A numerical simulation method for maintenance construction of semi-flexible pavement based on double-scale coupling

Abstract

This invention discloses a numerical simulation method for the maintenance and construction of semi-flexible pavement based on dual-scale coupling, comprising: determining the dimensions of the milled and repaved section based on the original pavement structure obtained by three-dimensional ground-penetrating radar and falling-weight deflectometer; establishing a finite element model for the original pavement structure; simulating the semi-flexible layer after milling and repaving using discrete element method and coupling it into the finite element model to construct a dual-scale coupled model; performing simulation calculations on the compaction and molding of large-void parent asphalt mixture; conducting simulation derivation of self-leveling cement slurry filling based on the construction process and the dual-scale coupled model; establishing a discrete element numerical model for cement curing to characterize cement curing behavior; and evaluating and guiding the maintenance of the semi-flexible pavement based on the dual-scale coupled model. This invention realizes process control and performance evaluation during the construction period of semi-flexible pavement and guidance for maintenance and repair during service life, making up for the shortcomings of high testing costs and insufficient theoretical data for different construction conditions in semi-flexible pavement.

Description

Technical Field

[0001] This invention belongs to the field of road engineering technology, specifically relating to a numerical simulation method for semi-flexible pavement maintenance construction based on dual-scale coupling. Background Technology

[0002] Semi-flexible pavement structures utilize porous asphalt mixtures infused with cement grout to form a semi-flexible composite material, replacing one or more layers of traditional asphalt pavement surface layers. This novel pavement structure has received significant attention in the field of road engineering in recent years. At intersections and long, steep sections of trunk highways, near ports and container terminals, bus stops on municipal roads, and in urban BRT and FRT lanes, the pavement surface material must withstand not only significant compressive stress but also high shear stress due to the frequent braking and acceleration of heavy-duty vehicles traveling at relatively low speeds. This combined effect of stress and temperature makes the asphalt surface layer prone to rutting in these sections. Semi-flexible pavement materials are composite pavement materials formed by infusing large-pore asphalt mixtures with specific cement grout, and their stiffness falls between that of asphalt concrete and cement concrete. Indoor and engineering practice studies have shown that semi-flexible pavement materials possess excellent shear strength, rutting resistance, and high-temperature stability, making them the preferred solution for pavement design and reconstruction in the aforementioned road sections. Currently, semi-flexible pavements have been applied to paving and reconstruction projects on some road sections in China, demonstrating good performance and excellent rutting resistance.

[0003] Conducting performance research on this type of new pavement material is of great theoretical and practical significance for improving the construction level of expressways and general roads in my country, especially for improving pavement performance, extending pavement service life, and reducing pavement construction investment. Summary of the Invention

[0004] Technical problem solved: This invention proposes a numerical simulation method for the maintenance and construction of semi-flexible pavement based on dual-scale coupling. By constructing a semi-flexible pavement simulation model with dual-scale coupling of finite element and discrete element methods, it studies the macroscopic and microscopic states of the semi-flexible pavement construction process based on simulation deduction and analysis, realizes process control and performance evaluation during the construction period of semi-flexible pavement, and provides guidance for maintenance and repair during service life. This method makes up for the shortcomings of high testing costs and insufficient theoretical data for different construction conditions in semi-flexible pavement.

[0005] Technical solution:

[0006] A numerical simulation method for semi-flexible pavement maintenance construction based on dual-scale coupling, the method comprising the following steps:

[0007] S1. Based on the original road surface structure scheme measured by three-dimensional ground penetrating radar and falling weight deflectometer, determine the size of the milled and repaved section;

[0008] S2. Establish a finite element model for the original pavement structure. After milling and repaving, use discrete element method to simulate the semi-flexible layer and couple it into the finite element model to construct a dual-scale coupled model.

[0009] S3. Based on the dual-scale coupled model constructed in step s2, the designed gradation parameters are set in the discrete element part of the pavement structure. The speed data of different rolling processes are simulated in the finite element model and coupled into the dual-scale coupled model to establish a mixture compaction and forming simulation model. The compaction and forming of large-void parent asphalt mixture is simulated and calculated.

[0010] S4. Based on the construction process and the dual-scale coupling model, conduct simulation and deduction of self-leveling cement grout filling to determine the self-leveling cement filling effect under different grouting volumes and construction conditions.

[0011] S5. Establish a discrete element numerical model for cement curing to characterize cement curing behavior, obtain the pavement state and data after semi-flexible pavement curing and shaping, and deduce the cement curing time to determine the expected time to open to traffic.

[0012] S6. Obtain the morphology and composition of the semi-flexible pavement material, optimize the parameters of the dual-scale coupled model, and deduce the mechanical behavior and deformation behavior of the semi-flexible pavement structure under repeated loads under different pavement load modes and different temperatures. Obtain the high and low temperature performance and fatigue performance of the pavement structure, and evaluate and guide the maintenance of the semi-flexible pavement based on the dual-scale coupled model.

[0013] Furthermore, in step S1, the process of determining the dimensions of the milled and repaved section based on the original pavement structure obtained by three-dimensional ground-penetrating radar and falling weight deflectometer includes the following steps:

[0014] Three-dimensional ground-penetrating radar was used for detection and analysis by 3dr-Examiner. The thicknesses of the original pavement structure layers were obtained according to the radar thickness inversion formula (1.1): top layer thickness h1, middle layer thickness h2, bottom layer thickness h3, base layer thickness h4, and subbase layer thickness h5. Based on the internal damage detection results obtained from the analysis, the void volume, settlement area, and crack rate were statistically analyzed.

[0015]

[0016]

[0017] In the formula, D is the distance traveled by the electromagnetic wave in each layer of the pavement structure, Δt is the time taken for the electromagnetic wave to travel through each layer of the pavement structure, h is the distance between the interfaces of two different media, i.e., the thickness of the structural layer, c is the speed of light in the medium, θ is the angle of refraction of the electromagnetic wave with the normal in the next layer of the pavement structure, and ε is the dielectric constant. The real-time response of the pavement under dynamic impact load, i.e., the measured deflection basin, is obtained using a falling weight deflectometer. The relationship between deflection and resilient modulus is established through the following formula (1.2). Based on the fitting degree between the measured deflection basin and the theoretical deflection basin, a modulus inversion system is established to obtain the pavement structural layer modulus result E based on the measured deflection basin.

[0018]

[0019] In the formula, p is the load applied by the bearing plate of the falling weight deflectometer, δ is the radius of the standard axle load equivalent circle, l is the calculated deflection of the old pavement, m1 is the ratio of the deflection value measured by a car with standard axle load on the original pavement to the rebound deformation value measured by the bearing plate under the same pressure conditions, i.e., the wheel-plate comparison value; m2 is the expansion coefficient of the equivalent rebound modulus of the old pavement.

[0020] By controlling the milling design criteria using the inversion modulus value E, the milling and resurfacing dimensions are determined: the minimum resilient modulus E of the pavement top surface that satisfies direct overlay is determined separately. s1 Minimum resilient modulus E of the top surface of the road surface before milling the upper layer s2 Minimum resilient modulus E of the top surface of the road surface before milling the upper and middle layers s3 Minimum resilient modulus E of the top surface of the road surface before milling a certain thickness si When E≥E s1 At this time, no milling is required before paving; when E s1 >E≥E si At that time, milling to h i Thickness base layer added later; when E s1 >E≥E s2 When milling down to the bottom of the upper layer, then add the next layer; when E s2 >E≥E s3 At that time, milling is carried out to the bottom of the middle layer and then adding the next layer; E < E s3 Then, after milling all the surface layers, another layer is laid to determine the milling thickness as h. 铣 .

[0021] Further, in step S2, a finite element model is established for the original pavement structure. After milling and repaving, a discrete element method is used to simulate the semi-flexible layer and couple it to the finite element model. The process of constructing a dual-scale coupled model includes the following steps:

[0022] A finite element model was established for the original road surface structure. The size of the finite element model was constructed and analyzed using a single-lane model of the full road surface structure. The road model is a three-dimensional solid structure. The road width B is set as the X-axis, the road structure depth H is set as the Y-axis, and the direction of vehicle travel is defined as the negative Z-axis.

[0023] After milling and repaving, the semi-flexible layer was simulated using discrete element method (DEM) and coupled in a finite element model to construct an initial dual-scale coupled model. The structures in the initial dual-scale coupled model, from top to bottom, were pre-defined as a semi-flexible layer, asphalt layer, water-stabilized layer, graded crushed stone layer, and soil base layer. Specifically, the thickness of the semi-flexible layer in the initial semi-flexible pavement structure was set to h0, and the thickness of the AC layer was set to h1 + h2 + h3 - h. 铣 The thickness of the water-stabilized layer is h4, and the thickness of the graded crushed stone layer is h5; the modulus of each layer corresponds to the modulus obtained in step S1.

[0024] The initial transient modulus E0 of the SFP semi-flexible material is selected, and its viscoelastic behavior is characterized using the modified Burgers constitutive model. The AC layer is set to the Mohr-Coulomb strain softening constitutive model, and the constitutive relation is set to linear elastic in the elastic stage of the surface layer. The aggregate is set to the Hertz-Medlin constitutive model. The base layer is set to the linear elastic constitutive model. The subgrade is set to the Mohr-Coulomb constitutive model.

[0025] Using a 3D laser scanning device, the scanned mesh was imported using software commands to obtain the 3D micro-profile of the aggregate, simulating the shape and edges of the aggregate. The particle contact type settings included: linear contact models between aggregates; parallel bonding models between cement stone, between cement stone and asphalt mortar, and between asphalt mortar and aggregates; and Burgers models between asphalt mortars. For the linear contact model, the contact was simplified to an elastic beam with both ends located at the center of the particle to obtain the relationship between the micro-parameters of the contact model and the macro-parameters of the material. The equivalent beam length was L, the cross-sectional area was A, and the moment of inertia was I. The shear modulus G was obtained by formula (2.3) using the values ​​of normal stiffness and tangential stiffness from equations (2.1) and (2.2). For the parallel bonding model between aggregates and asphalt mortar, the values ​​of normal stiffness and tangential stiffness were obtained from equations (2.4) and (2.5). For the Burgers model, its model parameters were obtained through uniaxial creep tests and uniaxial penetration tests.

[0026]

[0027]

[0028]

[0029]

[0030]

[0031] Where: k n1 k s1 These represent the normal stiffness and tangential stiffness of the linear contact model, respectively; E0 is the elastic modulus of the aggregate; v is Poisson's ratio; k n2 k s2 These represent the normal stiffness and tangential stiffness of the parallel bonded model, respectively. The elastic modulus of the parallel bond model. It is half the equivalent contact width of the parallel bonding model.

[0032] Furthermore, in step S3, the simulation calculation process for compacting and molding the large-void parent asphalt mixture includes the following steps:

[0033] Based on the variable K-method, the mixture is divided into main aggregate and fine aggregate by discontinuity points. Different k values ​​are used for the main aggregate and fine aggregate to design the proportion between different particle sizes within each grade. Then, the coarse and fine aggregate ratio is determined according to the volume design method to determine the gradation range of the mixture. The orthogonal gradation design scheme is proposed with porosity and stability as research indicators. An improved main aggregate filling method is used to design a reasonable porosity of the matrix asphalt mixture and obtain the amount of mineral powder and coarse and fine aggregate.

[0034] Based on the dual-scale coupled model constructed in step s2, the design gradation parameters are set in the discrete element part of the pavement structure. The speed data of different rolling processes are obtained by simulating different rolling processes in the finite element model and coupled into the dual-scale coupled model to establish a mixture compaction and molding simulation model. The state of the parent asphalt aggregate during the compaction process and the size and mechanical data after compaction are obtained by pavement structure simulation software to evaluate the compaction effect of different processes.

[0035] The stability and splitting strength of the parent asphalt mixture were tested by simulating the Marshall stability test and splitting strength test methods of the current standard asphalt mixture, to determine the optimal asphalt content, and to evaluate the physical and mechanical properties of the parent asphalt mixture under each grade of mix design based on void ratio and stability.

[0036] Furthermore, in step S4, the process of simulating and deducing the self-leveling cement grout filling based on the construction process and the dual-scale coupled model includes the following steps:

[0037] Based on orthogonal experiments on the influence of various factors on performance, considering the drying shrinkage, fluidity, flexural strength, compressive strength and bleeding of the grout, the optimal mix ratio of cement grouting material was determined, and relevant parameters of the parallel bond model were obtained based on macroscopic mechanical tests: effective modulus of particles and effective modulus of parallel bond.

[0038] The theoretical grouting volume of self-leveling cement grout was calculated by comparing the actual grouting volume under vibration conditions with the void ratio VV of the parent asphalt macadam and the size V of the rutted slab, providing data for the amount of cement grout required for simulation and construction.

[0039] Bingham model parameters of cement slurry were obtained using a coaxial cylindrical rotating concrete rheometer, which were then used to simulate the rheological behavior of self-leveling cement slurry in the discrete element method.

[0040] Cement stone particles were generated, and a two-scale model of cement self-leveling filling was used to simulate the filling effect of self-leveling cement under different grouting volumes and construction conditions.

[0041] Furthermore, in step S5, the process of establishing a discrete element numerical model for cement curing to characterize the cement curing behavior, obtaining the pavement state and data after the semi-flexible pavement has been cured and formed, and extrapolating the cement curing time to determine the expected time for opening to traffic includes the following steps:

[0042] The microstructure changes of self-leveling cement grouting materials at different curing ages were studied using SEM. A CFD model simulating the flow state of cement curing was established. The rheological parameters of cement and asphalt were input as material property parameters into the cement curing multiphase flow model for simulation. Based on this, the relationship between the flow properties of cement curing and time was studied.

[0043] In the discrete element model, the JKR cohesion model is used as the contact constitutive model of cement particles. Based on the experimental data of self-leveling cement fluidity, shrinkage and mechanical strength, the BP neural network machine learning method is used to train a limited dataset to establish a discrete element numerical model of cement curing to characterize the cement curing behavior.

[0044] Based on the Fish language logic settings, different working condition data call files are generated to create multiple status files. These files are then compared with the non-destructive testing data of the semi-flexible pavement in the field for verification. The initial parameters of the simulation model are improved, and the pavement status and data after the semi-flexible pavement has been cured and formed are obtained. The cement curing time is then estimated to determine the expected time for opening to traffic.

[0045] Furthermore, in step S6, the process of evaluating and providing maintenance guidance for semi-flexible pavements based on a dual-scale coupled model includes the following steps:

[0046] The morphology of semi-flexible pavement materials was obtained through X-ray CT scanning and image processing techniques, and the parameters of the dual-scale coupled model were optimized.

[0047] By studying the mechanical and deformation behavior of semi-flexible pavement structures under repeated loads under different pavement load modes and temperatures, the high and low temperature performance and fatigue performance of pavement structures are obtained.

[0048] The cohesion model is used to define the damage of asphalt, cement phase, asphalt-aggregate interface and cement-asphalt. The damage is characterized by batch insertion of CZM cohesion model elements and the quasi-static failure process is simulated. The construction of semi-flexible pavement is evaluated and service life maintenance and repair guidance is proposed.

[0049] Beneficial effects:

[0050] The present invention provides a numerical simulation method for the maintenance and construction of semi-flexible pavement based on dual-scale coupling. This method utilizes a dual-scale coupled finite element and discrete element simulation model to address the problems of limited full-scale test sections and insufficient theoretical data supporting different working conditions in existing semi-flexible pavement systems. By constructing a dual-scale coupled finite element and discrete element simulation model, the method studies the macroscopic and microscopic states of the semi-flexible pavement construction process based on simulation deduction and analysis. This enables process control and performance evaluation during the construction period and provides guidance for maintenance and repair during the service life of semi-flexible pavement, overcoming the shortcomings of high testing costs and insufficient theoretical data for different construction conditions. Attached Figure Description

[0051] Figure 1 This is a flowchart illustrating the technical process of the numerical simulation method for construction and maintenance of semi-flexible pavement based on dual-scale coupling, according to an embodiment of the present invention.

[0052] Figure 2 A schematic diagram of the modulus inversion system based on the fit between the measured deflection basin and the theoretical deflection basin;

[0053] Figure 3 This is a schematic diagram illustrating the process of establishing a dual-scale coupled model for semi-flexible pavement maintenance construction according to an embodiment of the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention.

[0055] Unless otherwise defined, the technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0056] Example 1

[0057] This invention discloses a numerical simulation method for semi-flexible pavement maintenance construction based on dual-scale coupling. The numerical simulation method for semi-flexible pavement maintenance construction includes the following steps:

[0058] S1. Based on the original road surface structure scheme measured by three-dimensional ground penetrating radar and falling weight deflectometer, determine the size of the milled and repaved section;

[0059] S2. Establish a finite element model for the original pavement structure. After milling and repaving, use discrete element method to simulate the semi-flexible layer and couple it into the finite element model to construct a dual-scale coupled model.

[0060] S3. Based on the dual-scale coupled model constructed in step s2, the designed gradation parameters are set in the discrete element part of the pavement structure. The speed data of different rolling processes are simulated in the finite element model and coupled into the dual-scale coupled model to establish a mixture compaction and forming simulation model. The compaction and forming of large-void parent asphalt mixture is simulated and calculated.

[0061] S4. Based on the construction process and the dual-scale coupling model, conduct simulation and deduction of self-leveling cement grout filling to determine the self-leveling cement filling effect under different grouting volumes and construction conditions.

[0062] S5. Establish a discrete element numerical model for cement curing to characterize cement curing behavior, obtain the pavement state and data after semi-flexible pavement curing and shaping, and deduce the cement curing time to determine the expected time to open to traffic.

[0063] S6. Obtain the morphology and composition of the semi-flexible pavement material, optimize the parameters of the dual-scale coupled model, and deduce the mechanical behavior and deformation behavior of the semi-flexible pavement structure under repeated loads under different pavement load modes and different temperatures. Obtain the high and low temperature performance and fatigue performance of the pavement structure, and evaluate and guide the maintenance of the semi-flexible pavement based on the dual-scale coupled model.

[0064] As a preferred technical solution of the present invention: In step S1, the original pavement structure scheme based on three-dimensional ground penetrating radar and FWD and the method for determining the size of the milled and repaved section are as follows: First, three-dimensional ground penetrating radar is used for detection and analysis by 3dr-Examiner. According to the radar thickness inversion formula (1.1), the thickness of the original pavement structure layers is obtained: the thickness of the upper layer h1, the thickness of the middle layer h2, the thickness of the lower layer h3, the thickness of the base layer h4, and the thickness of the subbase layer h5. Based on the internal damage detection results obtained from the analysis, the void volume, the settlement area, and the crack rate are statistically analyzed.

[0065]

[0066]

[0067] In the formula, D is the distance the electromagnetic wave travels through each layer of the road structure, Δt is the time the electromagnetic wave takes to travel through each layer of the road structure, h is the distance between the interface of two different media, i.e. the thickness of the structural layer, c is the speed of light in the medium, θ is the angle of refraction between the electromagnetic wave and the normal in the next layer of the road structure, and ε is the dielectric constant.

[0068] Secondly, a falling weight deflectometer is used to obtain the real-time response of the pavement under dynamic impact load, i.e., the measured deflection basin. The relationship between deflection and resilient modulus is established through the following formula (1.2). Based on the fitting degree between the measured deflection basin and the theoretical deflection basin, a modulus inversion system is established to obtain the pavement structural layer modulus result E based on the measured deflection basin:

[0069]

[0070] In the formula, p is the load applied by the bearing plate of the falling weight deflectometer, δ is the radius of the standard axle load equivalent circle, l is the calculated deflection of the old pavement, m1 is the ratio of the deflection value measured by a car with standard axle load on the original pavement to the rebound deformation value measured by the bearing plate under the same pressure conditions, i.e., the wheel-plate comparison value; m2 is the expansion coefficient of the equivalent rebound modulus of the old pavement.

[0071] Finally, the milling design criteria are controlled by the inversion modulus value E, thereby determining the milling and resurfacing dimensions: the minimum resilient modulus E of the pavement top surface that can be directly overlaid is determined respectively. s1 Minimum resilient modulus E of the top surface of the road surface before milling the upper layer s2 Minimum resilient modulus E of the top surface of the road surface before milling the upper and middle layers s3 Minimum resilient modulus E of the top surface of the road surface before milling a certain thickness si When E≥E s1 At this time, no milling is required before paving; when E s1 >E≥E si At that time, milling to h i Thickness base layer added later; when E s1 >E≥E s2 At that time, milling down to the bottom of the upper layer is followed by additional paving. When E s2 >E≥E s3 At that time, milling is carried out to the bottom of the middle layer and then adding the next layer; E < E s3 Then, the entire surface layer is milled and then re-laid. The final milling thickness is determined to be h. 铣 .

[0072] As a preferred technical solution of the present invention: In step S2, the method for constructing a dual-scale coupled model based on the finite element method and the discrete element method is as follows: First, a finite element model is established for the original pavement structure. The size of the finite element model is constructed and analyzed using the single-lane model of the full pavement structure obtained in step S1. The road model is a three-dimensional structure, with the road width set to B (X-axis), the road structure depth set to H (Y-axis), and the vehicle travel direction defined as the negative Z-axis. After milling and repaving, the semi-flexible layer is simulated and coupled in the finite element model using the discrete element method. The initial dual-scale coupled model is pre-set from top to bottom as a semi-flexible layer (SFP), an asphalt layer (AC), a water-stabilized layer (CSM), a graded crushed stone layer (GM), and a soil base layer (SG); wherein, the thickness of the semi-flexible layer in the initial semi-flexible pavement structure is set to h0, and the thickness of the AC layer is h1+h2+h3-h 铣 The thickness of the water-stabilized layer is h4, and the thickness of the graded crushed stone layer is h5; the modulus of each layer corresponds to the modulus obtained in S1. The initial transient modulus E0 of the SFP semi-flexible material is selected, and the modified Burgers constitutive model is used to characterize its viscoelastic behavior; the AC layer is set to a Mohr-Coulomb strain softening constitutive model, and the constitutive relation is set to linear elasticity in the elastic stage of the surface layer; the aggregate is set to a Hertz-Medlin constitutive model; the base layer is set to a linear elastic constitutive model; and the subgrade is set to a Mohr-Coulomb constitutive model. Using a 3D laser scanning device, the scanned mesh is imported using software commands to obtain the 3D micro-profile of the aggregate, simulating its shape and edges. Particle contact type settings: a linear contact model is set between aggregates; a parallel bonding model is set between cement paste, between cement paste and asphalt mortar, and between asphalt mortar and aggregate; and a Burgers model is set between asphalt mortar particles. For the linear contact model, the contact is simplified to an elastic beam with both ends located at the center of the particle to obtain the relationship between the microscopic parameters of the contact model and the macroscopic parameters of the material. The normal stiffness and tangential stiffness are obtained through equations (2.1) and (2.2), and the shear modulus G is obtained through equation (2.3). For the parallel bond model between aggregate and asphalt mortar, the normal stiffness and tangential stiffness are obtained through equations (2.4) and (2.5). For the Burgers model, its model parameters are obtained through uniaxial creep test and uniaxial penetration test.

[0073]

[0074]

[0075]

[0076]

[0077]

[0078] As a preferred technical solution of the present invention: In step S3, the numerical calculation method for compaction molding of the large-void matrix asphalt mixture is as follows: First, based on the variable "K" method, the mixture is divided into main aggregate and fine aggregate by discontinuity points. Different k values ​​are taken for the main aggregate and fine aggregate to design the proportion between each particle size within their respective ranges. Then, the coarse and fine aggregate ratio is determined according to the volume design method to determine the gradation range of the mixture. An orthogonal gradation design scheme is proposed with porosity and stability as research indicators. A reasonable porosity of the matrix asphalt mixture is designed using an improved main aggregate filling method to obtain the amount of mineral powder and coarse and fine aggregates. Second, based on the dual-scale model constructed in S2, the above-mentioned design gradation parameters are set in the discrete element part of the pavement structure. Different rolling processes are simulated in the finite element model to obtain speed data and coupled into the dual-scale model to establish a mixture compaction molding simulation model. The state of the matrix asphalt aggregate during the compaction process and the dimensions and mechanical data after compaction are obtained through pavement structure simulation software to evaluate the compaction effect of different processes. Finally, the stability and splitting strength of the parent asphalt mixture were tested by simulating the Marshall stability test and splitting strength test methods of the current standard asphalt mixture, so as to determine the optimal asphalt content, and evaluate the physical and mechanical properties of the parent asphalt mixture under each grade of mix design based on void ratio and stability.

[0079] As a preferred technical solution of the present invention: In step S4, the simulation and deduction method for self-leveling cement grout filling is as follows: Based on orthogonal experiments on the influence of various factors, considering the drying shrinkage, fluidity, flexural strength, compressive strength, and bleeding of the grout, the optimal mix ratio of cement grouting material is determined, and relevant parameters of the parallel bonding model, such as particle effective modulus and parallel bonding effective modulus, are obtained according to the above macroscopic mechanical tests. The theoretical grouting volume of self-leveling cement grout and the actual grouting volume under vibration conditions are compared by calculating the void ratio VV of the parent asphalt macadam and the rut slab size V, providing data for the amount of cement grout required for simulation and construction. The Bingham model parameters of the cement grout are obtained using a coaxial cylindrical rotating concrete rheometer, which is used to simulate the rheological behavior of self-leveling cement grout in the discrete element method. Cement stone particles are generated, and a two-scale model of cement self-leveling filling is simulated based on the construction processes of "pressing in", "road rake dragging", and "vibration" to determine the self-leveling cement filling effect under different grouting volumes and construction conditions, providing a basic model for subsequent cement curing.

[0080] As a preferred technical solution of the present invention: In step S5, the method for characterizing cement curing behavior is as follows: SEM is used to study the microstructural changes of self-leveling cement grouting materials at different curing ages; a CFD model simulating the flow state of cement curing is established; the rheological parameters of cement and asphalt are input as material property parameters into the cement curing multiphase flow model for simulation; and the relationship between the flow properties of cement curing and time is studied. In the discrete element model, the JKR cohesion model is used as the contact constitutive model of cement particles. Based on the test data of self-leveling cement flowability, shrinkage, and mechanical strength, a BP neural network machine learning method is used to train a limited dataset to establish a discrete element numerical model of cement curing behavior. Based on the Fish language logic, different working condition data call files are generated to create multiple state files, which are then verified against the non-destructive testing data of the semi-flexible pavement in the field. The initial parameters of the simulation model are improved, and the pavement state and data after curing of the semi-flexible pavement are obtained. The cement curing time is then extrapolated to determine the expected opening time to traffic.

[0081] As a preferred technical solution of the present invention: In step S6, the semi-flexible pavement performance evaluation method based on a dual-scale coupling model is specifically as follows: The morphology and composition of the semi-flexible pavement material are obtained through X-ray CT scanning and image processing technology, and the parameters of the dual-scale coupling model are optimized. After the model is established, the mechanical behavior and deformation behavior of the semi-flexible pavement structure under repeated loading are deduced under different pavement bearing modes and temperatures, thereby obtaining the high and low temperature performance and fatigue performance of the pavement structure. A cohesion model is used to define the damage of asphalt, cement phase, asphalt-aggregate interface, and cement-asphalt. CZM cohesion model elements are batch-inserted to characterize the damage and simulate the quasi-static failure process. This enables the evaluation of semi-flexible pavement construction and guidance for maintenance and repair during service life.

[0082] Example 2

[0083] like Figure 1-3 As shown, this invention proposes a numerical simulation method for semi-flexible pavement maintenance construction based on dual-scale coupling. This method uses a dual-scale coupling model as its core, and through the reconstruction of the three-dimensional structure of the semi-flexible pavement and the assignment and extraction of mesoscopic parameters, it can analyze the pavement forming under different material conditions and processes, and establish an evaluation system to guide construction and maintenance. Specifically, it includes the following steps:

[0084] S1. Determination of the original road surface structure scheme and milled and repaved section dimensions based on three-dimensional ground penetrating radar and FWD;

[0085] As shown in the figure, firstly, three-dimensional ground-penetrating radar was used for detection and analysis by 3dr-Examiner. According to the radar thickness inversion formula (1.1), the original pavement structure layer thicknesses were obtained: upper, middle, and lower layers were 4, 6, and 8 cm respectively; the water-stabilized base course was 25 cm thick; and the graded crushed stone layer was 15 cm thick. Based on the internal damage detection results obtained from the analysis, the void volume, settlement area, and crack rate were statistically analyzed.

[0086]

[0087]

[0088]

[0089] Secondly, the real-time response of the road surface under dynamic impact load, i.e., the measured deflection basin, is obtained by using a falling weight deflectometer (FWD). The relationship between deflection and resilient modulus is established by formula (1.2).

[0090] Based on the good fit between the measured and theoretical deflection basins, a modulus inversion system was established to obtain the pavement structure layer modulus results based on the measured deflection basin: ① Based on the three-dimensional ground-penetrating radar pavement interior detection results, the thickness of each structural layer, modulus inversion range, Poisson's ratio, and other parameters in the modulus inversion system were analyzed to obtain pavement structure combinations under different parameters. The theoretical deflection basin was then calculated using a layered elastic system mechanical model; ② The characteristic parameters of the deflection basin were determined; ③ The theoretical deflection basin was corrected using the EXP exponential function model, and the characteristic parameter values ​​of the theoretical deflection basin were calculated and their relationship with the pavement structure was obtained. The relationship between parameters; ④ Then, the EXP exponential function model is used to correct the measured deflection basin, and the characteristic parameter values ​​of the measured deflection basin are calculated. The pavement structure parameters of the measured deflection basin are obtained according to the relationship model between the pavement structure parameters and the characteristic parameters of the deflection basin in steps ② and ③; ⑤ The obtained pavement structure parameters are substituted into the layered system model to obtain the calculated deflection basin. The calculated deflection basin is compared with the measured deflection basin to obtain the error between the two; ⑥ Steps ③ to ⑤ are iterated until the relative error between the two meets the requirements. The pavement structure parameters obtained at this time are the modulus inversion results.

[0091] Finally, the milling design criteria are controlled by the inversion modulus value E, thereby determining the milling and resurfacing dimensions: the minimum resilient modulus E of the pavement top surface that can be directly overlaid is determined respectively. s1 Minimum resilient modulus E of the top surface of the road surface before milling the upper layer s2 Minimum resilient modulus E of the top surface of the road surface before milling the upper and middle layers s3 Minimum resilient modulus E of the top surface of the road surface before milling a certain thickness si When E≥E s1 At this time, no milling is required before paving; when E s1 >E≥E si At that time, milling to hi Thickness base layer added later; when E s1 >E≥E s2 At that time, milling down to the bottom of the upper layer is followed by additional paving. When E s2 >E≥E s3 At that time, milling is carried out to the bottom of the middle layer and then adding the next layer; E < E s3 Then, the entire surface layer is milled and then re-laid. The final milling thickness is determined to be h. 铣 .

[0092] S2. Construction of a dual-scale coupled model based on the finite element method and the discrete element method;

[0093] First, a finite element model was established for the original pavement structure. The model dimensions were constructed and analyzed using the single-lane model of the full pavement structure obtained in S1. The road model is a three-dimensional structure with a road width of 3m (X-axis) and a road structure depth of 3.8m (Y-axis). The direction of vehicle travel is defined as the negative Z-axis. After milling and repaving, the semi-flexible layer was simulated and coupled to the finite element model using discrete element methods. The initial dual-scale coupled model was pre-defined from top to bottom as follows: semi-flexible layer (SFP), asphalt layer (AC), water-stabilized layer (CSM), graded crushed stone layer (GM), and soil base layer (SG). The initial semi-flexible pavement structure was set with a semi-flexible layer thickness of 6cm, an AC layer thickness of 12cm, a water-stabilized layer thickness of 36cm, and a graded crushed stone layer thickness of 15cm. The modulus of each layer corresponds to the modulus obtained in S1. The initial transient modulus of the SFP semi-flexible material was selected as 3500 MPa, and its viscoelastic behavior was characterized using a modified Burgers constitutive model. The AC layer was set to a Mohr-Coulomb strain softening constitutive model, and the constitutive relation was set to linear elasticity during the elastic stage of the surface layer. The aggregate was set to a Hertz-Medlin constitutive model; the base course was set to a linear elastic constitutive model; and the subgrade was set to a Mohr-Coulomb constitutive model. Using a 3D laser scanning device, the scanned mesh was imported using software commands to obtain the 3D micro-profile of the aggregate, simulating its morphology and edges. Particle contact types were set as follows: linear contact model between aggregates; parallel bond model between cement paste, between cement paste and asphalt mortar, and between asphalt mortar and aggregate; and Burgers model between asphalt mortar particles. For the linear contact model, the contact is simplified to an elastic beam with both ends located at the center of the particle sphere to obtain the relationship between the microscopic parameters of the contact model and the macroscopic parameters of the material. The normal stiffness and tangential stiffness are obtained through equations (2.1-2.2), and the shear modulus G is obtained through equation (2.3). For the parallel bond model between aggregate and asphalt mortar, the normal stiffness and tangential stiffness are obtained through equations (2.4-2.5). For the Burgers model, its model parameters are obtained through uniaxial creep tests and uniaxial penetration tests. Based on the above methods and existing research, the parameters taken for each model in this example are shown in Tables 1-4.

[0094]

[0095]

[0096]

[0097]

[0098]

[0099] Table 1 Parameters of the Linear Contact Model

[0100]

[0101] Table 2 Parameters of the Parallel Bonding Model Between Aggregates and Asphalt Mortar

[0102]

[0103] Table 3 Dynamic parameters of the Burgers model in asphalt mortar compartment

[0104]

[0105]

[0106] Table 4 Static parameters of the Burgers model in asphalt mortar compartment

[0107]

[0108] S3. Numerical calculation of compaction molding of large-void parent asphalt mixture;

[0109] First, based on the variable "K" method, the mixture is divided into main aggregate and fine aggregate by discontinuity. The proportion of each particle size is designed by taking different k values ​​for the main aggregate and fine aggregate. Then, the ratio of coarse and fine aggregate is determined according to the volume design method, and the gradation range of the mixture is determined. One representative gradation is shown in Table 4. The orthogonal design scheme of gradation is proposed with porosity and stability as research indicators.

[0110] Table 4 Gradation of Large Porosity Matrix Asphalt Mixture

[0111]

[0112] An improved main aggregate filling method was used to design the void ratio of the matrix asphalt mixture and obtain the amounts of mineral powder and coarse and fine aggregates. The specific steps are as follows: ① Select materials with good performance and determine the apparent density and bulk density of the mineral aggregates; ② Select the gradations of coarse and fine aggregates; ③ Measure the void ratio of the main aggregates using method 2 above, determine the asphalt content and the powder-to-binder ratio, and calculate the amounts of mineral powder, coarse aggregates, and fine aggregates using the improved CAVF method; ④ Determine the asphalt content using the Kentenberg fly test and the Schellenberg asphalt leakage test; ⑤ Use the Marshall test to test the performance of the asphalt mixture. If the requirements are met, the mix design is complete; otherwise, change the coarse and fine aggregate gradations and repeat steps ③ and subsequent steps. Based on the dual-scale model constructed in s2, the above design gradation parameters are set in the discrete element part of the pavement structure. The compaction process for asphalt mixtures consists of three steps: one pass of vibratory compaction using a 12-ton steel wheel roller, 1-2 passes of static compaction using a 12-ton steel wheel roller, and final static finishing using a 7-ton steel roller. Velocity data is obtained by simulating this compaction process in a finite element model and coupled into a dual-scale model to establish a simulation model for asphalt mixture compaction. Road structure simulation software is used to obtain the state of the parent asphalt aggregate during compaction and the dimensional and mechanical data after compaction, evaluating the compaction effect. Finally, the stability and splitting strength of the parent asphalt mixture are tested using the Marshall stability test and splitting strength test methods of current asphalt mixture standards to determine the optimal asphalt content. The physical and mechanical properties of the parent asphalt mixture at each mix grade are evaluated based on porosity and stability as fundamental indicators.

[0113] S4. Simulation and deduction of self-leveling cement grout filling;

[0114] Based on orthogonal experiments considering the influence of various performance factors, the optimal mix ratio of cement grouting material was determined, taking into account the drying shrinkage, fluidity, flexural strength, compressive strength, and bleeding properties of the grout. Relevant parameters of the parallel bond model, such as particle effective modulus and parallel bond effective modulus, were obtained from the aforementioned macroscopic mechanical tests, as shown in Tables 5-6. The theoretical grouting volume of self-leveling cement grout and the actual grouting volume under vibration conditions were compared using the porosity VV of the parent asphalt aggregate and the rut slab size V, providing data for the cement grout usage required for simulation and construction. Bingham model parameters of the cement grout were obtained using a coaxial cylindrical rotating concrete rheometer, used to simulate the rheological behavior of self-leveling cement grout in the discrete element method. 1mm cement stone particles were generated, and a two-scale model of cement self-leveling filling was simulated based on construction processes such as "pressing," "road rake dragging," and "vibration" to determine the filling effect of self-leveling cement under different grouting volumes and construction conditions, providing a basic model for subsequent cement curing.

[0115] Table 5 Parameters of the parallel bonding model between cement stone particles

[0116]

[0117] Table 6. Parallel Bonding Model Parameters Between Cement Stone and Asphalt Mortar

[0118]

[0119] S5. Characterization of cement curing behavior;

[0120] SEM was used to study the microstructural changes of self-leveling cement grouting materials at different curing ages. A CFD model simulating the flow state of cement curing was established, with the rheological parameters of cement and asphalt used as material property parameters input into the cement curing multiphase flow model for simulation. Based on this, the relationship between the flow properties of cement curing and time was studied. In the discrete element model, the JKR cohesion model was used as the contact constitutive model of cement particles. Based on experimental data of self-leveling cement flowability, shrinkage, and mechanical strength, a BP neural network machine learning method was used to train a limited dataset to establish a discrete element numerical model of cement curing behavior. Using the FILE language logic, various state files were generated by calling data files under different working conditions. These files were then verified against non-destructive field test data of semi-flexible pavement to improve the initial parameters of the simulation model. The pavement state and data after curing of the semi-flexible pavement were obtained, and the cement curing time was estimated to determine the expected time for opening to traffic.

[0121] S6. Performance evaluation of semi-flexible pavement based on dual-scale coupling model;

[0122] The morphology and composition of semi-flexible pavement materials were obtained through X-ray CT scanning and image processing techniques, and the parameters of the dual-scale coupled model were optimized. After the model was established, the mechanical and deformation behaviors of the semi-flexible pavement structure under repeated loading were deduced under different pavement load modes and temperatures, thereby obtaining the high and low temperature performance and fatigue performance of the pavement structure. A cohesion model was used to define the damage of asphalt, cement phase, asphalt-aggregate interface, and cement-asphalt. CZM cohesion model elements were batch-inserted to characterize the damage and simulate the quasi-static failure process. This enables the evaluation of semi-flexible pavement construction and guidance for service-life maintenance and repair.

[0123] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A numerical simulation method for semi-flexible pavement maintenance construction based on dual-scale coupling, characterized in that, The numerical simulation method for semi-flexible pavement maintenance construction includes the following steps: S1. Based on the original pavement structure scheme measured by three-dimensional ground penetrating radar and falling weight deflectometer, determine the size of the milled and repaved section; In step S1, the process of determining the dimensions of the milled and repaved section based on the original pavement structure obtained by three-dimensional ground penetrating radar and falling weight deflectometer includes the following steps: Three-dimensional ground-penetrating radar (GPR) was used for detection and analysis using 3dr-Examiner, and the thickness was determined according to the radar thickness inversion formula. The original pavement structure layer thicknesses were obtained as follows: top layer thickness h1, middle layer thickness h2, bottom layer thickness h3, base layer thickness h4, and subbase layer thickness h5. Based on the internal damage detection results obtained from the analysis, the void volume, settlement area, and crack rate were statistically analyzed. ； ； In the formula, This represents the distance the electromagnetic wave travels through each layer of the road surface structure. This refers to the time it takes for electromagnetic waves to pass through each layer of the road surface structure. This refers to the distance between the interfaces of two different media, i.e., the thickness of the structural layer. The speed at which light travels through a medium. The angle of refraction between the electromagnetic wave and the normal in the layer below the road surface structure. The dielectric constant is used; the real-time response of the road surface under dynamic impact load, i.e., the measured deflection basin, is obtained using a falling weight deflectometer, and is expressed by the following formula. The relationship between deflection and resilient modulus is established. Based on the good fit between the measured and theoretical deflection basins, a modulus inversion system is established to obtain the pavement structural layer modulus result E based on the measured deflection basin. ； In the formula, Apply load to the bearing plate of the falling weight deflectometer. The radius of the standard axle load equivalent circle. Calculate the deflection for the old road surface. It is the ratio of the deflection value measured on the original road surface by a car with standard axle load to the rebound deformation value measured by a bearing plate under the same pressure conditions, i.e., the wheel plate comparison value. The expansion factor for the equivalent resilient modulus of the old road surface; By controlling the milling design criteria using the inversion modulus value E, the milling and resurfacing dimensions are determined: the minimum resilient modulus of the pavement top surface required for direct overlay is determined separately. Minimum resilient modulus of the top surface of the road before milling the upper layer Minimum resilient modulus of the top surface of the road surface before milling the upper and middle layers Minimum resilient modulus of the top surface of a road surface before milling a certain thickness ;when At this time, no milling is required before paving; when At that time, milling to Thickness base layer added later; when When milling down to the bottom of the upper layer, then add the next layer; when At that time, milling is carried out to the bottom of the middle layer and then adding the next layer; Then, after milling all the surface layers, another layer is laid to determine the milling thickness as h. 铣 ; S2. Establish a finite element model for the original pavement structure. After milling and repaving, use discrete element method to simulate the semi-flexible layer and couple it into the finite element model to construct a dual-scale coupled model. In step S2, a finite element model is established for the original pavement structure. After milling and repaving, a discrete element method is used to simulate the semi-flexible layer and couple it to the finite element model. The process of constructing a dual-scale coupled model includes the following steps: A finite element model was established for the original road surface structure. The size of the finite element model was constructed and analyzed using a single-lane model of the full road surface structure. The road model is a three-dimensional structure. The road width B is set as the X-axis, the road structure depth H is set as the Y-axis, and the direction of vehicle travel is defined as the negative Z-axis. After milling and repaving, the semi-flexible layer was simulated using discrete element method (DEM) and coupled in a finite element model to construct an initial dual-scale coupled model. The structures in the initial dual-scale coupled model, from top to bottom, were pre-defined as a semi-flexible layer, asphalt layer, water-stabilized layer, graded crushed stone layer, and soil base layer. Specifically, the thickness of the semi-flexible layer in the initial semi-flexible pavement structure was set to h0, and the thickness of the AC layer was set to h1 + h2 + h3 - h. 铣 The thickness of the water-stabilized layer is h4, and the thickness of the graded crushed stone layer is h5; the modulus of each layer corresponds to the modulus obtained in step S1. The initial transient modulus E0 of the SFP semi-flexible material is selected, and its viscoelastic behavior is characterized using the modified Burgers constitutive model. The AC layer is set to the Mohr-Coulomb strain softening constitutive model, and the constitutive relation is set to linear elastic in the elastic stage of the surface layer. The aggregate is set to the Hertz-Medlin constitutive model. The base layer is set to the linear elastic constitutive model. The subgrade is set to the Mohr-Coulomb constitutive model. Using a 3D laser scanning device, the scanned mesh was imported using software commands to obtain the 3D micro-profile of the aggregate, simulating the shape and edges of the aggregate. The particle contact type settings included: linear contact models between aggregates; parallel bonding models between cement stone, between cement stone and asphalt mortar, and between asphalt mortar and aggregates; and Burgers models between asphalt mortars. For the linear contact model, the contact was simplified to an elastic beam with both ends located at the center of the particle to obtain the relationship between the micro-parameters of the contact model and the macro-parameters of the material. The equivalent beam length was L, the cross-sectional area was A, and the moment of inertia was I. The shear modulus G was obtained by formula (2.3) using the values ​​of normal stiffness and tangential stiffness from equations (2.1) and (2.2). For the parallel bonding model between aggregates and asphalt mortar, the values ​​of normal stiffness and tangential stiffness were obtained from equations (2.4) and (2.5). For the Burgers model, its model parameters were obtained through uniaxial creep tests and uniaxial penetration tests. ； ； ； ； ； in: , These represent the normal stiffness and tangential stiffness of the linear contact model, respectively. The elastic modulus of the aggregate; Poisson's ratio; , These represent the normal stiffness and tangential stiffness of the parallel bonded model, respectively. The elastic modulus of the parallel bond model. It is half the equivalent contact width of the parallel bonding model; S3. Based on the dual-scale coupled model constructed in step s2, the designed gradation parameters are set in the discrete element part of the pavement structure. The speed data of different rolling processes are simulated in the finite element model and coupled into the dual-scale coupled model to establish a mixture compaction and forming simulation model. The compaction and forming of large-void parent asphalt mixture is simulated and calculated. S4. Based on the construction process and the dual-scale coupling model, conduct simulation and deduction of self-leveling cement grout filling to determine the self-leveling cement filling effect under different grouting volumes and construction conditions. S5. Establish a discrete element numerical model for cement curing to characterize cement curing behavior, obtain the pavement state and data after semi-flexible pavement curing and shaping, and deduce the cement curing time to determine the expected time to open to traffic. S6. Obtain the morphology and composition of the semi-flexible pavement material, optimize the parameters of the dual-scale coupled model, and deduce the mechanical behavior and deformation behavior of the semi-flexible pavement structure under repeated loads under different pavement load modes and different temperatures. Obtain the high and low temperature performance and fatigue performance of the pavement structure, and evaluate and guide the maintenance of the semi-flexible pavement based on the dual-scale coupled model.

2. The numerical simulation method for semi-flexible pavement maintenance construction based on dual-scale coupling according to claim 1, characterized in that, Step S3, the simulation calculation process for compacting large-void parent asphalt mixture includes the following steps: Based on the variable K-method, the mixture is divided into main aggregate and fine aggregate by discontinuity points. Different k values ​​are used for the main aggregate and fine aggregate to design the proportion between different particle sizes within each grade. Then, the coarse and fine aggregate ratio is determined according to the volume design method to determine the gradation range of the mixture. The orthogonal gradation design scheme is proposed with porosity and stability as research indicators. An improved main aggregate filling method is used to design a reasonable porosity of the matrix asphalt mixture and obtain the amount of mineral powder and coarse and fine aggregate. Based on the dual-scale coupled model constructed in step s2, the design gradation parameters are set in the discrete element part of the pavement structure. The speed data of different rolling processes are obtained by simulating different rolling processes in the finite element model and coupled into the dual-scale coupled model to establish a mixture compaction and molding simulation model. The state of the parent asphalt aggregate during the compaction process and the size and mechanical data after compaction are obtained by pavement structure simulation software to evaluate the compaction effect of different processes. The stability and splitting strength of the parent asphalt mixture were tested by simulating the Marshall stability test and splitting strength test methods of the current standard asphalt mixture, to determine the optimal asphalt content, and to evaluate the physical and mechanical properties of the parent asphalt mixture under each grade of mix design based on void ratio and stability.

3. The numerical simulation method for semi-flexible pavement maintenance construction based on dual-scale coupling according to claim 1, characterized in that, Step S4, based on the construction process and the dual-scale coupled model, includes the following steps in the simulation and deduction of self-leveling cement grout filling: Based on orthogonal experiments on the influence of various factors on performance, considering the drying shrinkage, fluidity, flexural strength, compressive strength and bleeding of the grout, the optimal mix ratio of cement grouting material was determined, and relevant parameters of the parallel bond model were obtained based on macroscopic mechanical tests: effective modulus of particles and effective modulus of parallel bond. The theoretical grouting volume of self-leveling cement grout was calculated by comparing the actual grouting volume under vibration conditions with the void ratio VV of the parent asphalt macadam and the size V of the rutted slab, providing data for the amount of cement grout required for simulation and construction. Bingham model parameters of cement slurry were obtained using a coaxial cylindrical rotating concrete rheometer, which were then used to simulate the rheological behavior of self-leveling cement slurry in the discrete element method. Cement stone particles were generated, and a two-scale model of cement self-leveling filling was used to simulate the filling effect of self-leveling cement under different grouting volumes and construction conditions.

4. The numerical simulation method for semi-flexible pavement maintenance construction based on dual-scale coupling according to claim 1, characterized in that, In step S5, the process of establishing a discrete element numerical model for cement curing to characterize cement curing behavior, obtaining the pavement state and data after the semi-flexible pavement has been cured and formed, and extrapolating the cement curing time to determine the expected time for opening to traffic includes the following steps: The microstructure changes of self-leveling cement grouting materials at different curing ages were studied using SEM. A CFD model simulating the flow state of cement curing was established. The rheological parameters of cement and asphalt were input as material property parameters into the cement curing multiphase flow model for simulation. Based on this, the relationship between the flow properties of cement curing and time was studied. In the discrete element model, the JKR cohesion model is used as the contact constitutive model of cement particles. Based on the experimental data of self-leveling cement fluidity, shrinkage and mechanical strength, the BP neural network machine learning method is used to train a limited dataset to establish a discrete element numerical model of cement curing to characterize the cement curing behavior. Based on the Fish language logic settings, different working condition data call files are generated to create multiple status files. These files are then compared with the non-destructive testing data of the semi-flexible pavement in the field for verification. The initial parameters of the simulation model are improved, and the pavement status and data after the semi-flexible pavement has been cured and formed are obtained. The cement curing time is then estimated to determine the expected time for opening to traffic.

5. The numerical simulation method for semi-flexible pavement maintenance construction based on dual-scale coupling according to claim 1, characterized in that, In step S6, the process of evaluating and providing maintenance guidance for semi-flexible pavements based on a dual-scale coupled model includes the following steps: The morphology of semi-flexible pavement materials was obtained through X-ray CT scanning and image processing techniques, and the parameters of the dual-scale coupled model were optimized. By studying the mechanical and deformation behavior of semi-flexible pavement structures under repeated loads under different pavement load modes and temperatures, the high and low temperature performance and fatigue performance of pavement structures are obtained. The cohesion model is used to define the damage of asphalt, cement phase, asphalt-aggregate interface and cement-asphalt. The damage is characterized by batch insertion of CZM cohesion model elements and the quasi-static failure process is simulated. The construction of semi-flexible pavement is evaluated and service life maintenance and repair guidance is proposed.

Images