallel bond model are both 2×10 6 Pa.

[0165] Table 9 Results of Key Parameters of the Parallel Bond Model

[0166]

[0167] Step Five: Complete the construction of the two-dimensional asphalt mixture model

[0168] Assign values to the asphalt mixture model established in step two according to the mesoscopic parameters; conduct virtual uniaxial compression tests to obtain the virtual stress-strain curve and virtual creep curve respectively; and compare them with the actual stress-strain curve in step S1 to correct the mesoscopic parameters, so that the obtained two-dimensional asphalt mixture model has high accuracy, as Figure 3 shown.

[0169] From Figure 3 it can be seen that for the two-dimensional asphalt mixture model based on the discrete element method constructed in this embodiment, the mechanical properties have an error less than 5% compared with the actual situation, and the error is small, and its accuracy is sufficient to analyze various properties of the asphalt mixture. It can be seen that the model established by the model construction method of the present invention has good accuracy and high reliability.

[0170] The above are several relatively preferred implementation manners of the preparation method of the present invention, but it cannot be used as a limitation to the technical solutions protected by the present invention. Any replacement solutions obtained by those of ordinary skill in the art based on the technical idea of the present invention without creative labor should fall within the protection scope of the present invention.

Claims

1. A method for constructing a two-dimensional asphalt mixture model based on the discrete element method, characterized in that, It includes the following steps: S1. Fabricate asphalt mixture specimens Fabricate asphalt mixture specimens with asphalt mortar and aggregates, and respectively obtain the actual stress-strain curve and actual creep curve of the asphalt mixture specimens through indoor uniaxial compression tests and uniaxial creep tests; S2. Establish a morphological model of asphalt mixture Utilize stereology, probability theory, and discrete element technology to establish a morphological model of asphalt mixture; S3. Assign contact constitutive models Use discrete element technology to respectively assign contact constitutive models to the morphological model of asphalt mixture; the contact constitutive models include linear contact model, parallel bond model, and Burgers model; S4. Obtain mesoscopic parameters Preliminarily assign values to all contact parameters corresponding to the contact constitutive models assigned in step S3, and in combination with the actual stress-strain curve and actual creep curve obtained in step S1, respectively use the empirical method, fitting, and inverse iteration to obtain the mesoscopic parameters of the contact constitutive models; S5. Complete the construction of the two-dimensional asphalt mixture model Assign values to the asphalt mixture model in step S2 according to the mesoscopic parameters in step S4, and conduct virtual uniaxial compression tests to respectively obtain virtual stress-strain curves and virtual creep curves; compare them with the actual stress-strain curve and actual creep curve in step S1 respectively, and correct the mesoscopic parameters in step S4 according to the comparison error to complete the construction of the two-dimensional asphalt mixture model; the two-dimensional asphalt mixture model includes the overall morphological model of asphalt mixture, contact constitutive model, and mesoscopic parameters.

2. The method for constructing a two-dimensional asphalt mixture model based on the discrete element method according to claim 1, wherein In step S1, the asphalt mixture specimen is a cylindrical structure with a diameter of 100 mm and a height of 150 mm; the asphalt mortar is formed by fine aggregates, mineral powder, and asphalt.

3. The method for constructing a two-dimensional asphalt mixture model based on the discrete element method according to claim 2, wherein In step S2, the process of establishing the morphological model of asphalt mixture is as follows: S2.

1. Establish the basic model of coarse aggregates Utilize stereology and probability theory to perform dimensional transformation on the three-dimensional gradation of asphalt mixture, obtain the two-dimensional gradation of asphalt mixture adapted to discrete element method and use disks for placement, and establish the basic model of coarse aggregates; S2.

2. Establish the morphological clump model of coarse aggregates Based on discrete element technology, use random algorithms to construct polygons with various shapes in the basic model of coarse aggregates to obtain the morphological clump model of coarse aggregates; S2.

3. Establish the morphological model of asphalt mixture Use regularly arranged small particle units to simulate asphalt mortar and replace the morphological clump model of coarse aggregates to establish the morphological model of asphalt mixture.

4. The method for constructing a two-dimensional asphalt mixture model based on the discrete element method according to claim 3, wherein In step S2.1, the formula for performing dimensional transformation on the three-dimensional gradation of asphalt mixture is: Where: D is the aggregate particle size; D0 is the grid diameter; D max is the maximum aggregate particle size; P c is the passing rate; P k is the percentage of the total volume of aggregates in the total volume of the asphalt mixture specimen, taking 75%; In step S2.1, before placement, aggregates with a particle size below 2.36 mm need to be removed; use the balldistribute command to randomly place the corresponding gradation disks, and then make the coarse aggregates bounce off each other through servo balance to form the basic model of coarse aggregates.

5. The method for constructing a two-dimensional asphalt mixture model based on the discrete element method according to claim 3, wherein, In step S2.2, the construction of polygons with various shapes in the basic model of coarse aggregates using random algorithms is carried out according to the following formula: r ik = (1 - aλ)R i Where: r ik represents the distance from the k-th vertex of the i-th coarse aggregate to the center of the coarse aggregate; a represents the reduction coefficient of the coarse aggregate radius, taking 0.24; λ represents a random number from 0 to 1; R i represents the radius of the i-th coarse aggregate before transformation; x ik represents the abscissa of the k-th vertex of the i-th coarse aggregate; y ik represents the ordinate of the k-th vertex of the i-th coarse aggregate; X i represents the abscissa of the i-th coarse aggregate; Y i represents the ordinate of the i-th coarse aggregate; N represents the number of sides of the irregular polygon, randomly selected from 4 to 10; k represents the serial number of the vertex of the i-th coarse aggregate.

6. The method for constructing a two-dimensional asphalt mixture model based on the discrete element method according to claim 3, wherein In the step S2.3, the specific process of establishing the morphological model of asphalt mixture by using the clump model that simulates asphalt mortar with regularly arranged small particle units and replaces the morphology of coarse aggregates is as follows: S2.3.

1. Generate a uniformly distributed asphalt mortar model with small particle units in the simulated asphalt mortar area. The arrangement method is square regular arrangement, and the radius of the small particle units is 0.4 mm. S2.3.

2. Characterize the voids in the two-dimensional asphalt mixture by randomly deleting 4% of the asphalt mortar. S2.3.

3. Use small particle units of the same size as the asphalt mortar model to replace at the positions where coarse aggregates are generated to obtain the morphological model of asphalt mixture.

7. The method for constructing a two-dimensional asphalt mixture model based on the discrete element method according to claim 3, wherein In the step S3, the specific process of assigning the contact constitutive model is as follows: Adopt the discrete element technology to assign a linear contact model to the contact between coarse aggregates; assign a parallel bond model to the contact between coarse aggregates and asphalt mortar; assign the Burgers model and the parallel bond model to the contact between asphalt mortars according to the ratio of 1:

3.

8. The method for constructing a two-dimensional asphalt mixture model based on the discrete element method according to claim 7, characterized in that In the step S4, the specific process of obtaining the mesoscopic parameters is as follows: S4.

1. Initially assign values to the contact parameters corresponding to each contact constitutive model in the asphalt mixture. S4.

2. Set the ratio of the normal stiffness to the tangential stiffness in the linear contact model to 1:1, and use the empirical method to adjust the initial assignment to obtain the basic contact parameters corresponding to the linear contact model. S4.

3. For the Burgers model, substitute the initial assignment into the creep compliance formula and continuously fit the actual creep curve obtained in step S1 until the macroscopic parameters of the Burgers model are finally obtained; then convert the obtained macroscopic parameters of the Burgers model into the corresponding microscopic parameters according to the macroscopic-microscopic conversion formula group of the Burgers model. S4.

4. For the parallel bond model, set the ratio of the normal stiffness to the tangential stiffness in the parallel bond model to 1:1, and based on the results of the actual stress-strain curve in step S1, perform reverse iterative adjustment on the initial assignment to obtain the key parameters of the parallel bond model. S4.

5. The basic contact parameters, microscopic parameters, and key parameters obtained in steps S4.2 to S4.4 are the mesoscopic parameters.

9. The method for constructing a two-dimensional asphalt mixture model based on the discrete element method according to claim 8, wherein In the step S4.3, the creep compliance formula is as follows: Where: J(t) is the creep compliance, E1, E2, η1, and η2 are the parameters of the macroscopic Burgers model; t is the particle thickness, taking 1 mm. The macroscopic-microscopic conversion formula group of the Burgers model is as follows: Among them: C mn , K mn , C kn and K kn are the parameters of the Burgers model at the normal mesoscopic scale; C ms , K ms , C ks and K ks are the parameters of the Burgers model at the tangential mesoscopic scale; v is the Poisson's ratio of the asphalt mortar, taking 0.

5.

10. The method for constructing a two-dimensional asphalt mixture model based on the discrete element method according to claim 8, wherein In the step S4.5, the key parameters include the bond tensile strength, contact bond strength, normal stiffness, and tangential stiffness; the basic contact parameters include the friction coefficient, aggregate density, normal stiffness, tangential stiffness, damping coefficient, and friction angle.

Images