Asphalt mixture three-dimensional discrete element model construction method, storage medium and equipment

Abstract

The method, storage medium and equipment for constructing a three-dimensional discrete element model of asphalt mixture belong to the technical field of digital design of asphalt mixture. In order to solve the problem that the existing method for constructing a three-dimensional discrete element model of asphalt mixture cannot take into account both accuracy and efficiency, the present invention obtains slice images of asphalt mixture based on industrial CT, and uses digital image processing technology to identify, extract and segment aggregate components; then the aggregate particle model is reconstructed in three dimensions, the surface mesh model of a single aggregate is derived, and the template library of aggregate particle model is created by software; then a skeleton structure of a specified grading type is constructed through a random generation algorithm; ball particles are filled between aggregate particles to simulate asphalt mortar, and a specified number of balls are randomly deleted according to the void ratio to simulate voids; finally, the model is given corresponding microscopic contact parameters to simulate the mechanical behavior of asphalt mixture, and the peak load error rate is used as an indicator to evaluate the accuracy of the discrete element model to realize the construction of a three-dimensional discrete element model.

Description

Technical Field

[0001] The present invention belongs to the technical field of digital design of asphalt mixtures, and in particular relates to a method for constructing a three-dimensional discrete element model of an asphalt mixture, a storage medium and a device. Background Art

[0002] In terms of structural material composition, asphalt mixture is a multiphase composite material with internal voids composed of aggregate, asphalt, and filler, with obvious inhomogeneous anisotropy, particle properties, and nonlinear characteristics. The difference in mechanical properties between different components makes the internal contact force chain transmission distribution of asphalt mixtures uneven under external loads, and the stress deformation is complex and has large differences. Traditional macro-performance tests cannot consider the complexity of the micro-structure of asphalt mixtures, nor can they deeply understand the relationship between the micro-structure and mechanical behavior of the mixture, so they have certain limitations.

[0003] The discrete element method (DEM) is a numerical method for simulating the mechanical behavior of granular materials. It is based on the theory of discontinuous media and is particularly suitable for the microscopic mechanical properties and stress-deformation analysis (non-uniform, discontinuous, and large deformation problems) of dispersed or cemented materials. It can quantify the internal contact force chain transmission distribution when the material is loaded. The force exerted on each particle and the displacement generated can be expressed intuitively, so that the internal mechanical mechanism of the material under stress can be better understood from the microscopic structural level.

[0004] The current discrete element modeling of asphalt mixtures is mostly two-dimensional, which cannot truly reflect the stress state of asphalt mixtures; the three-dimensional discrete element model can reflect the internal stress state of the structure and provide a powerful tool for the study of internal force chains, but there is no effective means to construct the model, and there are still problems such as accuracy and efficiency. The current methods for constructing three-dimensional discrete element models of asphalt mixtures can be roughly divided into two categories. One is based on random generation algorithms. Most of these methods use multi-faceted generation algorithms to simulate aggregate particles and randomly place them to simulate the skeleton structure. The advantages are high modeling efficiency and accurate control of the gradation composition of the generated model. However, the aggregate particle model constructed by the algorithm cannot fully represent the surface morphology of the real aggregate particles, and there is a large gap with the actual aggregate, and the authenticity of the force chain cannot be guaranteed. The other is based on industrial CT and digital image technology to establish in-situ discrete element digital specimens of asphalt mixtures. This method can restore the real microstructure of asphalt mixtures to a certain extent, but due to the error transmission in the image acquisition, processing, and segmentation process, there is a certain error between the constructed digital specimen and the real structure. At the same time, because it guarantees in-situ construction, it greatly prolongs the model construction time and has no practical application value. At present, the treatment of the interface position between aggregate and mortar and between aggregate and aggregate is not clear, and the accuracy of the force chain result is not ideal. In view of the limitations of the current construction of discrete element models of asphalt mixtures, this paper proposes a method for constructing three-dimensional discrete element models of asphalt mixtures that takes into account both model authenticity and modeling efficiency. Summary of the invention

[0005] In order to solve the problem that the existing asphalt mixture three-dimensional discrete element model construction method cannot take into account both accuracy and efficiency, the present invention further provides an asphalt mixture three-dimensional discrete element modeling method that takes into account both model authenticity and modeling efficiency.

[0006] A method for constructing a three-dimensional discrete element model of asphalt mixture comprises the following steps:

[0007] 1. Perform X-Ray scanning on the standard Marshall specimen of asphalt mixture based on industrial CT to obtain the slice image of asphalt mixture;

[0008] 2. Import the slice image obtained in step 1 into the digital image processing software, and use the digital image processing software to extract and segment the aggregate components;

[0009] 3. Based on step 2, the aggregate particle model is reconstructed in three dimensions to derive the surface mesh model of a single aggregate;

[0010] Fourth, the surface mesh files of aggregate particles of various particle sizes are imported into the discrete element software, and the template of aggregate particle model is created by using the Clump algorithm to establish a template library of aggregate particle model;

[0011] 5. Calculate the volume of aggregates of each size according to the volume, gradation type, asphalt-stone ratio, void ratio, aggregate density, and asphalt density of the Marshall specimen, call the aggregate particle model in the template library constructed in step 4, and place aggregate particle groups of the specified gradation type in the specified cylindrical wall space through a random generation algorithm. Use a loop command to eliminate the overlap between particles, set gravity to allow the aggregate particle groups to accumulate naturally, and finally apply a speed to the upper surface wall to compact the aggregate particle groups to a specified height to form a skeleton structure;

[0012] 6. Use the language program to fill ball particles between aggregate particles to simulate asphalt mortar, and randomly delete a specified number of balls to simulate the void structure according to the set void ratio, so as to realize the construction of a three-dimensional digital specimen of asphalt mixture;

[0013] 7. Assign the corresponding contact model and micro-contact parameters to the constructed digital specimen of asphalt mixture to simulate the mechanical behavior of asphalt mixture;

[0014] 8. Conduct virtual indirect tensile tests and indoor macroscopic mechanical tests, and use the peak load error percentage p as an evaluation index for the accuracy of the discrete element model. If the p value exceeds the modeling requirements, adjust the discrete element microscopic model parameters in step 7 until the error percentage is less than the error requirement, thereby completing the construction of the three-dimensional discrete element model of asphalt mixture.

[0015] Preferably, the process of extracting and segmenting the aggregate components using digital image processing software in step 2 is as follows:

[0016] Firstly, the aggregate components are identified and extracted through non-local mean filtering, grayscale equalization, interactive threshold segmentation and background detection correction algorithm. Then, the segmentation of the adhesion aggregate particles is achieved through watershed segmentation, morphological opening operation, hole filling algorithm and three-dimensional contact segmentation algorithm.

[0017] A computer storage medium stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement the method for constructing a three-dimensional discrete element model of asphalt mixture.

[0018] A device for constructing a three-dimensional discrete element model of asphalt mixture, the device comprising a processor and a memory, the memory storing at least one instruction, the at least one instruction being loaded and executed by the processor to implement the method for constructing a three-dimensional discrete element model of asphalt mixture as described.

[0019] Compared with the existing modeling methods, the present invention has the following advantages:

[0020] The method of the present invention can efficiently and quickly obtain a large number of aggregate particle model templates to fully characterize the real surface morphology of aggregate particles. The asphalt mixture molding process is accompanied by the crushing and stacking of aggregates. There is a certain difference between the aggregates of the asphalt mixture after molding and the aggregates before molding. Compared with directly scanning the aggregate particles, the aggregates extracted based on the asphalt mixture Marshall specimen are in the real state after molding, which ensures the accuracy of model construction. An image processing and segmentation method is proposed to accurately distinguish between aggregates and mortar, ensuring the authenticity of the aggregate morphology. Establishing a template library of aggregate particle models for calling greatly shortens the model construction time and improves the simplicity of model construction. Compared with directly constructing the model, the method of setting gravity to make the aggregate particle group naturally stacked truly simulates the specimen molding process, and the authenticity is greatly improved. The method of the present invention improves the accuracy of the three-dimensional discrete element model of asphalt mixture. The constructed aggregate call library ensures the authenticity and convenience of model construction. The random generation algorithm can accurately control the gradation type of the constructed digital specimen, taking into account the authenticity, accuracy and efficiency of the modeling results. The modeling method of the three-dimensional discrete element model of asphalt mixture is optimized, which provides an effective support for the study of the mechanical properties of asphalt mixture and helps to promote further in-depth research on the digital design of asphalt mixture. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 Reconstruction effect of asphalt mixture aggregate structure.

[0022] Figure 2 This is a flowchart for image processing using Avizo.

[0023] Figure 3 A digital image of the asphalt mixture processing process.

[0024] Figure 4 Aggregate particle template library.

[0025] Figure 5 This is the rendering of the constructed three-dimensional discrete element digital specimen of asphalt mixture.

[0026] Figure 6 This is the cross-sectional view of the asphalt mixture three-dimensional discrete element digital specimen constructed.

[0027] Figure 7 Comparison chart of mechanical curves between virtual test and indoor test. DETAILED DESCRIPTION Specific implementation method one:

[0029] This embodiment is a method for constructing a three-dimensional discrete element model of asphalt mixture, comprising the following steps:

[0030] 1. Perform X-Ray scanning on the standard Marshall specimen of asphalt mixture based on industrial CT to obtain the slice image of the Marshall specimen of asphalt mixture.

[0031] 2. Import the slice image obtained in step 1 into the digital image processing software Avizo; first, identify and extract the aggregate components through non-local mean filtering, grayscale equalization, interactive threshold segmentation, and background detection correction algorithm, and then segment the adhesion aggregate particles through watershed segmentation, morphological opening operation, hole filling algorithm processing and three-dimensional contact segmentation algorithm.

[0032] 3. Based on step 2, the aggregate particle model is reconstructed in three dimensions through the volume rendering command, and the surface mesh generation command is used to construct and export the surface mesh file (STL file) of the single aggregate model.

[0033] Fourth, the surface mesh files of aggregate particles of various particle sizes are imported into the discrete element software, and the Clump algorithm is used to create the aggregate particle model template, and a template library of the aggregate particle model is established.

[0034] In this implementation, the key parameters ratio and distance for controlling the Clump algorithm are set to 0.3 and 150 respectively.

[0035] 5. Calculate the volume of aggregates of each size according to the volume, gradation type, oil-stone ratio, void ratio, aggregate density, and asphalt density of the Marshall specimen. Randomly call the aggregate particle model in the template library constructed in step 4. Use a random generation algorithm to place aggregate particle groups of the specified gradation type in the specified cylindrical wall space. Use the cycle command to eliminate the overlap between particles. Set gravity to allow the aggregate particle groups to accumulate naturally. Finally, apply a speed to the upper surface wall to compact the aggregate particle groups to the specified height to form a skeleton structure.

[0036] In this embodiment, the height of the cylindrical wall is 200 mm.

[0037] 6. Use the fish language to write a program to fill ball particles between aggregate particles to simulate the asphalt mortar components. According to the set void ratio, a specified number of ball particles are randomly deleted to simulate the void structure, thereby realizing the construction of a three-dimensional digital specimen of asphalt mixture.

[0038] In this embodiment, the ball particle radius of the simulated asphalt mortar is 0.8 mm.

[0039] 7. Assign corresponding contact models and micro-contact parameters to the constructed digital asphalt mixture specimens to simulate the mechanical behavior of the asphalt mixture.

[0040] In this embodiment, the microscopic contact parameters of the discrete element model include contact stiffness parameters between aggregate units, contact stiffness and bonding parameters between aggregate and mortar units, and contact stiffness and bonding parameters between mortar units.

[0041] 8. Conduct virtual indirect tensile tests and indoor macroscopic mechanical tests, and use the peak load error percentage p as an evaluation index for the accuracy of the discrete element model. If the p value exceeds the modeling requirements, adjust the discrete element microscopic model parameters in step 7 until the error percentage is less than the error requirement, thereby completing the construction of the three-dimensional discrete element model of asphalt mixture.

[0042] The indirect tensile test was carried out in accordance with the Test Procedure for Asphalt-Based Asphalt Mixtures for Highway Engineering (JTG E20-2011).

[0043] In this implementation manner, the discrete element modeling error is required to be less than 10%, and then the discrete element model is successfully constructed. Specific implementation method 2:

[0045] This embodiment is a computer storage medium, in which at least one instruction is stored. The at least one instruction is loaded and executed by a processor to implement the method for constructing a three-dimensional discrete element model of asphalt mixture.

[0046] It should be understood that any method described in the present invention may be provided as a computer program product, software or computerized method, which may include a non-transitory machine-readable medium having instructions stored thereon, and the instructions may be used to program a computer system or other electronic device. The storage medium may include, but is not limited to, magnetic storage media, optical storage media; magneto-optical storage media include: read-only memory ROM, random access memory RAM, erasable programmable memory (e.g., EPROM and EEPROM) and flash memory layer; or other types of media suitable for storing electronic instructions. Specific implementation method three:

[0048] This embodiment is a device for constructing a three-dimensional discrete element model of asphalt mixture, and the device includes a processor and a memory. It should be understood that the device includes any device including a processor and a memory described in the present invention, and the device may also include other units and modules that perform display, interaction, processing, control, etc. and other functions through signals or instructions;

[0049] At least one instruction is stored in the memory, and the at least one instruction is loaded and executed by the processor to implement the method for constructing a three-dimensional discrete element model of asphalt mixture.

[0050] Example:

[0051] A method for constructing a three-dimensional discrete element model of asphalt mixture in this embodiment is implemented according to the following steps:

[0052] 1. Perform X-Ray scanning on the standard Marshall specimen of asphalt mixture based on industrial CT to obtain the slice image of the Marshall specimen of asphalt mixture.

[0053] 2. Import the slice image obtained in step 1 into the digital image processing software Avizo. The Avizo image processing process is as follows: Figure 2 shown.

[0054] Firstly, the aggregate components are identified and extracted through non-local mean filtering, grayscale equalization, interactive threshold segmentation, and background detection correction algorithm. Then, the segmentation of the adhesion aggregate particles is achieved through watershed segmentation, morphological opening operation, hole filling algorithm processing and three-dimensional contact segmentation algorithm. Among them, the morphological opening operation processes the aggregate model, which can simplify the image data while maintaining the basic shape. Finally, label analysis is performed to mark the separated parts (distinguished by color, and the volume coordinates can be read out at the same time).

[0055] It should be noted that, in fact, the processing process of the present invention is not limited to that of the present invention, as long as segmentation can be achieved. However, the processing process of the present invention can process images very well, and the segmentation effect is very good. For example, the digital image of the asphalt mixture processing process is Figure 3 As shown, it can be seen that the present invention can achieve clear segmentation of particles of different particle sizes.

[0056] 3. Based on step 2, the aggregate particle model is reconstructed in three dimensions through volume rendering commands, such as Figure 1 As shown, the surface mesh generation command is used to construct and export the surface mesh file (STL file) of the single aggregate model.

[0057] Fourth, the surface mesh files of aggregate particles of various particle sizes are imported into the discrete element software, and the aggregate particle model template is created using the Clump algorithm (the key parameters ratio and distance are set to 0.3 and 150 respectively), and the template library of the aggregate particle model is established, such as Figure 4 As shown, a total of 100 different aggregate particle templates were constructed in this embodiment.

[0058] 5. According to the volume, gradation type, asphalt-stone ratio, void ratio, aggregate density and asphalt density of the Marshall specimen, the volume of aggregate of each size is calculated according to formula (1), and a cylindrical wall with a height of 200 mm and a diameter of 101.6 mm is constructed. The aggregate particle model in the template library constructed in step 4 is randomly called, and the aggregate particle group of the specified gradation type is placed in the constructed cylindrical wall through the random generation algorithm clump distribute command. The overlap between particles is eliminated by the cycle command, and the gravity is set to make the aggregate particle group accumulate naturally. Finally, the aggregate particle group is compacted to a specified height of 63.5 mm by applying a speed of -1 m / s to the upper surface wall to form a skeleton structure.

[0059]

[0060] Where: V Ln ——the volume of aggregate with the nth particle size;

[0061] P n+1 , P n ——Respectively the passing percentage of the particle size of the n+1th and nth gears

[0062] V S ——the volume occupied by fine aggregate (less than 2.36 mm);

[0063] V L ——Volume occupied by coarse aggregate (2.36 mm and above);

[0064] V——specimen volume;

[0065] ρ a - density of asphalt;

[0066] ρ g - aggregate density;

[0067] α——oil-stone ratio;

[0068] VV——void ratio.

[0069] 6. Use the fish language to write a program to fill balls with a radius of 0.8 mm between aggregate particles to simulate asphalt mortar components. According to the void ratio determined in step 5, a specified number of ball particles are randomly deleted to simulate the void structure, thereby realizing the construction of a 3D digital specimen of asphalt mixture. The software effect diagram of the 3D digital specimen of asphalt mixture is shown in the figure below. Figure 5 As shown; the cross-sectional view of the three-dimensional digital specimen of asphalt mixture is shown in Figure 6 shown.

[0070] 7. The corresponding contact model and micro-contact parameters are assigned to the constructed digital specimen of asphalt mixture to simulate the mechanical behavior of asphalt mixture. The linear stiffness contact model is used for the contact between aggregates, the linear contact bonding model is used for the contact model between asphalt mortar, and the linear contact bonding model is used for the contact model between aggregates and asphalt mortar. The corresponding micro-parameters of the model are shown in Tables 1 to 3.

[0071] Table 1 Contact parameters between aggregates

[0072]

[0073] Table 2 Contact parameters inside asphalt mortar

[0074]

[0075] Table 3 Contact parameters between asphalt mortar and aggregate

[0076]

[0077]

[0078] Where A is the cross-sectional area of ​​the contact elastic beam, in m 2 , L is the length of the contact elastic beam, in m. A and L can be calculated by traversing all contacts using the fish language and obtaining the radius of the solid unit at both ends of the contact. Let A / L=l.

[0079] 8. Conduct virtual indirect tensile test, take the mechanical curve of indoor macro test as verification index, adjust the micro contact parameters to make the mechanical curve of virtual test and indoor test match well, the peak load error is 6%, and the model is successfully established. Figure 7 shown.

[0080] The present invention can efficiently and quickly obtain a large number of aggregate particle model templates to fully characterize the true surface morphology of aggregate particles, improve the accuracy of the three-dimensional discrete element model of asphalt mixture, and use a random generation algorithm to accurately control the gradation type of the constructed digital specimen, taking into account the true accuracy and efficiency of modeling, optimizing the modeling method of the three-dimensional discrete element model of asphalt mixture, and helping to promote further in-depth research on the digital design of asphalt mixture.

[0081] The above calculation examples of the present invention are only used to explain the calculation model and calculation process of the present invention in detail, and are not intended to limit the implementation methods of the present invention. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made based on the above description. It is impossible to list all the implementation methods here. All obvious changes or modifications derived from the technical solution of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for constructing a three-dimensional discrete element model of asphalt mixture. It is characterized in that The following steps are involved:

1. Perform X-Ray scanning on the standard Marshall specimen of asphalt mixture based on industrial CT to obtain the slice image of asphalt mixture; 2. Import the slice image obtained in step 1 into the digital image processing software, and use the digital image processing software to extract and segment the aggregate components; 3. Based on step 2, the aggregate particle model is reconstructed in three dimensions to derive the surface mesh model of a single aggregate. Fourth, the surface mesh model of aggregate particles of various particle sizes is imported into the discrete element software, and the template of the aggregate particle model is created by using the Clump algorithm to establish a template library of the aggregate particle model; 5. Calculate the volume of aggregates of various particle sizes based on the volume, gradation type, asphalt-stone ratio, void ratio, aggregate density, and asphalt density of the Marshall specimen: Where: V Ln ——the volume of aggregate with the nth particle size; P n+1 , P n ——respectively the passing percentage of the particle size of the n+1th and nth gears; V S ——Volume occupied by fine aggregate; V L ——Volume occupied by coarse aggregate; V——Test specimen volume; ρ a ——Asphalt density; ρ g ——aggregate density; α——oil-stone ratio; VV——void ratio; Call the aggregate particle model in the template library constructed in step 4, place the aggregate particle group of the specified gradation type in the specified cylindrical wall space through a random generation algorithm, use the loop command to eliminate the overlap between particles, set the gravity to make the aggregate particle group accumulate naturally, and finally apply a speed to the upper surface wall to compact the aggregate particle group to the specified height to form a skeleton structure; 6. Use the language program to fill ball particles between aggregate particles to simulate asphalt mortar, and randomly delete a specified number of balls to simulate the void structure according to the set void ratio, so as to realize the construction of a three-dimensional digital specimen of asphalt mixture; 7. Assign the corresponding contact model and micro-contact parameters to the constructed digital specimen of asphalt mixture to simulate the mechanical behavior of asphalt mixture; 8. Conduct virtual indirect tensile tests and indoor macroscopic mechanical tests, and use the peak load error percentage p as an evaluation index for the accuracy of the discrete element model. If the p value exceeds the modeling requirements, adjust the microscopic contact parameters of step 7 until the error percentage is less than the error requirement, thereby completing the construction of the three-dimensional discrete element model of asphalt mixture.

2. A method for constructing a three-dimensional discrete element model of asphalt mixture as claimed in claim 1, It is characterized in that The process of extracting and segmenting the aggregate components using digital image processing software in step 2 is as follows: Firstly, the aggregate components are identified and extracted through non-local mean filtering, grayscale equalization, interactive threshold segmentation and background detection correction algorithm. Then, the segmentation of the adhesion aggregate particles is achieved through watershed segmentation, morphological opening operation, hole filling algorithm and three-dimensional contact segmentation algorithm.

3. A method for constructing a three-dimensional discrete element model of asphalt mixture as claimed in claim 1 or 2, It is characterized in that Step 4: The key parameters of the Clump algorithm, ratio and distance, are set to 0.3 and 150 respectively.

4. A method for constructing a three-dimensional discrete element model of asphalt mixture as claimed in claim 3, It is characterized in that The number of aggregate particle templates constructed in step 4 is 100.

5. A method for constructing a three-dimensional discrete element model of asphalt mixture as claimed in claim 4, It is characterized in that In step 5, the cylindrical wall has a height of 200 mm, a diameter of 101.6 mm, and a compaction height of 63.5 mm.

6. A method for constructing a three-dimensional discrete element model of asphalt mixture as claimed in claim 5, Features The ball particle radius of the simulated asphalt mortar in step six is ​​0.8 mm.

7. A method for constructing a three-dimensional discrete element model of asphalt mixture as claimed in claim 6, Features The micro-contact parameters in step seven are the stiffness parameters between aggregates, the stiffness and bonding parameters between asphalt mortars, and the stiffness and bonding parameters between aggregates and asphalt mortars.

8. A method for constructing a three-dimensional discrete element model of asphalt mixture as claimed in claim 7, Features If the modeling error of the discrete element model in step eight is less than 10%, the discrete element model is successfully constructed.

9. A computer storage medium, It is characterized in that The storage medium stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the method for constructing a three-dimensional discrete element model of asphalt mixture as described in any one of claims 1 to 8.

10. A device for constructing a three-dimensional discrete element model of asphalt mixture. It is characterized in that The device includes a processor and a memory, wherein the memory stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the method for constructing a three-dimensional discrete element model of asphalt mixture as described in any one of claims 1 to 8.

Images