Construction method, system and storage medium of three-dimensional microscopic model of asphalt mixture

Abstract

The present invention relates to the field of road engineering technology, and discloses a method, system, and storage medium for constructing a three-dimensional microscopic model of an asphalt mixture. The method generates a three-dimensional convex polyhedron aggregate geometric model by controlling the gradation of the mixture and preset random factors, and grids the placement space. This effectively solves two problems existing in the pick-and-place method, including that the success rate of picking points and the placement efficiency both decrease with the increase of the aggregate placement volume, thereby greatly improving the construction efficiency of the three-dimensional microscopic model. In addition, the preset random factors are determined according to the relationship between the random range and stability, so that the model has strong stability and randomness.

Description

Technical Field

[0001] The present invention relates to the technical field of road engineering, and in particular to a method, system and storage medium for constructing a three-dimensional microscopic model of an asphalt mixture. Background Art

[0002] Asphalt mixtures are an important building material and are often used in urban roads and highways of all grades. However, their properties, such as cracking and mechanical behavior, are affected by numerous factors, including composition, porosity, temperature, and moisture, making them difficult to accurately predict using theoretical models. With the rapid development of the discrete element method (DEM) and the finite element method (FEM), numerous researchers have developed a series of numerical models that can predict asphalt mixture properties, such as cracking and thermal conductivity. Existing numerical models for asphalt mixtures can be divided into two categories based on their scale: two-dimensional (2D) and three-dimensional (3D). While the former offers ease of construction and minimal computational effort, these two-dimensional stochastic models still exhibit significant differences from the actual asphalt mixture, resulting in significant errors in the prediction of some properties. The latter can more accurately describe the geometric characteristics of asphalt mixtures, but the construction efficiency of existing 3D models remains relatively low, and the resulting aggregate geometry models suffer from poor stability. Therefore, existing methods for constructing numerical models for asphalt mixtures suffer from inefficiency. Summary of the Invention

[0003] The present invention provides a method, system and storage medium for constructing a three-dimensional mesoscopic model of asphalt mixture, so as to solve the problem of low efficiency of the existing method for constructing the numerical model of asphalt mixture.

[0004] In order to achieve the above object, the present invention is implemented through the following technical solutions:

[0005] In a first aspect, the present invention provides a method for constructing a three-dimensional mesoscopic model of an asphalt mixture, comprising:

[0006] Generate a three-dimensional convex polyhedron aggregate geometric model according to the gradation of the mixture and a preset random factor, wherein the value of the preset random factor is determined according to the relationship between the random range and stability;

[0007] The placement space is gridded, and each aggregate in the three-dimensional convex polyhedron aggregate geometric model is placed into the placement space according to the conflict judgment method of convex polyhedron aggregates, until all aggregates in the three-dimensional convex polyhedron aggregate geometric model are placed to obtain the asphalt mixture geometric model.

[0008] Optionally, the preset random factors include a shape coefficient, a sphere radius, a number of vertices, and a vertex random distribution factor.

[0009] Optionally, the value of the preset random factor is determined as follows:

[0010] Determining a relationship between a stability index and a preset random factor, wherein the stability index includes the number of small edges and small faces of the aggregate model;

[0011] The range of the preset random factors is limited based on the reduction in the mean value of the stability index, the reduction in the range of the extreme fluctuation range, and the reduction in the random range to determine the final preset random factors.

[0012] Optionally, generating a three-dimensional convex polyhedron aggregate geometric model according to the mixture gradation and preset random factors includes:

[0013] Determine the geometric parameters of the placement space and the gradation of the mixture, and calculate the target volume of aggregate of each particle size grade based on the gradation and the volume of the placement space;

[0014] Substitute the preset random factors into the aggregate vertex coordinate generation formula to generate the vertex coordinates of the aggregate;

[0015] The aggregate is triangulated according to the vertex coordinates of the aggregate using a triangulation function to obtain the aggregate volume. The target volume of the aggregate is used as a constraint condition to control the total volume of the aggregate model of each particle size grade to obtain a three-dimensional convex polyhedron aggregate geometric model.

[0016] Optionally, gridding the delivery space includes:

[0017] The deployment space is discretized into a series of numbered spatial points, and the numbers of the occupied spatial points and the numbers of the unoccupied spatial points are recorded through the array setpoint and the array unsetpoint respectively.

[0018] Optionally, placing each aggregate in the three-dimensional convex polyhedron aggregate geometric model into the placement space according to the convex polyhedron aggregate conflict judgment method includes:

[0019] Each time aggregate is placed, a point is randomly selected from the numbered set array unsetpoint of unoccupied space points as the target point, the coordinates of the target point are obtained through the numbered coordinate conversion function, and the aggregate is moved to the corresponding position according to the coordinates of the target point. The number of the space point occupied by the aggregate and whether the aggregate placement is conflicting are calculated through the conflict judgment method for convex polyhedron aggregates. If there is a conflict between aggregates, points are repeatedly taken from the array unsetpoint until there is no conflict between the aggregates, and the aggregate placement is considered successful.

[0020] Optionally, the calculating of the numbers of the space points occupied by the aggregates and whether the aggregate placements conflict by using a conflict judgment method for convex polyhedron aggregates includes:

[0021] The surface triangulation information of convex aggregate is obtained through triangulation function;

[0022] Splitting the convex aggregate into dozens of tetrahedrons with the surface triangles as bases and the center points as vertices according to the surface triangle information, and analyzing each tetrahedron by using the vertex equation of the tetrahedron;

[0023] The analytical equations of all tetrahedrons are combined to form the analytical equation system of convex aggregates;

[0024] Substitute any spatial point into the analytical equation group for calculation. If the calculation result is non-negative, the point belongs to the convex aggregate, otherwise it does not belong to the convex aggregate;

[0025] Substitute the spatial point coordinates in the placement space into the analytical equation group for judgment to obtain a series of spatial point coordinates owned by convex aggregates. Convert the spatial point coordinates owned by the aggregates into numbers through the coordinate number conversion formula and store them using the variable point. If the number of conflicting points between the variable point and the set of placed points is greater than 0, it is determined that there is a conflict between the aggregates and the aggregates are re-placed.

[0026] Optionally, the method further includes:

[0027] Repeatedly apply small displacements of size [I, J, K] to the successfully placed aggregates until the aggregates collide again or reach the boundary. The total displacement of the aggregates is recorded through the array variable dir. Then, use the rotation matrix to repeatedly apply random rotations to the aggregates, and record the total rotation of the aggregates through rot. If the number of rotations reaches the preset upper limit and the conflict between the aggregates disappears, stop the rotation and start placing the next aggregate. If the number of rotations reaches the preset upper limit and the aggregates still conflict, refund all rotations rot and the last applied small displacement [I, J, K], and then place the next aggregate.

[0028] In a second aspect, the present application provides a system for constructing a three-dimensional microscopic model of asphalt mixture, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method described in the first aspect when executing the computer program.

[0029] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method steps described in the first aspect.

[0030] Beneficial effects:

[0031] The present invention provides a method for constructing a three-dimensional mesoscopic model of asphalt mixture. The method generates a three-dimensional convex polyhedron aggregate geometric model by controlling the gradation of the mixture and preset random factors, and grids the placement space. It effectively solves two types of problems existing in the pick-and-place method, including that the success rate of picking points and the placement efficiency both decrease with the increase of aggregate placement volume, thereby greatly improving the construction efficiency of the three-dimensional mesoscopic model. In addition, the preset random factors are determined according to the relationship between random range and stability, so that the model has strong stability and randomness. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is one of the flow charts of a method for constructing a three-dimensional mesoscopic model of asphalt mixture according to a preferred embodiment of the present invention;

[0033] Figure 2 This is a second flow chart of a method for constructing a three-dimensional mesoscopic model of asphalt mixture according to a preferred embodiment of the present invention;

[0034] Figure 3 The gradation curve used for the asphalt mixture specimen in the preferred embodiment of the present invention;

[0035] Figure 4 : The fitting curves between the random factors and the defect evaluation index of the preferred embodiment of the present invention; wherein (a) is the case when the random factor is the number of vertices, (b) is the case when the random factor is the radius of the sphere, (c) is the case when the random factor is the shape coefficient, and (d) is the case when the random factor is the vertex random distribution factor;

[0036] Figure 5 This is a schematic diagram of convex polyhedron aggregate splitting according to a preferred embodiment of the present invention;

[0037] Figure 6 This is a flow chart of an aggregate movement and densification algorithm according to a preferred embodiment of the present invention;

[0038] Figure 7 A schematic diagram of an asphalt mixture geometric model generated for a preferred embodiment of the present invention;

[0039] Figure 8 Schematic diagram of Boolean operation segmentation of asphalt mixture specimens according to a preferred embodiment of the present invention;

[0040] Figure 9 A schematic diagram of a steady-state plate method according to a preferred embodiment of the present invention;

[0041] Figure 10 Schematic diagram of a numerical model generated in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following is a clear and complete description of the technical solutions of the present invention. It should be understood that the embodiments described are only a portion of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are intended to fall within the scope of protection of the present invention.

[0043] Unless otherwise defined, the technical or scientific terms used in the present invention shall have the usual meanings understood by persons of ordinary skill in the field to which the present invention belongs. The words "first", "second" and similar terms used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "a" do not indicate a quantity limitation, but rather indicate the existence of at least one. Words such as "connected" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship also changes accordingly.

[0044] See Figure 1 The present application provides a method for constructing a three-dimensional microscopic model of asphalt mixture, comprising:

[0045] A three-dimensional convex polyhedron aggregate geometric model is generated according to the mixture gradation and the preset random factors, wherein the value of the preset random factors is determined according to the relationship between random range and stability;

[0046] The placement space is gridded, and each aggregate in the three-dimensional convex polyhedron aggregate geometric model is placed into the placement space according to the conflict judgment method of convex polyhedron aggregates, until all aggregates in the three-dimensional convex polyhedron aggregate geometric model are placed to obtain the asphalt mixture geometric model.

[0047] The above-mentioned method for constructing a three-dimensional mesoscopic model of asphalt mixture generates a three-dimensional convex polyhedron aggregate geometric model through the gradation of the mixture and preset random factors, and grids the placement space, effectively solving two types of problems existing in the pick-and-place method, including that the success rate of picking points and the placement efficiency both decrease with the increase of aggregate placement volume, thereby greatly improving the construction efficiency of the three-dimensional mesoscopic model. In addition, the preset random factors are determined according to the relationship between random range and stability, so that the model has strong stability and randomness.

[0048] Optionally, the preset random factors include a shape coefficient, a sphere radius, a number of vertices, and a vertex random distribution factor.

[0049] Optionally, the value of the preset random factor is determined as follows:

[0050] Determining a relationship between a stability index and a preset random factor, wherein the stability index includes the number of small edges and small faces of the aggregate model;

[0051] The range of the preset random factors is limited based on the reduction in the mean value of the stability index, the reduction in the range of the extreme fluctuation range, and the reduction in the random range to determine the final preset random factors.

[0052] In this embodiment, the number of small edges and small faces refers to the edges in the generated aggregate whose side length is less than 0.1 times the radius of the aggregate, and the triangular faces whose area is less than 0.01 times the square of the radius of the aggregate.

[0053] In one example, the detailed steps for determining the final preset random factor are as follows:

[0054] S1.1 Determine the geometric parameters L, W, and H of the rectangular drop area and the gradation of the mixture and determine the target volume of aggregate for each size grade using the aggregate volume formula:

[0055]

[0056] In the above formula, Vi represents the target volume of aggregate of the i-th grade particle size; p i 、p i+1 It represents the passing rate of the i-th sieve hole from large to small in the screening test; Vair is the porosity of the specimen; Vspecimen is the total volume of the specimen; ma is the mass fraction of asphalt; ρi is the density of the i-th grade aggregate; Vall is the percentage of aggregate volume to the sum of asphalt and aggregate volume; N is the grade number of aggregate particle size.

[0057] S1.2 uses the number of small edges and faces of the aggregate model, defect_n, as a stability index to limit the value ranges of the four random factors, specifically:

[0058] S1.2.1 Control the other random factors and change only the random factors to be studied. Under this condition, the target volume generated in each simulation is 50,000 mm. 3 The aggregate is calculated and the defect_n value is counted. The simulation is repeated 100 times for each value point and the mean value is taken as the representative value of the point.

[0059] S1.2.2 Analyze the relationship between each random factor and the stability index, and use the reduction in the mean value of the stability index, the reduction in the range of fluctuation of the stability index, and the reduction in the random range as the basis for limiting the random range of the random factors;

[0060] S1.3 Randomly select values ​​for random factors such as shape coefficient η, sphere radius r, number of vertices n, and vertex random distribution factor ζ according to the random range determined in S1.2.

[0061] In this way, by studying the relationship between random range and stability, the random range of each factor is restricted to ensure that the model has both strong stability and randomness.

[0062] Optionally, generating a three-dimensional convex polyhedron aggregate geometric model according to the mixture gradation and preset random factors includes:

[0063] Determine the geometric parameters of the placement space and the gradation of the mixture, and calculate the target volume of aggregate of each particle size grade based on the gradation and the volume of the placement space;

[0064] Substitute the preset random factors into the aggregate vertex coordinate generation formula to generate the vertex coordinates of the aggregate;

[0065] The aggregate is triangulated according to the vertex coordinates of the aggregate using a triangulation function to obtain the aggregate volume. The target volume of the aggregate is used as a constraint condition to control the total volume of the aggregate model of each particle size grade to obtain a three-dimensional convex polyhedron aggregate geometric model.

[0066] In this optional embodiment, the Monte Carlo method is used to select random factors within a random range. With the coordinate system of the placement space as the reference, the centroid of the aggregate is set at the origin of the placement space coordinate system and substituted into the aggregate vertex coordinate generation formula to generate the aggregate vertex coordinates. Specifically, the aggregate vertex coordinates are generated using the aggregate vertex coordinate generation formula as follows:

[0067] Among them, the formula for generating aggregate vertex coordinates is:

[0068]

[0069]

[0070]

[0071]

[0072] In the above formula, the round() function is used to round the variables in the brackets, n i is the total number of vertices in the ith aggregate, and its value is an integer between 16 and 20; ζ is the random distribution factor, and its value is a random number between -1 and +1; w is the fluctuation factor, the larger its value, the greater the difference in the distribution of vertices in the upper and lower hemispheres. In this example, w is 2. i with n2 i are the number of vertices in the upper and lower hemispheres of the ith aggregate. minwith d max are the lower and upper limits of aggregate particle size, respectively; ri is the radius of the ith spherical aggregate, which is 0.5d min to 0.5d max A random number between is the azimuth coordinate value of the vertex before stratification, φ1 ij ,θ1 ij 、φ2 ij θ2 ij The azimuth and zenith angle coordinates of the jth point on the i-th aggregate sphere 1 and sphere 2 are respectively ij are the coordinates of the jth vertex of the ith aggregate, and π is a constant.

[0073] S1.4 Use the target volume fractional particle size calculated according to the method in S1.1 as the control condition to generate aggregates and record the vertex information and quantity of the aggregates, specifically:

[0074] S1.4.1 Set the total volume of generated aggregate VV = 0, the number of particle size classes i = 1, and the number of aggregates of particle size i number(i) = 0;

[0075] S1.4.2 Calculate the volume V of the aggregate based on the aggregate vertex coordinates generated in S1.3 and the MATLAB triangulation function DelaunayTriangulation. Set VV = VV + V, number(i) = number(i) + 1, and use the variable facet to record the triangulation and coordinate information of the aggregate.

[0076] S1.4.3 determines whether VV exceeds the target volume. If not, return to S1.4.2; otherwise, continue;

[0077] S1.4.4 determines whether the particle size grade number is greater than the set value. If so, continue; otherwise, set i=i+1, VV=0 and return to S1.4.2.

[0078] Optionally, gridding the delivery space includes:

[0079] The deployment space is discretized into a series of numbered spatial points, and the numbers of the occupied spatial points and the numbers of the unoccupied spatial points are recorded through the array setpoint and the array unsetpoint respectively.

[0080] In this optional implementation, the detailed steps of meshing are as follows:

[0081] S2.1.1 Determine the spacing between discrete points in the X, Y, and Z directions and represent them as components 1, 2, and 3 of the three-dimensional array coordi;

[0082] S2.1.2 Calculate the number of discrete points in the X, Y, and Z directions based on the discrete point spacing determined in S2.1.1. The formula for calculating the number of discrete points is:

[0083]

[0084] In the above formula, NumX, NumY, and NumZ represent the number of discrete points in the X, Y, and Z directions respectively, L, W, and H represent the length, width, and height of the cube placement space respectively, and coordi(1), coordi(2), and coordi(3) represent the first, second, and third components of coordi respectively.

[0085] S2.1.3 Number each point in space according to its coordinates, first in the X direction, then in the Y direction, and finally in the Z direction. The coordinate number conversion formula is:

[0086] num=x÷coordi(1)+y÷coordi(2)×NumX+z÷coordi(3)×NumX× (7)

[0087] NumY+1

[0088] In the above formula, x, y, z represent the coordinates of the spatial point, and num represents the number of the spatial point.

[0089] S2.1.4 Create a one-dimensional array setpoint of size NumX × NumY × NumZ to store all placed points. The position corresponds to the point number. A value of 0 indicates that the point is unoccupied, and a value of 1 indicates that the point is occupied. All values ​​in the array are set to 0. Create a one-dimensional array unsetpoint of size NumX × NumY × NumZ to store the numbers of unplaced points. Aggregate number j = 0, and particle size grade number i = 0.

[0090] S2.2 randomly selects a point in the set unsetpoint as the placement point and moves the generated j-th aggregate to the placement point.

[0091] Optionally, placing each aggregate in the three-dimensional convex polyhedron aggregate geometric model into the placement space according to the convex polyhedron aggregate conflict judgment method includes:

[0092] Each time the aggregate is placed, a point is randomly selected from the numbered set array unsetpoint of unoccupied space points as the target point. The coordinates of the target point are obtained through the numbered coordinate conversion function. The aggregate is moved to the corresponding position at the target point according to the coordinates of the target point. The number of the space point occupied by the aggregate and whether the aggregate placement is conflicting are calculated through the conflict judgment method for convex polyhedron aggregates. If there is a conflict between the aggregates, points are repeatedly taken from the array unsetpoint until there is no conflict between the aggregates, and the aggregate placement is considered successful.

[0093] Among them, the corresponding position at the target point refers to the coordinate position of this target point.

[0094] Furthermore, the delivery step further includes:

[0095] S2.3 uses the aggregate conflict judgment method for complex convex polyhedrons to determine whether the newly placed aggregate conflicts with the already placed aggregate. If there is a conflict between the aggregates, return to S2.2 to take points until there is no conflict between the aggregates, that is, the aggregate placement is successful.

[0096] It should be noted that in actual placement, some points in the placement space may not be occupied. However, since the distance between points is close, if the aggregate is placed at this point, the vertex of the placed aggregate may be in the point space adjacent to this point, which will cause conflict. Based on this, by judging whether the aggregate placement is conflicting, this situation can be avoided.

[0097] Optionally, the calculating of the numbers of the space points occupied by the aggregates and whether the aggregate placements conflict by using a conflict judgment method for convex polyhedron aggregates includes:

[0098] The surface triangulation information of convex aggregate is obtained through triangulation function;

[0099] Splitting the convex aggregate into dozens of tetrahedrons with the surface triangles as bases and the center points as vertices according to the surface triangle information, and analyzing each tetrahedron by using the vertex equation of the tetrahedron;

[0100] The analytical equations of all tetrahedrons are combined to form the analytical equation system of convex aggregates;

[0101] Substitute any spatial point into the analytical equation group for calculation. If the calculation result is non-negative, the point belongs to the convex aggregate, otherwise it does not belong to the convex aggregate;

[0102] Substitute the spatial point coordinates in the placement space into the analytical equation group for judgment, and obtain a series of spatial point coordinates owned by convex aggregates. Convert the spatial point coordinates owned by the aggregates into numbers through the coordinate number conversion formula and store them using the variable point. If the value of the expression sum(setpoint(point)) is greater than 0, determine whether there is a conflict between the aggregates. Specifically, the expression sum(setpoint(point)) specifically refers to the size of the intersection of the variable point and the set of placed points setpoint, that is, the number of conflict points between the two. That is, if the number of conflict points between the variable point and the set of placed points is greater than 0, it is determined that there is a conflict between the aggregates and the aggregates are re-placed.

[0103] In this optional embodiment, it also includes:

[0104] S2.3.1 Divide the convex polyhedron aggregate into several central tetrahedrons in the form of center point-surface triangles. Substitute the coordinates of the suspicious spatial point into the central tetrahedron analytical formula for calculation. When the result of any equation is non-negative, the point belongs to the aggregate. The central tetrahedron analytical formula is:

[0105] |||f1|-2f2|+|f1|+f3-3v|+||f1|-2f2|+|f1|-3v (9)

[0106]

[0107]

[0108]

[0109]

[0110] In the above formula, f1, f2, and f3 represent three intermediate variables, v represents the volume of the tetrahedron, x1, y1, z1, x2, y2, z2, x3, y3, and z3 represent the coordinates of the three vertices of the triangle on the surface of the convex polyhedron; x4, y4, and z4 represent the center point of the convex polyhedron, and their values ​​are all 0;

[0111] S2.3.2 Calculate whether the value of the expression sum(setpoint(point)) is greater than 0. If it is greater than 0, there is a conflict between the aggregates and return to S2.2. Otherwise, there is no conflict and continue.

[0112] In this way, the convex polyhedron aggregate is split to obtain its analytical expression, which is convenient for further judgment of the conflict relationship of the aggregates.

[0113] Optionally, the method further includes:

[0114] Repeatedly apply small displacements of size [I, J, K] to the successfully placed aggregates until the aggregates collide again or reach the boundary. The total displacement of the aggregates is recorded through the array variable dir. Then, use the rotation matrix to repeatedly apply random rotations to the aggregates, and record the total rotation of the aggregates through rot. If the number of rotations reaches the preset upper limit and the conflict between the aggregates disappears, stop the rotation and start placing the next aggregate. If the number of rotations reaches the preset upper limit and the aggregates still conflict, refund all rotations rot and the last applied small displacement [I, J, K], and then place the next aggregate.

[0115] In this optional embodiment, it also includes:

[0116] S2.4.1 Continue applying small random displacements [I, J, K] to the successfully placed aggregate until the value of the equation sum(setpoint(point)) is greater than 1, and record the total displacement of the aggregate using the variable dir, where I, J, and K are random integers between -1 and 1;

[0117] S2.4.2 Set the success index and the number of attempts k, where index = 1 and k can be set arbitrarily;

[0118] S2.4.3 Apply random rotations to the aggregate using a rotation matrix, and use rot to record the total rotation of the aggregate, that is, rot = rot * M. The expression of the rotation matrix is:

[0119]

[0120] In the above formula, M is the rotation matrix, α, β, and γ represent the rotation angles in the X, Y, and Z directions respectively.

[0121] S2.4.4 uses the method described in S2.3 to determine whether there is a conflict in the aggregates. If there is a conflict, decrement k by 1, i.e., k = k-1. Otherwise, return to S2.4.4 and set index = 0.

[0122] If S2.4.5 determines that both k=0 and index=0 are false, return to S2.4.3; otherwise, proceed to the next step.

[0123] S2.4.6 Determine whether index = 1 is true. If so, return the displacement dir and rotation rot. Otherwise, continue. The return expression is:

[0124] direction ij =direction ij / rot-dir (15)

[0125] S2.4.7 Obtain the numbers of the space points occupied by the aggregates using the method described in S2.3.1, change the value corresponding to the number in setpoint to 1, and delete the number corresponding to the number in unsetpoint.

[0126] S2.5 Let j = j + 1 and determine whether j < number(i) is true. If it is true, return to the step of re-placing the aggregates; otherwise, continue.

[0127] S2.6 Let i = i + 1 and determine whether i > 3 is true. If it is false, let j = 0 and return to the step of re-placing the aggregates; otherwise, output the triangular surface information stored in facet in stl format for subsequent further use of this information.

[0128] In this way, using the moving compaction algorithm to process the successfully placed aggregates can make the distribution of the aggregates more dense.

[0129] In a complete example, see Figure 2 , the construction method of the above three-dimensional mesoscopic model of asphalt mixture specifically includes the following steps:

[0130] S1. Randomly generate a three-dimensional convex aggregate geometric model according to the gradation

[0131] S1.1 Determine the geometric parameters L, W, H of the rectangular placement area, and determine the target volume of each particle size grade aggregate through the gradation;

[0132] Set the geometric parameters of the placement area of the asphalt mixture specimen as L = 100, W = 100, H = 60, and the gradation curve is as attached Figure 3 shown. Calculate the target volumes of the first three gradations as 47055.2 mm 3 、118185.3 mm 3 、118732.4 mm 3 .

[0133] S1.2 Use the number of small sides and small faces defect_n of the aggregate model as a stability index to evaluate and control the four random factors within a reasonable range of values.

[0134] Control several other random factors unchanged, only change the random factor to be studied, and under this condition, simulate and generate aggregates with a target volume of 50000 mm 3 each time, and count the defect_n value. Repeat the simulation 100 times for each value point and take its average value as the representative value of this point. Analyze the relationship between each random factor and defect_n, and the results are as attached Figure 4As shown in the figure, (a) shows the case when the random factor is the number of vertices, (b) shows the case when the random factor is the sphere radius, (c) shows the case when the random factor is the shape coefficient, and (d) shows the case when the random factor is the vertex random distribution factor. Based on the fitting relationship between each random factor and the defect index, the reduction in its mean, the reduction in the range of the extreme fluctuation range, and the reduction in the random range are used as the basis for limiting the random range of the random factors. Ultimately, the random range of the modulus length |η| of the shape coefficient is limited to between 1.458 and 1.73, and the random range of the vertex random distribution factor ζ is limited to between 0.4 and 0.9.

[0135] S1.3 Generate the vertex coordinates of the aggregate using the aggregate vertex coordinate generation formula within the range of each random factor:

[0136] The Monte Carlo method is used to randomly select values ​​for random factors, and the vertex coordinates of aggregates are generated using the aggregate vertex coordinate generation formula.

[0137] S1.4 Generate aggregates based on target volume:

[0138] Aggregates are randomly generated within each particle size grade, and the aggregates are triangulated using MATLAB's triangulation function DelaunayTriangulation. The variable facet is used to save the vertex information and triangulation information of each aggregate, and its volume is calculated. When the total volume of the generated aggregate exceeds the target volume, the generation of aggregates of the current particle size grade is terminated, and the generation of aggregates of the next particle size grade begins. In this embodiment, the first three particle size grades of aggregates are generated, and their number number = [42, 203, 651].

[0139] S2 uses the background grid and the improved pick-and-place method of the random moving densification algorithm and the calculation method for complex convex polyhedron aggregate conflict to build an asphalt mixture geometric model

[0140] S2.1 Set the spacing of discrete points in the X, Y, and Z directions of the placement space to 0.5 mm, that is, coordi = [0.50.50.5], then the number of discrete points in the X, Y, and Z directions is calculated to be 201, 201, and 121 respectively. Number each point using the coordinate number conversion formula, and create two array variables setpoint and unsetpoint respectively. Setpoint is used to store all placed points, and its position represents the point with the corresponding number. A value of 0 means that the point is not occupied, and a value of 1 means that the point is occupied. The initial value of all positions is set to 0, and unsetpoint is used to store the number of unplaced points, so that the aggregate number j = 0 and the particle size grade number i = 0. The coordinate number conversion formula is:

[0141] num=x÷coordi(1)+y÷coordi(2)×NumX+z÷coordi(3)×NumX×NumY+1 (16)

[0142] S2.2 randomly selects a point in the set unsetpoint as the placement point, and moves the j-th aggregate generated in S1 to the placement point.

[0143] S2.3 uses the aggregate conflict judgment method for complex convex polyhedrons to determine whether the newly placed aggregate conflicts with the already placed aggregate. If so, return to S2.2. Otherwise, record the points owned by the aggregate through facet and continue.

[0144] S2.3.1 Assemble convex polyhedrons Figure 5 The center point-surface triangle structure shown is split into dozens of tetrahedrons. The resulting system of equations, derived from the simultaneous analytical expressions for all tetrahedrons, is used as the analytical equation for the convex polyhedron. Substituting any point into the equation yields a non-negative result, indicating that the point belongs to the aggregate. Using the analytical equation for the aggregate, the positions and numbers of all points belonging to the aggregate in the placement space can be calculated. The intermediate variable point is used to store the aggregate point number.

[0145] S2.3.2 Calculate whether the value of sum(setpoint(point)) is greater than 0. If it is greater than 0, it means that there is a conflict between the aggregates; otherwise, there is no conflict.

[0146] S2.4 uses the moving densification algorithm to process the aggregates that have been successfully placed to make the distribution of the aggregates more dense. The flow chart of the moving densification algorithm is as follows: Figure 6 shown.

[0147] S2.4.1 Apply a random small displacement [I, J, K] to the successfully placed aggregate and record the total displacement of the aggregate through the variable dir, that is, dir = dir + [I, J, K], where I, J, K are random numbers between -1 and 1.

[0148] S2.4.2 Use the method described in S2.3.2 to determine whether there is conflict among the aggregates. If there is no conflict, continue to apply displacement to the aggregates according to S2.4.1 until conflict occurs.

[0149] S2.4.3 Set the success indicator index and the number of attempts k, where index = 1 and k can be set arbitrarily.

[0150] S2.4.4 Apply random rotations to the aggregate using the rotation matrix M and record the total rotation of the aggregate using rot.

[0151] S2.4.5 Use the method described in S2.3 to determine whether there is a conflict in the aggregates. If there is a conflict, decrement k by 1; otherwise, return to S2.4.4 and set index = 0.

[0152] S2.4.6 If both k = 0 and index = 0 are false, return to S2.4.4; otherwise, proceed to the next step.

[0153] S2.4.7 Determine whether index = 1 is true. If it is true, refund the displacement and rotation; otherwise, continue.

[0154] S2.4.8 Use the method described in S2.3.1 to obtain the numbers of the points owned by the aggregates, set setpoint(point) = 1, and delete the corresponding numbers in unsetpoint.

[0155] S2.5 Let j = j + 1 and determine whether j < number(i) is true. If it is true, return to S2.2; otherwise, continue.

[0156] S2.6 Let i = i + 1 and determine whether i > 3 is true. If it is, continue; then let j = 0 and return to S2.2.

[0157] S2.7 The asphalt mixture aggregate model generated after the placement is as Figure 7 shown.

[0158] In another example, a prediction model for the thermal conductivity of asphalt mixture specimens is constructed based on the steady-state hot plate method to verify the accuracy of the model, which specifically includes the following steps:<00,00372>

[0159] First, use the MATLAB program to output the aggregate triangulation information and coordinate information stored in the facet file in stl format, and convert it to stp format through a model format conversion website. Secondly, import the stp file into the commercial finite element software Abaqus, and use the Boolean operation as Figure 8 shown to divide the model into cylindrical asphalt mixture specimens with a diameter of 100 mm and a height of 60 mm, and construct a geometric model according to the steady-state hot plate method. Figure 9 is a schematic diagram of the steady-state hot plate method. Then, perform attribute assignment, mesh division, creation of analysis steps, interaction, load conditions, and load boundary conditions on the geometric model of the steady-state hot plate method. Among them, the thermal conductivity of the asphalt mortar is set to be 1.898 (W / (m·K)), the specific heat capacity is set to be 859 (J / (kg·K)), the thermal conductivity of the coarse aggregate is set to be 2.850 (W / (m·K)), the specific heat capacity is set to be 850 (J / (kg·K)), the mesh is divided using DC3D10 thermal conduction units, a heat flux of 1 W / m2 is applied to the upper surface, a temperature boundary condition of 15 °C is set for the lower surface, and the interaction is set to be thermal contact. Finally, the formed model is as Figure 10 The numerical model shown in the figure is created and submitted for solution. After the solution is completed, the temperature difference between the upper and lower surfaces of the model is obtained through the Volume Weighted Average Temperature plug-in. In this example, the temperature difference between the upper and lower surfaces is 14.867°C. Substituting this into the governing equation of the one-dimensional steady-state heat transfer process, the thermal conductivity of the asphalt mixture specimen is calculated to be 2.018. The thermal conductivity of the same graded asphalt mixture measured through testing is 2.012, which meets the accuracy requirements. The governing equation is:

[0160]

[0161] Where λ is the thermal conductivity of the material, W / (m·K); q is the heat flow through the material, W / m 2 ; h is the height of the specimen, m; △T represents the temperature difference between the upper and lower surfaces of the specimen, K.

[0162] The present application also provides a system for constructing a three-dimensional mesoscopic model of an asphalt mixture, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor, when executing the computer program, implements the steps of the aforementioned method. This system for constructing a three-dimensional mesoscopic model of an asphalt mixture can implement each embodiment of the aforementioned method and achieve the same beneficial effects, which will not be described in detail here.

[0163] The present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method steps described above. This computer-readable storage medium can implement various embodiments of the method described above and achieve the same beneficial effects, which are not described in detail here.

[0164] The above describes in detail the preferred embodiments of the present invention. It should be understood that those skilled in the art can make numerous modifications and variations based on the concepts of the present invention without inventive effort. Therefore, any technical solutions that can be derived by those skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A method for constructing a three-dimensional microscopic model of asphalt mixture, characterized in that: include: Generate a three-dimensional convex polyhedron aggregate geometric model according to the gradation of the mixture and a preset random factor, wherein the value of the preset random factor is determined according to the relationship between the random range and stability; The placement space is gridded, and each aggregate in the three-dimensional convex polyhedron aggregate geometric model is placed into the placement space according to the conflict judgment method of convex polyhedron aggregates, until all aggregates in the three-dimensional convex polyhedron aggregate geometric model are placed to obtain the asphalt mixture geometric model; The method of generating a three-dimensional convex polyhedron aggregate geometric model according to the gradation of the mixture and preset random factors includes: Determine the geometric parameters of the placement space and the gradation of the mixture, and calculate the target volume of aggregate of each particle size grade based on the gradation and the volume of the placement space; Substitute the preset random factors into the aggregate vertex coordinate generation formula to generate the vertex coordinates of the aggregate; The aggregate is triangulated according to the vertex coordinates of the aggregate using a triangulation function to obtain the aggregate volume, and the target volume of the aggregate is used as a constraint condition to control the total volume of the aggregate model of each particle size grade to obtain a three-dimensional convex polyhedron aggregate geometric model; Among them, the formula for generating aggregate vertex coordinates is: In the above formula, the round() function is used to round the variables in the brackets, n i is the total number of vertices in the ith aggregate, ζ is the random distribution factor, w is the fluctuation factor, n1 i with n2 i are the number of vertices in the upper and lower hemispheres of the ith aggregate, d min with d max are the lower and upper limits of aggregate particle size, r i is the radius of the i-th spherical aggregate, is the azimuth coordinate value of the vertex before stratification, φ1 ij ,θ1 ij 、φ2 ij θ2 ij are the azimuth and zenith angle coordinates of the jth point on the i-th aggregate sphere 1 and sphere 2, respectively. ij are the coordinates of the jth vertex of the ith aggregate, and π is a constant.

2. The method for constructing a three-dimensional mesoscopic model of asphalt mixture according to claim 1, characterized in that: The preset random factors include shape coefficient, sphere radius, number of vertices and vertex random distribution factor.

3. The method for constructing a three-dimensional microscopic model of asphalt mixture according to claim 1, characterized in that: The value of the preset random factor is determined as follows: Determining a relationship between a stability index and a preset random factor, wherein the stability index includes the number of small edges and small faces of the aggregate model; The range of the preset random factors is restricted based on the reduction in the mean value of the stability index, the reduction in the range of the extreme fluctuation range, and the reduction in the random range to determine the final preset random factors.

4. The method for constructing a three-dimensional microscopic model of asphalt mixture according to claim 1, characterized in that: The method of placing each aggregate in the three-dimensional convex polyhedron aggregate geometric model into the placement space according to the conflict judgment method of the convex polyhedron aggregate includes: Each time the aggregate is placed, a point is randomly selected from the numbered set array unsetpoint of unoccupied space points as the target point. The coordinates of the target point are obtained through the numbered coordinate conversion function. The aggregate is moved to the corresponding position at the target point according to the coordinates of the target point. The number of the space point occupied by the aggregate and whether the aggregate placement is conflicting are calculated through the conflict judgment method for convex polyhedron aggregates. If there is a conflict between the aggregates, points are repeatedly taken from the array unsetpoint until there is no conflict between the aggregates, and the aggregate placement is considered successful.

5. The method for constructing a three-dimensional microscopic model of asphalt mixture according to claim 4, characterized in that: The method for determining the conflict of convex polyhedral aggregates to calculate the number of the space points occupied by the aggregates and whether the aggregate placement is conflicting includes: The surface triangulation information of convex aggregate is obtained through triangulation function; Splitting the convex aggregate into dozens of tetrahedrons with the surface triangles as bases and the center points as vertices according to the surface triangle information, and analyzing each tetrahedron by using the vertex equation of the tetrahedron; The analytical equations of all tetrahedrons are combined to form the analytical equation system of convex aggregates; Substitute any spatial point into the analytical equation group for calculation. If the calculation result is non-negative, the point belongs to the convex aggregate, otherwise it does not belong to the convex aggregate; Substitute the spatial point coordinates in the placement space into the analytical equation group for judgment to obtain a series of spatial point coordinates owned by convex aggregates. Convert the spatial point coordinates owned by the aggregates into numbers through the coordinate number conversion formula and store them using the variable point. If the number of conflicting points between the variable point and the set of placed points is greater than 0, it is determined that there is a conflict between the aggregates and the aggregates are re-placed.

6. The method for constructing a three-dimensional microscopic model of asphalt mixture according to claim 1, characterized in that: The method further comprises: Repeatedly apply small displacements of size [I, J, K] to the successfully placed aggregates until the aggregates collide again or reach the boundary. The total displacement of the aggregates is recorded through the array variable dir. Then, use the rotation matrix to repeatedly apply random rotations to the aggregates, and record the total rotation of the aggregates through rot. If the number of rotations reaches the preset upper limit and the conflict between the aggregates disappears, stop the rotation and start placing the next aggregate. If the number of rotations reaches the preset upper limit and the aggregates still conflict, refund all rotations rot and the last applied small displacement [I, J, K], and then place the next aggregate.

7. A system for constructing a three-dimensional microscopic model of asphalt mixture, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method steps according to any one of claims 1 to 6 are implemented.

Images